69203 v2 The World Bank Least Cost Electricity Master Plan, Djibouti Final Report Volume 2 of 2 Appendices Submitted by November 2009 The World Bank Least Cost Electricity Master Plan, Djibouti Final Report Volume 2 of 2 Appendices Submitted by November 2009 THE WORLD BANK LEAST COST ELECTRICITY MASTER PLAN, DJIBOUTI FINAL REPORT VOLUME 2 of 2 APPENDICES November 2009 Prepared by Prepared for Parsons Brinckerhoff Ltd The World Bank (UK) Report Title : Least Cost Electricity Master Plan, Djibouti (Volume 2 of 2: Appendices) Report Status : Final Report Job No : 63579 A Date : November 2009 Prepared by : Mo Deif, Mark Fraser, Richard Gadsden...... Checked by : Richard Butlin .................................................. Approved by : Bruce Stedall .................................................... CONTENTS VOLUME 2: APPENDICES Page LIST OF ABBREVIATIONS APPENDIX A: Study Terms of Reference A.1 APPENDIX B: Sales Forecast B.1 APPENDIX C: Review of Geothermal Resources in Djibouti C.1 APPENDIX D: Review of Wind Resources in Djibouti D.1 APPENDIX E: Description of ASPLAN Least Cost Generation Planning Software E.1 APPENDIX F: Generation Planning Results F.1 APPENDIX G: Load Flow Results G.1 APPENDIX H: Transient Stability Results H.1 APPENDIX I: Distribution Network I.1 Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 LIST OF ABBREVIATIONS AMT: Audio Magnetic-Telluric, C.11 ARGeo: African Rift Geothermal Facility, C.14 BRGM: Bureau de Recherches Géologiques et Minières, C.9 CADLF: Coincident After Diversity Load Factor, B.16 CAPEX: Capital Expenditure, C.32 CERD: Centre d’Etudes et de Recherche de Djibouti, C.3 COUE: Cost of Unserved Energy, F.3 CRC: Crystallizers Reactor Clarifier, C.22 EdD: Electricite de Djibouti, B.3 ELDC: Equivalent Load Duration Curve, E.3 ENS: Energy Not Served, E.4 EUR: Euro, D.7 EWEA: European Wind Energy Association, D.13 g/l: grams per litre, C.7 GCCU: Geothermal Combined Cycle Unit, C.24 GDA: Geothermal Development Associates, C.17 GDP: Gross Domestic Product, B.4 GEF: Global Environment Facility, D.3 Hz: Hertz, C.21 ISOR: Islenskar Orkurannsoknir, C.3 kg/s: kg per second, C.13 km: kilometre, C.33 kV: kilovolt, C.33 kWh: kilowatt hour, C.26 LOLE: Loss of Load Expectation, F.3 LOLP: Loss of Load Probability, E.4 LRMC: Long Run Marginal Cost, F.3 LV: Low Voltage, B.6 MoU: Memorandum of Understanding, C.4 MT: Magneto-Telluric, C.14 MV: Medium Voltage, B.12 MW: Megawatt, C.21 NEA: National Energy Authority, C.14 O&M: Operation & Maintenance, C.4 OPEX: Operational Expenditure, D.7 PB: Parsons Brinckerhoff, C.15 Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 PPA: Power Purchase Agreement, C.4 RALF: Regression Analysis Load Forecast, B.14 REI: Reykjavik Energy Invest, C.4 t/h: tonnes per hour, C.10 TDS: Total Dissolved Solids, C.10 TEM: Transient Electro-Magnetic, C.14 UE: Unserved Energy, F.3 UNDP: The United Nations Development Programme, C.3 USc: US cents, C.26 USD: US Dollar, D.7 UTC: United Technologies Company, C.26 VA: Value Added, F.3 WWA: World Wind Atlas, D.3 Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 APPENDIX A STUDY TERMS OF REFERENCE Republic of Djibouti Consultant Services for the Preparation of an Electricity Sector Least-Cost Master Plan TERMS OF REFERENCE I. Context of the Assignment I.1. Sector Background Djibouti is characterized by a large urban population. About 70% of the population lives in the main town of Djibouti-Ville, 11% live in secondary towns and the remainder in a rural setting, including a substantial nomadic population. The country’s electrification rate is about 50%. Electricité de Djibouti (EDD), the national state-owned utility, reports about 38,000 electricity connections for the Djibouti-Ville metropolitan area. Djibouti is also characterized by a total reliance on imported oil products as the fuels for electricity generation and has no hydroelectric potential. This has implied very high costs of production and of electricity generation in particular. In the context of electricity, Djibouti’s main (Boulaos) and secondary (Marabout) power plants, all running on costly diesel, currently offer together a total installed generation capacity of 95 megawatt (MW) to the only main interconnected system that serves Djibouti-Ville and surrounding areas, including the nearby towns of Arta, Oueah and Damerjog. There are a further four small diesel-fueled generating stations serving the southern region, consisting of Dikhil and Ali-Sabieh, which are interconnected and between them have a nominal installed capacity of 3,328 kilowatt (kW); and the northern region, serving Tadjourah (2,240 kW) and Obock (1,760 kW), amounting together to another total installed capacity of 7.3 MW. Despite already very high electricity tariffs (over 30 cent/kWh in average), the financial performance of the sector, and of EDD in particular, has been deteriorating over the past few years (negative net profit since 2004), due to various inefficiencies, staffing costs, and, by far the biggest burden, high fuel costs, which are severely impacting the utility’s cash-flows and capacity to ensure proper maintenance and new investments. I.2. Specificities of Djibouti’s electricity sector Very high cost of the electricity for Djibouti. Djibouti’s tariff for electricity is very high (US$0.30/KWh on average) and among Africa’s highest. The main reasons behind the high tariff are increasingly expensive production costs due to a total reliance on imported diesel and units pushed well beyond acceptable operating conditions, persistently high oil prices, and inefficiencies in administration and distribution of power and overstaffing. A sector in financial trouble. Despite already very high electricity tariffs, the financial performance of the sector, and of the public utility Electricité de Djibouti (EDD) in particular, has been deteriorating over the past four years (net loss of US$ 12 millions estimated for 2007), due to various inefficiencies, staffing costs, and, by far the biggest burden, high fuel costs (which doubled over 2004-2007, now representing 40% of EDD operating costs). Such situation has serious macro implication, as the government must mobilize a fair amount of its national budget in various subsidies and transfers to EDD. Low access rate. Due to the high cost of electricity and high connection fees, the electrification rate remains relatively low and mostly available to the privileged, while performance of critical social and commercial sectors are hampered. In 2007, only 38,000 customers were connected to the electricity grid and the electrification rate was about 57% in urban areas. A very small electric system. The transport network is constituted of about 250 km of medium-voltage 20KV medium-voltage lines, as well as one 4.8 km 63KV high- voltage line linking the capital’s two main stations (Boulaos and Marabout). Nine 20KV lines depart from each main station toward the city’s surroundings (half of which underground). Close to 300 HT/LT substations are spread throughout the city. The 350-km distribution network is mostly airborne and an average of 2,000 new connections is added every year. The exact amount of loss is uncertain and should be clarified upon the conclusion of an ongoing loss study, but it is currently estimated at about 22% for 2006. Environmental costs. As many as 90% of the population connected to the network are estimated to be using kerosene and/or fuel-wood for cooking due to the high cost, and everyone without a connection is using either kerosene or fuel-wood. In-door air pollution is also increasingly being recognized as a cause of health problems. Strong demand growth. Over recent months, Djibouti’s economy has been faced with an increasing inflow of foreign capital, primarily from Gulf Countries investors, coming to support numerous new projects, some expected to consume large quantities of electricity. Such projects range from new developments in the very active free zone, at the fast-expanding port terminal facilities, or at the new regional livestock triage center, to the various real estate, industrial and hotel projects; Generation capacity stretched thin. The average available capacity in 2006, when taking into account the derating for ambient operating conditions and age of plants, the operating rules applied, and an increased number of technical problems, dropped to about half its installed potential (55 MW) in 2006. That same year, peak demand reached 53 MW. Today’s situation has worsened and available capacity is now regularly insufficient to cover demand. Given the current rate of economic growth, including various large-scale ongoing infrastructure developments, new capacity is needed urgently. Ongoing investments in the electric interconnection with Ethiopia (AfDB project) should fulfill a fair amount of such need in the mid-term (mid 2010) and renewed momentum on the development of geothermal (with the support of Argeo, IGA, Proparco and IFC among others) should do so in the longer term, while the short-term remains highly uncertain. Djibouti is well-endowed with wind resources. Wind offers promising short-and- medium-term opportunities, as comprehensive wind data and feasibility work for one site in Arta is readily available while partial data and pre-feasibility work for about 10 other sites exist, some of these showing very good wind conditions. Djibouti also possesses good potential for geothermal. A lot of preliminary exploration work has been carried, but further testing up-front is needed, which is rather expensive. Until a complete investigation of that resource has been undertaken, it is not possible to know the exact size of a plant that could be supported, nor what the final investment costs will be. However, a Memorandum between the Government of Djibouti and the Icelandic Reykjavik Energy Invest Ltd (REI) has recently been signed, which paves the way for further drilling and lays out the tentative terms for an IPP arrangement, should it be proved that exploitation is possible and profitable. Better planning and a comprehensive least-cost electricity sector master plan are overdue. Djibouti has been pursuing numerous opportunities to improve its capacity to respond to its electricity needs, current and future, both in terms of infrastructure and financial viability (interconnection with Ethiopia; wind development; replacement and/or expansion of its traditional generation capacity; desalination; geothermal temporary licensing). These initiatives, however, often lack coordination and prioritizing. In fact, at the moment, no serious electricity sector planning capacity exists at the Government level. The last true energy-wide master plan was primarily designed locally and dates back to 1987, while the last load forecast, produced by EdF, dates back to 1997. The only proxy to sector planning today can be found at EDD, but it remains produced manually and on an ad-hoc basis, without using any modeling nor, by extension, cost and technology optimization. Energy sector is a top-priority for the government. Improving the performance and the organization of the energy sector has, and remains, a key priority for Djibouti, and the Government in particular. This level of priority is driven by the high burden of energy costs on the economy, on public finances, and, by extension, on a large share of the population. II. Objectives The main objectives of the assignment are: To define the least-cost investment program for the development of Djibouti’s electric generation, transmission and distribution system for the next 25 years, particularly taking into consideration the country’s resources and recent economic and sector developments. Particular attention and the greatest level of details should be given for the short-term forecast period (first 5 years) of the plan. The study will primarily include the main grid, but also the few isolated systems. To provide to EDD and the Government of Djibouti a comprehensive report, model and database for the further development of its systems and updates of the plan as needed; To provide to EDD and the Ministry of Energy & Natural Resources (MENR) with some basic planning capacity and tools to update some key components of the master plan as needed. Ultimately, the master plan would aim at improving the sector’s financial performance and fiscal impact on government budget through a more rational and economical planning of investments, while equipping the sector with a strategic, integrated, and cost- efficient long-term plan of the sector's priority investments. This plan will also be relied upon by the Government in all future discussions with donors or investors interested in the sector. III. Scope of Services The consultant will provide the services, reports, goods and other items as set froth herein and any other particulars as may be reasonable required to carry out the preparation of the electricity sector master plan and perform the work in accordance with prevailing international standards and best practices in integrated resource planning for electric utilities. III.1. Data Collection and Planning Criteria III.1.a. Data Collection The Consultant shall first compile information and data as well as review previous reports, all readily available upon starting the assignment. It will also meet with EDD, the Ministry of Energy & Natural Resources, the Ministry of Finances, CERD, and if possible, the World Bank, and carry out such other work as it deems necessary to carry out the study and develop a detailed plan and schedule for the study. Previous Studies – The Consultant will review, and where missing collect, all available data and reports relevant to the study, including previous master plan and load forecast work, donor-funded studies and ongoing project documents (particularly materials relating to the Ethiopia-Djibouti interconnection work, the wind and geothermal development, and ongoing electric system loss technical assistance), assessments and data on indigenous energy resources, etc. A comprehensive list (and copies) of these resources will be provided to the Consultant upon signing of the contract. Energy Resources - The Consultant will review reports on indigenous energy resources such as wind, geothermal, solar, issued since the last Master Plan was prepared and assess to which degree each indigenous energy resource should be included in the development of the optimum Least-Cost Plan. Demand-side Management – Where deemed relevant, the consultant will review opportunities for selected and most appropriate demand-side management programs (ex. shift to CFL lamps, change in public lighting specifications) and factor both the estimated cost and benefits of these programs into the least-cost plan. Existing Power System - review and evaluate operating and maintenance data, and technical characteristics (ex. available vs. installed capacity, type and age of systems, fuels used, rate of power outages, etc.), for existing power plants (or stand-alone diesel units in isolated areas), and transmission and distribution lines. III.1.a. Planning Criteria EDD wishes the Consultants to give particular attention to reviewing and recommending criteria for defining reliability of power supply in Djibouti. Attention should be given to the reliability and likely availability of thermal generating units and EDD’s total lack of experience with interconnections and any other sources of generation. The Consultants should review the planning criteria used in previous studies and currently adopted by EDD and should recommend changes if considered appropriate. These criteria should include determination of dependable Vs installed capacity, outage rate, reserve requirements, Loss of Load Probability (LOLP), utilization factor for thermal power stations and similar criteria. The consultant should provide a review of alternative probabilistic planning criteria and recommend a criteria that is suited and applicable to the EDD power system. III.2. Load Forecast The Consultants should review the growth of energy demand on the EDD system in recent years and all available information related to the ongoing, committed or envisaged initiatives that are expected to impact the demand, such as the important ongoing expansion plans at the Doraleh port, including planned refinery facilities, various real estate (industrial, residential and tourism) projects and business activities, and the construction of a self-generation diesel plant in the free zone (which could lead to a drop in demand from the main grid). The Consultants should then develop realistic energy and maximum demand projections; firstly at national level and then disaggregated to branch, sub-stations, grid and national levels, taking explicit account of specific revisions in industrial and commercial expansion plans. Typical weekday and weekend load curves for the present transmission grid system and for representative isolated systems should be established, along with load-duration diagram. Projected future changes to these curves should be established. In carrying out the forecasts the following will be taken into account: The effects of transmission grid extensions, interconnections, power imports, distribution system expansion plans and system losses (and realistic reductions thereof) should be determined. Three forecasts should be prepared, namely, "base ", "high� and "low", and the basis of each forecast should be explained. The contributions of the individual components underlying the global forecasts, including the effects of interconnection and possible exports or imports, should be separately identified and tabulated. Each forecast scenario should clearly differentiate the short-term (5 first years) from the medium and long-term (5 to 25 years). Short-term forecasts should include projected loads at branch, sub-stations, grid, isolated load centers where applicable, and national levels on a year-by-year basis for the first five years, broken down into major load categories, down to the 63kV network and the 63/20kV substation level. Where applicable, the short-term demand forecasts should take into account review of committed or planned investments in each major consumption category. The Consultant shall compile information and data as well as review previous reports, all readily available upon starting the assignment. It will also meet with EDD, the Ministry of Energy & Natural Resources, the Ministry of Finances, CERD, and if possible, the World Bank, and carry out such other work as it deems necessary to carry out the study and develop a detailed plan and schedule for the study. Demand side management in household, government, expatriated (including military), commercial, and industry categories should be examined and its impact on the load forecast determined. III.3. Supply-Side Options - Electricity Generation The Consultant will review existing data and studies and define a set of generation candidates; define the technical characteristics, capital cost, operating cost, efficiency and the lead-time for each candidate; and assess the technical feasibility of the candidate and include them in the development of the optimum least-cost plan. If relevant, the estimated costs of supply-side options proposed in previous studies should be revised to January 2008 price levels. Such assessment will look at the suitability, availability, financial viability, and lead-time of increasing existing capacity (diesel-fueled) and of pursuing relevant alternative technologies that would optimally contribute to additional generation capacity, should the preparation of their development be already well advanced in Djibouti (wind, interconnection, geothermal), be considered (co-generation, desalination) or represent potentially good new candidates (ex. solar, natural gas, the autonomous generation plant developed by the Doraleh port). As part of this evaluation, attention should also be given to the potential for using solar and wind power as sources for isolated power plants systems. Throughout this step, considerable attention will need to be given to the planned 220 kV electric interconnection line between Ethiopia and Djibouti, expected to be operational by early 2010. This important development should have a major impact on Djibouti’s supply mix, as well as on its current grid, as it will extend transmission lines to new areas and load centers. Giving attention to the recently-signed MOU and draft PPA with REI on geothermal energy will also be essential. Similarly, evaluating the feasibility and potential for efficiency improvement of pursuing efforts to promote the development of additional installed capacity by Doraleh port authorities, in the context of an electricity purchase agreement with EDD that would allow the latter to benefit from cheaper and investment-free electricity, will be necessary. The Consultant will also assess scope and indicative costs of rehabilitation of present generating thermal units including the possible conversion of existing thermal units to other fuels (ex. natural gas). The recommendations for rehabilitation should take into consideration the economic cost of future maintenance. It will be critical that environmental, social and economic impacts, as well as an analysis of other associated risks, are also taken into consideration when evaluating these various supply options. III.4. Review of Supply-side Options A two tiered approach to the analysis of least-cost supply-side options will be applied in order to identify the best power options available to meet EDD’s load forecast, taking into consideration the multi-purpose nature of some of the available options: First, a screening process of all technologies available will be undertaken based on factors such as availability of the technology to meet the identified project objectives, availability of resource requirements (at a macro-level), suitability in a particular situation, financial viability, and, where relevant, the potential institutional and legal prerequisites, etc. Secondly, screening of alternative locations will consider the availability to meet demand, resource requirements for short-listed technologies, and broad environmental, social and economic considerations. This second screening will allow to eliminate those power options with unacceptable social and environmental consequences such as significant conversion or degradation of natural habitats, significant involuntary resettlement/loss of livelihoods and significant regional impacts up- and downstream. This second screening process should define a realistic range of alternatives for further consideration in the next steps. A methodology to compare options at different levels of maturity will be developed to avoid unjustifiably eliminating options which may be good but could be handicapped if they are compared with options that are at a more advanced study level. III.5. Review of Transmission and Distribution Options Based on projected demand and optimal supply options, a comparative analysis and least- cost optimization of electric transmission and distribution development options will be carried out, down to the 63kV network and the 63/20kV substation level. Such assessment will have to give particular consideration to existing planned rehabilitation plans, load flows, and system stability constraints. Similarly to the supply-side, it will be critical that environmental, social and economic impacts, as well as an analysis of other associated risks, are also taken into consideration when evaluating these various transmission/distribution options; III.6. Training The Government, and the Ministry of Energy & Natural Resources and EDD in particular, recognize the need to better integrate the few existing planning functions (mostly at EDD today) while creating an embryo of planning capability at the Ministry of Energy & Natural Resources, closely assisted by EDD (and, if relevant, the CERD), to enable it to carry out more planning activities in-house in the future. As part of a “learning by doing� approach, the Working Group will identify 4 counterpart staff members issued from the Ministry of Energy & Natural Resources and from EDD, who will participate in the Consultant work during its in-the-field visit, as well as benefit from its training. The training provided by the consultant will include: A 2-day training course on load forecasting methodologies, generation reliability criterion and generation planning techniques; A 3-day training course to introduce the counterpart staff to the proposed security rules, network planning and the basic fundamentals of network analysis and the necessary data requirements. The courses should highlight the deficiencies in the existing planning capability of EdD and introduce the techniques the consultant will use to update the least cost plan. These courses would take place during the consultant’s mission to Djibouti and would comprise both formal lectures and hands-on training exercises. The content of the course will be dictated primarily by the competencies of the assigned counterpart staff. It is understood that all counterpart staff must be computer literate and be fully familiar with the use of Microsoft Office Applications, particularly MS Excel. Ideally, by the end of the course, the trained staff trained by the Consultant may have a limited ability to update the load forecast model and be familiar with the concepts on which the generation plan is based. The Consultants shall, at the end of the study, hand over to EDD all computer models, computer applications, and the database used for the execution of the study. It should also supervise the installation, startup and initial operation of the computer systems at EDD's offices. Licenses for proprietary network modeling software will not be included, with the exception of one @Risk Professional license The Consultants shall also conduct one information dissemination workshop in Djibouti to (i) briefly educate the key sector stakeholders, in simple terms, to the methodology, process, requirements, and challenges behind the Master Plan; (i) report on the key findings and conclusions of its final report; and (iii) give the audience an opportunity to seek clarifications or ask questions. IV. Main Deliverables Inception Report: Confirming planning criteria & training arrangements (by e-mail) Technical Memorandum: Preliminary Load Forecast & Supply-Side Options with assumptions Interim Report: Finalized Load Forecast & Supply-Side Options + Review of the supply options, transmission and distribution options, and their associated institutional and financial requirements Final Report: Final report integrating the above and comments received from Client, including a strategic executive summary (with clear investment plans under main scenarios), as well as any documentation and manuals relating to the planning tools developed during the assignment in order to allow for future plan updates. In addition to the reports above, the Consultant will ensure that the agreed training is provided and, in particular, that: The computer planning model and applications, as well as any agreed licenses and manuals are provided to the Working Group. It leads one dissemination workshop to present its key findings to the key sector stakeholders (most likely officials from the various ministries and agencies represented on the High-Level Working Group and utility staff) V. Reporting & Validation V.1. Reporting The Consultant’s primary reporting line is the World Bank. However, given the strategic stake of this exercise for the government, it will also work in very close collaboration and consultation with the High-Level Working Group that has been setup to ensure as much ownership, coordination, integration, and ownership of this effort. This Working Group is chaired by the Secretary General of the Ministry of Energy & Natural Resources, who will also serve as the main liaison agent with the Consultant and the Bank, and is composed of representatives from: Ministry of Finances Prime Minister's Office (“Primature�) EDD CERD (National Center for Research and Studies) UGPE (PMU of the current IDA-funded power project) Other entities, as needed These representatives, or colleagues they would delegate, would be the primary beneficiaries of all planning knowledge transfer. This Working Group will be particularly instrumental in the data collection phase and in any sensitive validation step of the exercise. V.2. Validation In order to avoid any loss of trust in the Consultant’s deliverables and of ownership of the planning process on the Djiboutian side, close collaboration with the Working Group during the assignment and a validation process of key deliverables will be necessary. Therefore, while all deliverables will be shared by the Bank with the Working Group for its comments, the Bank will seek the Working Group’s official validation of the Technical Memorandum and the Final Report. VI. Estimated Effort While several aspects of the assignment will consist in desk work, the Consultant is also expected to spend as much time as deemed necessary in the field and with the Djiboutian counterparts. The total effort for this assignment has been estimated at between 6 and 7 men months. VII. Resources Provided by the Beneficiary In order to facilitate the Consultant’s work in the field, the current Project Implementation Unit (UGPE) for IDA-financed projects in Djibouti, in coordination with the Ministry of Energy & Natural Resources, will ensure that car transportation, as well as access to their office (located in one of the Utility’s premises) for access to basic secretarial support, the Internet and local phone, is made available as needed. VIII. Terms of Payments The following terms of payment will be followed: Receipt of Inception Report: 20% of contract amount Validation of the Technical Memorandum: 20% of contract amount Receipt of Interim Report: 25% of contract amount Validation of Final Report: 30% of contract amount Workshop delivered: 5% of contract amount APPENDIX B SALES FORECAST APPENDIX B SALES FORECAST B. SALES FORECAST B.1 Introduction In this Appendix we present the results of the analysis for the regression based sales forecast prepared as part of this study. We then discuss the inclusion of specific ‘additional loads’ as identified by Electricité de Djibouti (EdD). Finally, we bring together the regression based sales forecast and the ‘additional loads’ to produce a modified sales forecast. B.2 Regression based sales forecasts B.2.1 Social The Social consumer category is a lifeline tariff category and as such does not exhibit the same characteristics as other consumer categories. For this reason a regression based demand forecast has not been developed for this consumer category. Instead, for this study we have projected social electricity sales to develop in line with national population, whilst assuming a constant level of specific consumption. It is reasonable to assume that the specific consumption of each connection in this consumer category is unlikely to alter significantly. This assumption is based on the notion that if this type of consumer were able to consume more, then they would no longer remain in the Social consumer category. In order to derive the forecast of social electricity sales it has been assumed that there are approximately 12 persons per defined connection. By multiplying the number of registered connections by the assumed number of people per household it is possible to identify the total number of people that have access to electricity in this consumer category (approximately 206,664). For this forecast the ratio between the total number of people that have access to electricity in the social consumer category to the national population (approximately 24 per cent in 2008) is assumed to remain constant into the future. Using the national population forecast and the defined ratio above, it is possible to determine the number of people in the social consumer category with access to electricity and in turn (using the assumed number of people per household) the number of social consumer connections. By assuming a constant level of specific consumption (assumed as the 13 year average between 1996 and 2008 of 1392 kWh per connection) and multiplying this by the number of connections, the forecast of social sales can be derived. This forecast of sales is presented in Table B.1 below. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.3 APPENDIX B SALES FORECAST Table B.1: Social sales forecast Social Sales Growth Year (GWh) (%) 2008 25.0 2009 24.4 -2.4% 2010 24.8 1.7% 2011 25.2 1.7% 2012 25.7 1.7% 2013 26.1 1.6% 2014 26.5 1.6% 2015 26.9 1.6% 2016 27.3 1.6% 2017 27.7 1.5% 2018 28.2 1.5% 2019 28.6 1.5% 2020 29.0 1.5% 2021 29.5 1.6% 2022 29.9 1.6% 2023 30.4 1.6% 2024 30.9 1.6% 2025 31.4 1.5% 2026 31.8 1.5% 2027 32.3 1.4% 2028 32.7 1.4% 2029 33.2 1.4% 2030 33.7 1.4% 2031 34.1 1.3% 2032 34.5 1.3% 2033 35.0 1.3% 2034 35.4 1.2% 2035 35.8 1.2% B.2.2 Domestic We have projected the level of electricity sales for the domestic sector using the multi-variable regression analysis model with the economic driver of total Gross Domestic Product (GDP). Recognising the impact of the fire (in 1998) on domestic energy sales, we have included a ‘dummy Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.4 APPENDIX B SALES FORECAST variable’ which essentially neutralises the effect of the ‘fire’ on the variability in sales, such that a reasonable statistical relationship could be identified. Table B.2 shows the results of this analysis. Table B.2: Equation for Domestic Sales Regression Statistics Multiple R 0.888577132 R Square 0.789569319 Adjusted R Square 0.747483183 Standard Error 0.049806596 Observations 13 ANOVA df SS MS F Significance F Regression 2 0.093079688 0.046539844 18.76079368 0.000412615 Residual 10 0.02480697 0.002480697 Total 12 0.117886658 Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept -4.30486603 1.63978916 -2.62525582 0.025367865 -7.958543953 -0.651188112 -7.958543953 -0.651188112 Log Total GDP (DJFm) 1.247750157 0.337050605 3.70196682 0.004095192 0.496754612 1.998745702 0.496754612 1.998745702 Log Fire Dummy -0.19169038 0.054421085 -3.52235504 0.005516713 -0.312948117 -0.07043265 -0.312948117 -0.07043265 The analysis shows a good level of statistical fit (R2 = 0.79), particularly when considering the high degree of variability in the historical sales of electricity during the past 13 years in this sector. The elasticities are appropriate in terms of both the sign and magnitude and the P value is low. Using the statistical output for the given relationship the forecast for domestic sales was calculated as shown in Table B.3. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.5 APPENDIX B SALES FORECAST Table B.3: Forecast of Domestic Sales Domestic Sales Growth Year (GWh) (%) 2008 69.1 2009 76.3 10.3% 2010 79.6 4.4% 2011 83.1 4.4% 2012 86.7 4.4% 2013 90.5 4.4% 2014 94.5 4.4% 2015 98.7 4.4% 2016 103.0 4.4% 2017 107.5 4.4% 2018 112.2 4.4% 2019 117.1 4.4% 2020 122.3 4.4% 2021 127.5 4.3% 2022 132.9 4.3% 2023 138.6 4.3% 2024 144.5 4.3% 2025 150.6 4.3% 2026 156.7 4.0% 2027 163.0 4.0% 2028 169.5 4.0% 2029 176.3 4.0% 2030 183.3 4.0% 2031 190.2 3.8% 2032 197.4 3.8% 2033 204.8 3.8% 2034 212.5 3.8% 2035 220.5 3.8% B.2.3 LV Djibouti We have projected the level of electricity sales for the Low Voltage (LV) Djibouti consumer category using the multi-variable regression analysis model with the economic driver of population and, recognising the impact of the fire in 1998 on general < 36 energy sales, we have introduced a dummy variable to neutralise the effect of the fire on the variability in sales. Table B.4 shows the results of this analysis. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.6 APPENDIX B SALES FORECAST The analysis shows an excellent level of statistical fit (R2 = 0.95), and the elasticities are appropriate in terms of both the sign and magnitude. The P value is very low indicating that the inclusion of population as a driver is accurate particularly as we would expect the level of commercial activity to increase as the population increases. Table B.4 : Equation for LV Djibouti Sales Regression Statistics Multiple R 0.974141397 R Square 0.948951462 Adjusted R Square 0.938741754 Standard Error 0.028353729 Observations 13 ANOVA df SS MS F Significance F Regression 2 0.149444885 0.074722442 92.94599726 3.4667E-07 Residual 10 0.008039339 0.000803934 Total 12 0.157484224 Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept -5.39117745 0.652778221 -8.25881943 8.89964E-06 -6.845657963 -3.936696945 -6.845657963 -3.936696945 Log Population (000's) 2.438523835 0.226659197 10.75854793 8.10096E-07 1.933495674 2.943551996 1.933495674 2.943551996 Log Fire Dummy -0.14368509 0.031039436 -4.629114 0.000937442 -0.212845258 -0.074524914 -0.212845258 -0.074524914 Using the statistical output for the given relationship the forecast for General (<36 kVA) sales was calculated as shown in Table B.5. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.7 APPENDIX B SALES FORECAST Table B.5: Forecast of LV Djibouti Sales LV Djibouti Sales Growth Year (GWh) (%) 2008 63.0 2009 58.9 -7% 2010 61.3 4.2% 2011 63.9 4.2% 2012 66.5 4.1% 2013 69.2 4.0% 2014 71.9 4.0% 2015 74.7 3.9% 2016 77.6 3.8% 2017 80.5 3.8% 2018 83.5 3.7% 2019 86.5 3.7% 2020 89.7 3.6% 2021 93.3 4.0% 2022 97.0 4.0% 2023 100.8 3.9% 2024 104.6 3.8% 2025 108.6 3.8% 2026 112.5 3.6% 2027 116.5 3.5% 2028 120.6 3.5% 2029 124.7 3.4% 2030 128.9 3.4% 2031 133.0 3.2% 2032 137.2 3.1% 2033 141.5 3.1% 2034 145.8 3.1% 2035 150.2 3.0% B.2.4 Public Lighting The sales input data for the public lighting sector illustrates a high degree of variability, with no clear trend and as such has proved difficult to model. However, the forecast level of electricity sales for the pubic lighting consumer category has still been developed using the multi-variable regression analysis Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.8 APPENDIX B SALES FORECAST model with an economic driver of total GDP. A dummy variable to represent the fire has also been included in this regression. Table B.6 shows the results of this analysis. Table B.6: Equation for Public Lighting Sales Regression Statistics Multiple R 0.770495184 R Square 0.593662829 Adjusted R Square 0.512395394 Standard Error 0.07714461 Observations 13 ANOVA df SS MS F Significance F Regression 2 0.086948977 0.043474488 7.305051944 0.011077271 Residual 10 0.059512908 0.005951291 Total 12 0.146461885 Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept -4.52478902 2.539842226 -1.78152366 0.105166365 -10.18391014 1.134332094 -10.18391014 1.134332094 Log Total GDP (DJFm) 0.998415953 0.522052091 1.912483392 0.084856202 -0.164788588 2.161620495 -0.164788588 2.161620495 Log Fire Dummy -0.21667759 0.084291916 -2.57056192 0.027863797 -0.404491681 -0.028863497 -0.404491681 -0.028863497 The analysis shows a low level of statistical fit (R2 = 0.59), but the elasticities are appropriate in terms of both the sign and magnitude. Given the relative contribution to total sales (approximately 1 per cent) it is considered acceptable to use this regression to forecast electricity sales. Using the statistical output for the given relationship the forecast for Public Lighting sales was calculated as shown in Table B.7. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.9 APPENDIX B SALES FORECAST Table B.7: Forecast of Public Lighting Sales Public Lighting Growth Year Sales (GWh) (%) 2008 2.7 2009 2.7 -1% 2010 2.8 3.5% 2011 2.9 3.5% 2012 3.0 3.5% 2013 3.1 3.5% 2014 3.2 3.5% 2015 3.3 3.5% 2016 3.4 3.5% 2017 3.5 3.5% 2018 3.6 3.5% 2019 3.8 3.5% 2020 3.9 3.5% 2021 4.0 3.4% 2022 4.2 3.4% 2023 4.3 3.4% 2024 4.4 3.4% 2025 4.6 3.4% 2026 4.7 3.2% 2027 4.9 3.2% 2028 5.1 3.2% 2029 5.2 3.2% 2030 5.4 3.2% 2031 5.5 3.0% 2032 5.7 3.0% 2033 5.9 3.0% 2034 6.1 3.0% 2035 6.2 3.0% B.2.5 Chantier We have forecast the level of electricity sales for the ‘chantier’ sector using the multi-variable regression analysis model with the economic driver of population and a dummy variable to isolate the impact of the fire which occurred in 1998. Table B.8 shows the results of this analysis. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.10 APPENDIX B SALES FORECAST Table B.8: Equation for ‘chantier’ Sales Regression Statistics Multiple R 0.609592418 R Square 0.371602917 Adjusted R Square 0.2459235 Standard Error 0.188565924 Observations 13 ANOVA df SS MS F Significance F Regression 2 0.210267128 0.105133564 2.956752399 0.097987528 Residual 10 0.355571078 0.035557108 Total 12 0.565838206 Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept -10.3501622 4.341288891 -2.38412197 0.038341588 -20.02315662 -0.677167817 -20.02315662 -0.677167817 Log Population (000's) 3.608068732 1.507392592 2.393582634 0.037727017 0.249388746 6.966748717 0.249388746 6.966748717 Log Fire Dummy 0.237363619 0.206427165 1.149866197 0.276955439 -0.222584765 0.697312002 -0.222584765 0.697312002 The analysis shows a low level of statistical fit (R2 = 0.37), particularly when considering the high degree of variability in the historical sales of electricity during the past 10 years in this sector. The elasticities are appropriate in terms of both the sign and magnitude, whilst the P value is low. Given the relative contribution to total sales (approximately 1 per cent) it is considered acceptable to use this regression to forecast electricity sales. Using the statistical output for the given relationship the forecast for ‘chantier’ sales was calculated as shown in Table B.9. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.11 APPENDIX B SALES FORECAST Table B.9: Forecast of ‘chantier’ Sales Chantier Sales Growth Year (GWh) (%) 2008 2.3 2009 1.8 -23% 2010 1.9 6.3% 2011 2.0 6.2% 2012 2.1 6.1% 2013 2.2 6.0% 2014 2.4 5.9% 2015 2.5 5.8% 2016 2.6 5.7% 2017 2.8 5.6% 2018 3.0 5.5% 2019 3.1 5.5% 2020 3.3 5.4% 2021 3.5 6.0% 2022 3.7 5.9% 2023 3.9 5.8% 2024 4.1 5.7% 2025 4.4 5.7% 2026 4.6 5.4% 2027 4.8 5.3% 2028 5.1 5.2% 2029 5.3 5.1% 2030 5.6 5.1% 2031 5.9 4.7% 2032 6.2 4.7% 2033 6.4 4.6% 2034 6.7 4.6% 2035 7.0 4.5% B.2.6 MV Djibouti We have projected the level of electricity sales for the Medium Voltage (MV) Djibouti consumer category using the multi-variable regression analysis model with the economic driver of population and a dummy variable to isolate the impact of the fire which occurred in 1998. Table B.10 shows the results of this analysis. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.12 APPENDIX B SALES FORECAST The analysis shows a strong level of statistical fit (R2 = 0.82), and elasticities which are both appropriate in terms of the sign and magnitude. The P value is low indicating that the inclusion of population as a driver is accurate. The inclusion of population as a driver for MV Djibouti sales is in line with our expectations that, given the size of the MV Djibouti category (the sector accounts for almost half of all sales); any increase in population will drive the level of MV Djibouti sales. It is worth noting however that other regressions were identified, however, given the rapid growth in this sector in recent years, the modelling of these relationships would have led to unrealistic growth rates which would not be sustainable over the medium to long term. Table B.10: Equation for MV Djibouti Sales Regression Statistics Multiple R 0.906440157 R Square 0.821633758 Adjusted R Square 0.78596051 Standard Error 0.049283983 Observations 13 ANOVA df SS MS F Significance F Regression 2 0.111886378 0.055943189 23.03221033 0.000180536 Residual 10 0.02428911 0.002428911 Total 12 0.136175488 Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept -5.25162567 1.08046354 -4.8605302 0.00066089 -7.659048448 -2.844202883 -7.659048448 -2.844202883 Log Population (000's) 2.467893954 0.375740356 6.56808329 6.32704E-05 1.630692272 3.305095636 1.630692272 3.305095636 Log MV Djibouti Specific Dumm -0.08437774 0.0380021 -2.22034405 0.050664999 -0.169051691 0.000296219 -0.169051691 0.000296219 Using the statistical output for the given relationship the forecast for MV Djibouti sales was calculated as shown in Table B.11. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.13 APPENDIX B SALES FORECAST Table B.11: Forecast of MV Djibouti Sales MV Djibouti Sales Growth Year (GWh) (%) 2008 80.5 2009 99.0 23% 2010 103.2 4.3% 2011 107.6 4.2% 2012 112.0 4.1% 2013 116.6 4.1% 2014 121.3 4.0% 2015 126.0 3.9% 2016 130.9 3.9% 2017 135.9 3.8% 2018 141.0 3.8% 2019 146.2 3.7% 2020 151.6 3.6% 2021 157.8 4.1% 2022 164.1 4.0% 2023 170.6 4.0% 2024 177.2 3.9% 2025 184.0 3.8% 2026 190.7 3.6% 2027 197.6 3.6% 2028 204.5 3.5% 2029 211.7 3.5% 2030 218.9 3.4% 2031 226.0 3.2% 2032 233.2 3.2% 2033 240.5 3.1% 2034 247.9 3.1% 2035 255.5 3.1% B.3 Additional loads Using the Regression Analysis Load Forecast (RALF) model we have developed regression based sales forecasts for each of the six defined consumer categories (Social, Domestic, LV Djibouti, Public Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.14 APPENDIX B SALES FORECAST 1 Lighting, Chantier and MV Djibouti ). We have been made aware that there are a number of relatively large committed projects which will come ‘on line’ within the next ten years. These include: • The Kempinski hotel development (to be completed in early 2009), • The Delta Doraleh residential development, • Heron City development, • Haramous City development, • A water desalination plant, • American Embassy, • The free zone, • Universities at Balbala and Malayslenne, • Business centres A and B • The Doraleh container, and, • A large cement factory. Table B.12 below presents EdD’s estimate of the size (demand for power, MW) and timing of these committed projects. 1 It should be noted that the forecast of Social sales has not been developed using regression analysis. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.15 APPENDIX B SALES FORECAST Table B.12: EdD estimate of additional loads Load Growth Distribution (MW) Total Development Load 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 (MW) Kempinski 1.0 1.0 Haramous City 35.0 1.0 2.0 2.0 2.0 3.0 5.0 5.0 5.0 5.0 5.0 Free Zone 10.0 2.0 3.0 5.0 Delta Doraleh 2.0 2.0 Doraleh Container 8.0 8.0 Ali Sableh Cement 5.0 5.0 Desalination Plant 30.0 3.0 27.0 Universite de Balbala 3.0 1.0 2.0 American Embassy 1.0 1.0 Heron Al Boum 34.0 2.0 3.0 4.0 5.0 5.0 5.0 5.0 5.0 Business Center A & B 4.0 4.0 Universite Malayslenne 2.0 2.0 135.0 4.0 7.0 22.0 39.0 13.0 10.0 10.0 10.0 10.0 10.0 0.0 0.0 Analysis of the additional load data provided by EdD highlights two key issues. Firstly, the sizing of loads induced by some of the projects identified seem to be excessive (particularly for maximum demand planning purposes), and secondly, some of project loads identified are likely to be incorporated within our regression based approach to sales forecasting and therefore the estimates of power demand need to be either revised downwards or removed from the list of additional loads. In line with the above, we present our assumption relating to the size and timing of additional loads incurred from significant committed projects that are not assumed to have been incorporated within the regression based sales forecast in Table B.13 below. We have based these estimates on our knowledge and professional judgement of the specific committed projects and similar projects that have been implemented elsewhere. It should be noted here however, that the additional load incurred by the development of the desalination plant has been modelled as a new consumer category with its own Coincident After Diversity Load Factor (CADLF) (assumed to be 75 per cent). The desalination plant is a significant load to be added to such a small system and by including it within the MV Djibouti (which has a CADLF of 100 per cent) would result in the over-estimation of its load factor and thus incorrectly increase the system load factor significantly. For this reason, the additional load incurred by the development of a desalination plant is modelled as a separate consumer category. The assumed ‘additional loads’ for this study are presented in Table B.13 below. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.16 APPENDIX B SALES FORECAST Table B.13: Summary of additional Loads Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.17 APPENDIX B SALES FORECAST Following the initial regression analysis, we have therefore modified the level of sales in the appropriate consumer categories to account for the extra demand for electricity created by these projects. B.4 Modified sales forecast Collating the results of the regression based sales forecasts and allowing for the identified additional loads above, the ‘modified’ sales forecast is shown below in Table B.14. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.18 APPENDIX B SALES FORECAST Table B.14: Modified Sales Forecast LV Public MV Desalination Year Social Domestic Chantier Total Djibouti Lighting Djibouti Plant 1996 20.1 46.3 31.5 2.6 1.1 57.4 158.9 1997 20.1 52.4 30.2 2.4 0.9 56.6 162.7 1998 12.7 32.5 24.2 1.2 1.3 45.0 116.9 1999 17.2 40.3 32.8 1.8 0.6 47.5 140.2 2000 18.6 51.6 40.9 1.8 0.9 58.8 172.6 2001 15.8 62.0 40.1 1.4 0.7 70.0 190.1 2002 18.8 62.7 43.0 2.0 0.7 66.1 193.3 2003 19.0 63.7 46.9 2.0 0.9 74.3 206.8 2004 20.2 63.9 44.4 2.3 1.1 82.9 214.9 2005 20.4 63.2 50.3 2.3 1.3 83.3 220.8 2006 21.3 64.1 50.3 2.2 1.2 92.5 231.6 2007 22.4 68.4 51.3 2.7 3.7 96.9 245.4 2008 25.0 69.1 63.0 2.7 2.3 80.5 242.6 2009 24.4 82.0 58.9 2.7 1.8 116.5 286.2 2010 24.8 85.3 61.3 2.8 1.9 147.0 323.1 2011 25.2 100.2 63.9 2.9 2.0 177.7 13.1 384.9 2012 25.7 115.2 72.2 3.0 2.1 243.4 26.3 487.8 2013 26.1 130.4 92.0 3.1 2.2 274.3 59.1 587.1 2014 26.5 140.1 106.1 3.2 2.4 296.5 92.0 666.6 2015 26.9 155.6 108.9 3.3 2.5 301.2 124.8 723.2 2016 27.3 171.3 111.7 3.4 2.6 306.1 157.7 780.2 2017 27.7 187.2 114.7 3.5 2.8 311.1 157.7 804.7 2018 28.2 203.3 117.7 3.6 3.0 316.2 157.7 829.6 2019 28.6 213.9 120.7 3.8 3.1 321.4 157.7 849.2 2020 29.0 224.8 123.8 3.9 3.3 326.8 157.7 869.2 2021 29.5 230.0 127.4 4.0 3.5 333.0 157.7 885.0 2022 29.9 235.4 131.1 4.2 3.7 339.3 157.7 901.3 2023 30.4 241.1 134.9 4.3 3.9 345.8 157.7 918.1 2024 30.9 247.0 138.8 4.4 4.1 352.4 157.7 935.4 2025 31.4 253.1 142.8 4.6 4.4 359.2 157.7 953.1 2026 31.8 259.2 146.7 4.7 4.6 365.9 157.7 970.6 2027 32.3 265.4 150.7 4.9 4.8 372.8 157.7 988.5 2028 32.7 272.0 154.7 5.1 5.1 379.7 157.7 1007.0 2029 33.2 278.8 158.9 5.2 5.3 386.9 157.7 1025.9 2030 33.7 285.8 163.1 5.4 5.6 394.1 157.7 1045.4 2031 34.1 292.7 167.2 5.5 5.9 401.2 157.7 1064.3 2032 34.5 299.9 171.4 5.7 6.2 408.4 157.7 1083.7 2033 35.0 307.3 175.6 5.9 6.4 415.7 157.7 1103.5 2034 35.4 315.0 180.0 6.1 6.7 423.1 157.7 1123.9 2035 35.8 323.0 184.4 6.2 7.0 430.7 157.7 1144.8 Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page B.19 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI C. REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI This Appendix presents our review of the geothermal resources in Djibouti. The following list of data 2 was reviewed or collected during our visit to the EdD offices (March 2009) and has been used in our analysis of the geothermal resources in Djibouti: • Virkir-Orkint (1990) - Djibouti geothermal scaling and corrosion study (Final report) • Aquater (1988) - Djibouti Assal 3 Well Final report GEOT A3237 • Aquater (1987) - Assal 3 well Report IDRO A2876 • Jalludin M (1992) - Synthese reservoir superficial du rift d Assal (CERD report) • Jalludin M - State of knowledge of geothermal provinces of Djibouti (CERD report) • Jalludin M - An overview of the geothermal prospection in Djibouti. Results and perspectives (CERD report) • A M Houmed - Republique de Djibouti, Projet de Développement géothermique (EDD report) • CFG (1993) - Champ géothermique de Assal, Djibouti, Synthese de donnees 93CFG06 • GDA (2000) - Feasibility study of the Assal Geothermal Project • The United Nations Development Programme (UNDP) DJI/86/012 - Plan Energetique National for ISERST • ISOR (2008) - Magneto-telluric survey results at Lac Assal (reviewed with Mr Jalludin in Centre d’Etudes et de Recherche de Djibouti (CERD) office, power point presentation on 30 March and quick review of report contents and conclusions on 1 April 2009 in the CERD office, no copies made available due to confidentiality agreements with Islenskar Orkurannsoknir (ISOR)). In this Appendix, we discuss the characteristics of the previously identified potential geothermal sites within Djibouti and consequently evaluate and rank their suitability and prospects for future power development. We also provide a discussion on the possible geothermal conversion technologies that 2 Some items were only partially collected. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.3 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI could be adopted in Djibouti, the typical project life cycle for a geothermal power development and the cost estimates of geothermal plant specific to those identified sites which are deemed suitable for power generation development. The unit cost of generation for each geothermal development opportunity is estimated from the short-list of options using data in existing reports and from our extensive knowledge of operation and maintenance (O&M) costs. It should be noted that we are not in a position to review and comment on the Memorandum of Understanding (MoU) signed recently with Reykjavik Energy Invest (REI) from Iceland and the draft Power Purchase Agreement (PPA) associated with potential geothermal development(s), as these documents were said to be confidential during the site visit and were not made available for review. C.1 Geothermal Power Geothermal power is defined as the power extracted from heat that is stored in the earth. This geothermal energy source originates from the original formation of the planet, from the solar energy absorbed from the sun at the earths’ surface and from the radioactive decay of minerals. Geothermal energy has been used for heating and bathing for centuries but this source of energy is now better known for its use in electricity generation. Electricity generation from geothermal resources can be produced from both the heat emanating from deep within the Earth as well as shallower sources of heat closer to the surface. Initially geothermal power production was limited to deep exploration, utilising the heat content of the Earth. The heat trapped in the Earths surface naturally flows up to the surface by conduction (and is typically replaced, although at a lesser rate, by radioactive decay). The earths crust effectively acts as a thick insulating blanket which must be pierced by fluid conduits in order to release the heat underneath. For electricity production, the heat must be carried to the surface by fluid circulation, either through magma conduits, hot springs, hydrothermal circulation, oil wells, drilled water wells, or a combination of these. This circulation sometimes exists naturally in the most favourable areas where the crust is thin: magma conduits bring the heat close to the surface, and naturally occurring hot springs bridge the last gap to the surface. If no hot spring is available, a well must be drilled into a hot aquifer. Geothermal power has historically been restricted to the above form of production and geographically limited to areas near tectonic plate boundaries. Recent technological advances however, have dramatically expanded the range and size of viable resources. Electricity production from geothermal resources typically required very high temperatures that could only come from deep underground. The evolution of binary generators however, have allowed lower temperatures to be harnessed for electricity production and in turn have allowed for shallow and more geographically diverse exploration of geothermal resources. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.4 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Geothermal power is cost effective, reliable and although geothermal wells tend to release greenhouse gases trapped deep within the earth, these emissions are much lower than those of conventional fossil fuel power plants. As a result, geothermal power has the potential to help mitigate climate change/global warming. Geothermal power is considered to be a sustainable energy resource because the heat extraction is small compared to the overall heat content of the earth. Nevertheless, it is essential that extraction is monitored to avoid local depletion. It is also essential that any by-product residues are carefully managed as these can have a negative environmental impact. Although geothermal sites are capable of providing heat for many decades, individual wells may cool down or even run out of water. If production is reduced, it is theoretically possible to re-inject water and thus the full potential of the well could be recovered. It should be noted however, that construction of the power plants can adversely affect land stability in the local region. For example, subsidence and earthquakes have been known to occur due to the presence of geothermal power production and its associated process (injection of water etc). Geothermal power production does not incur any fuel costs, but at present, capital costs tend to be high. The highest proportion of the costs of electrical plants comes from the drilling stage of production, whilst the majority of the financial risk of a geothermal development is during the exploration of deep resources. C.2 Geothermal Resources in Djibouti Djibouti is situated where three major tectonic structures meet. It is at the frontier that separates the African and Arabic tectonic plates. It is also delineated by an Oceanic Rift Valley (with divergent movements with amplitudes exceeding 10 mm a year) and the great African Rift Valley (with divergent movements of the order of 10 to 15 mm a year). Figure C.1 below show the tectonic plates movements and the geographical location of Djibouti. The combination of Djibouti’s location on the borders of active tectonic zones and the presence of natural fluids in the underground rocks (evidenced by surface manifestations) at the most favourable geographical zones of the territory have aroused the interest of geologists, geochemists and geophysicists since the late 1960’s. The exploration of Djibouti’s geothermal resources has since intensified, but this exploration has not yet resulted in the implementation of an industrial application for this energy. This is in contrast to Djibouti’s neighbouring countries, where several geothermal power plants are in operation in Kenya and one small binary plant is in operation in Ethiopia. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.5 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Figure C.1: Tectonic plate movement and map of Djibouti Figure C.2 below shows the identified geothermal sites in Djibouti at present. These are situated in the continuation of the Ethiopian rift valley from Lake Abhe (in the South West of Djibouti) up towards Obock (in the North East of the Djibouti). It should be noted that (with the exception of the Arta site) the existing EdD electricity network is concentrated around the capital Djibouti-Ville and connected to Ethiopia via a 230 kV double circuit) is not located in close proximity to any of the potential geothermal sites. Any future development will consequently have to take into account this constraint. Table C.1 below provides a summary of the main characteristics of each identified geothermal site, 3 including the type of surface manifestations (hot springs or fumaroles ), the type and extent of the investigations made to this day and if any drillings that have occurred at the site. As indicated in Table C.1, geological drilling has only taken place in the Asal and Hanlé regions of Djibouti. Table C.2 below presents the results of these drillings. It should be noted that all existing drillings shown in the table below are vertical ones and are not ‘diverted’. 3 A fumarole (Latin fumus, smoke) is an opening in Earth's (or any other astronomical body's) crust, often in the neighbourhood of volcanoes, which emits steam and gases such as carbon dioxide, sulphur dioxide, hydrochloric acid, and hydrogen sulphide) - www.wikipedia.org Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.6 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI It is important to note that the drillings at the Asal rift detailed in the Table indicate the presence of a 4 superficial reservoir with a good level of permeability and that this reservoir could be targeted for future geothermal developments. These could, in particular, be aimed at developing binary applications for this shallow resource (depth of 400 m to 600 m) as previous geophysical studies and drilling programs report. Figure C.2: Geothermal sites in Djibouti 4 The drillings indicated interesting conditions for development (salinity reduced to approximately 50 grams per litre (g/land temperatures between 130°C and 190°C). The flows are to be confirmed but they are reported to be high. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.7 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Table C.1: Site characteristics Surface Exploration geo-sciences complete Exploration manifestations drillings Geothermal Area Hot Fumaroles Geology Geo- Geo- No Years springs chemistry physics 1 Lac Assal ºº º ºº ºº ººº 6 75, 87, 88 2 Lake Abhe ººº ºº ºº 3 Hanle ºº ºº ºº ºº 2 87, 88 4 Gaggade º ººº ºº ºº ºº 5 Ghoubet º º ºº ºº ºº 6 Obock º º º º 7 Arta º ºº ºº ºº 8 Dorra º º º 9 Tadjoura º º º º 10 Allol-Sakalol ºº º º º Table C.2: Drilling record in Djibouti Drillings Year Depth Temperature Temperature Mass Salinity (bottom) Gradient flow (t/h) (g/l) (Celcius/100m) Asal Asal 1 1975 1,146 260 18 135 120 Asal2 1975 1,554 233 14.3 - - Asal 3 1987 1,316 264 15.5 350 130 Asal 4 1987 2,013 359 15.2 - 180 Asal 5 1988 2,105 359 15.2 - - Asal 6 1988 1,761 265 12.7 150 130 Hanle Hanle 1 1987 1,623 80 <5 - - Hanle 2 1988 2,038 120 <6 - - 3 gradient holes 1982 - 450 130 <3 - - 1984 Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.8 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI C.3 Explorations Explorations have been conducted at the Lac Assal, Hanlé, Gaggade, North Goubhet, Arta, Obock and Lake Abhe sites. We provide further details of the explorations carried out at these sites in the sub-sections below. The other sites listed by the Centre d’Etudes et de Recherches de Djibouti (CERD) (such as Sakkalol, Tamataco, Dora, Sud-Goubhet and Wead) have not been investigated (no existing studies are reported) and therefore no further details of these other sites have been provided. C.3.1 Lac Assal and Hanle/Gaggade area Figure C.3 below shows an aerial view Lac Assal. The first exploration campaigns in the Lac Assal 5 region were conducted in the early 1970’s. Following various scientific studies and some thermal gradient well drilling, two deep wells were drilled in 1975 (namely Asal 1 and Asal 2). Asal 1 produced excessively salty geothermal fluids (Salinity superior to 110 g/l) whilst Asal 2 was unproductive. Figure C.3: Aerial west view of Lac Assal After new geological, geo-chemical and geophysics studies, three sets of gradient drillings were executed in 1985 (by Géothermica) in the plain of Hanlé, situated approximately 20 km South West of Lac Assal. The results of these drillings showed a maximum temperature of 87°C for one of three drillings at 450 metres depth. On the basis of these initials findings in the Hanlé region, a program of deep drillings was authorised and the drillings Hanlé 1 (1623 m) and Hanlé 2 (2038 m) took place at the beginning of 1987. These drillings did not meet the expectations of scientists as the maximum temperatures recorded were below what was forecasted. The maximum temperature reached only 5 Mainly led by the Bureau de Recherches Géologiques et Minières (BRGM) - the French geological and mining Institute. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.9 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI 72°C at 1420 metres depth for Hanlé 1 and 124°C at 2020 metres depth for Hanlé 2. Permeability was limited to superficial mineral layers, which disappeared beyond a depth of 800 metres for Hanlé 1 and a depth of 450 metres for Hanlé 2. It is interesting to note however, that fumaroles existing in the plain of Hanlé could be bound to a magmatic activity at deep depths (although this was not apparent from the 2 drillings conducted at Hanlé) and that the fluids encountered at shallower depths during the drilling of Hanlé 1 and Hanlé 2 presented very low dissolved solid concentrations (TDS of 2 g/l). The superficial fluids (with low salinity) which were evidenced at Hanlé but also at Lac Assal did not give rise to any particular interest at the time of the drillings. This was due to the fact that the "binary" technology was not mature and well managed at the time. On the basis of the results at the Hanlé site, it was decided to pursue the regional exploration program in the zone of Lac Assal. As a result, the initial program which was started in 1970s was re- initiated with the belief that new exploitation techniques would be capable of handling highly saline fluids, as previously evidenced at Lac Assal. Four new drillings were undertaken during the years 1987 to 1989 at Lac Assal (namely Asal 3, Asal 4, Asal 5 and Asal 6). Two of these drillings produced flows as high as 350 tonnes per hour (t/h) (Asal 3) and 200 t/h (Asal 6). Continuous discharge testing led to a significant reduction in the discharge flows (Asal 3 flow decreasing to 140 t/h after 2 periods of discharge at the end of 1987 and in 1989). The salinity of the produced fluids (of the order of 130 gm/l) was superior to that of Asal 1 and confirmed that the chemistry of the fluids would be an essential parameter to take into account if an industrial application of the deep geothermal fluids at Lac Assal were to be taken advantage of. Figure C.4 below shows the location of the wells Asal 1 to Asal 6. Asal 4 and Asal 5, which were drilled towards the interior of the rift zone and not on the edge of the rift and proved to be unproductive at the time, although minimum temperatures of 345°C and 360 °C were recorded for these two wells respectively. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.10 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Figure C.4: Location of wells A1 to A6 in Asal rift zone Lac Assal Rift Indian Ocean B.3.2 North Goubhet North Goubhet is situated South East of the Asal rift valley, at an elevation of between 500 and 600 metres and is in the continuity of the Asal rift faulted zone on the active geological structures heading South East to North West with secondary orientations from South to North in the region of Makarassou. Geo-chemical sampling and the subsequent analysis made on the existing fumaroles at North Goubhet led to reservoir temperature estimates of between 170°C and 220°C (Geothermica, 1987). 6 Three geophysical studies were also carried out on this zone by BRGM of France in 1983 including gravimetric methods, basic rectangle electrical methods and AMT (audio magneto-telluric). Given that these studies were carried out more than 25 years ago, it is not advisable to start a field exploratory program based on these preliminary studies and additional geo-scientific work would be necessary. Some existing interpretations concerning the superficial layers however, may be relevant and of interest for further studies. 6 BRGM is France's leading public institution involved in the Earth Science field for the sustainable management of natural resources and surface and subsurface risks. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.11 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI C.3.2 Arta The Arta site is the closest to Djibouti-Ville (located approximately 40 km west of the capital town). Geological studies were performed at the beginning of 1980’s (Arthaud, Geothermica) followed by geo-chemical sampling and analyses (Geothermica, 1982). All of these studies failed however, to define the exact nature and the origin of the collected fluids. Two geophysical studies were also undertaken on this zone (using gravimetric and electric methods). These studies revealed the existence of a thick basalt dominated zone. This zone helps to explain the absence of a geothermal reservoir in the geo-chemical analysis. Geothermica suggested that the Arta site contained a reservoir of a depth exceeding 2 km and that this site is not a superficial resource. C.3.3 Obock Hot springs and one fumarole can be seen at low tide at the Obock geothermal site, located on the North East shore of Djibouti. Electric studies were carried out on this site as early as 1965 (by CGG) and were followed later by incomplete magnetic measurements carried out by Essrich and Brunel in 1990. Geo-chemical studies revealed a main constituent of oceanic origin in the collected fluids and reservoir temperatures estimated with geo-thermometers exceed 200°C (Houssein, 1993). No geological study or large-scale geophysical survey is available on this zone. CERD noted in a recent publication however, that an existing geophysical and marine tectonics study (carried out by Manighetti, 1993) suggested that Obock may be in the continuity of the Tadjoura Gulf faulting zone. Further investigations (geology, geophysics) are therefore necessary in this area to confirm such a hypothesis. C.3.4 Lake Abhe Numerous surface manifestations exist in the Lake Abhe geological zone, namely hot springs (reaching 90°C) as well as fumaroles on a flat land whose surface exceeds 100 km². No geo- scientific study is reported to exist in this area. It is understood that a company from India is ready to start a scientific exploration program in the Lake Abhe area and that the local authorities of Djibouti were in the process of granting rights to this company (March 2009). Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.12 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI C.3.5 Scaling and corrosion study on Asal deep fluids (Virkir-Orkint, 1990) The explorations and investigations carried out in the 1970’s and 1980’s seem to indicate that the geothermal fluids in Djibouti have a strong salinity. In an attempt to find potential industrial applications for highly saline geothermal fluids, an extensive study addressing the phenomena of geo- chemical deposits and corrosion induced by the discharged fluids was contracted by EdD to the Icelandic company Virkir-Orkint. A report on this topic was published in August 1990. The objective of the Virkir-Orkint study was to determine the physical and chemical characteristics of the scaling noticed during the extraction of the fluids, to test inhibitors capable of avoiding, delaying or strongly reducing these phenomena and finally to evaluate the reservoir for the use of these products. This study included approximately 30 weeks of site tests during which the Asal 3 well discharged continuously for more than 12 weeks at an average flow of 31.6 kg per second (kg/s) and an uncorrected average enthalpy of 1070 kJ/kg, with a salinity of the order of 12 per cent. The characteristics of the extracted fluid were saturated in a diverse array of minerals (silica, iron carbonates, metallic sulphites) with a pH ranging between 4 and 5. The major findings made during these discharge tests were that the deposit rates at the well head increased drastically from 1 to 2 microns per hour if discharge pressure was kept above 18 bar g and that deposition increased 6 fold for pressures lower than 16 bar g. The solid deposits evidenced when lowering the discharge pressure were lead sulphites, followed by iron silicates in the immediate pressures (12 to 16 bar g) and finally amorphous silica at even lower pressures. Two inhibitors developed by the Italian company NADAR were tested in-situ, namely Nadar 4093 and Nadar 1008. The first inhibitor was a trapping agent for heavy metals and appears under an aqueous shape of aliphatic polymers. The second inhibitor is a trapping and dispersal agent more specific for calcium, magnesium and Silica. These two inhibitors were tested but not adapted to the specificities of the Asal site. They showed themselves effective against the first quoted scaling products (lead sulphites) in the pressures above 18 bar g but also induced side effects such as iron silicates and calcite deposits which were not evidenced without the use of these inhibitors. The study concluded that these standard products should be tested in a laboratory to precisely define the components and dosing particularly adapted to this site and to this resource. Virkir-Orkint also recommended that such an inhibitor be injected continuously at the bottom of the well in order to achieve a higher efficiency. It is important to note that during these site tests the flow rate of the Asal 3 well dropped by more than 25 per cent in 3 months of testing and that the head pressure reduced by 6.5 bar during the same period. Similarly, the same drop was evidenced for the Asal 6 well (situated only 300 m from the Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.13 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Asal 3 well) with a pressure decline of 4.5 bar. Virkir-Orkint concluded in their report that on the basis of these discharge test results, the reservoir feeding these wells is limited in size and in capacity. Finally the report also addressed the issue of mechanical cleaning in order to remove scaling and compared the costs of cleaning out a well of a standard diameter with those cost associated with cleaning out a well with a widened diameter (assuming an interval of 6 months for the standard diameter compared with 30 months for the larger diameter). The study concluded that the well with the larger diameter was 20 per cent more economical, when considering the initial costs and the subsequent maintenance costs. All in all, the field study was an expensive exercise and did not lead to convincing or definitive results. As a result, very little came of the report in the immediate years to follow its completion. C.3.6 Recent investigations using more advanced techniques The African Rift Geothermal Facility (ARGeo) program (aimed at developing geothermal resources in 7 the East African regions) in line with the Icelandic Government and the Icelandic companies REI and ISOR re-initiated the geothermal development program with the aim to finally validate an exploitable geothermal resource. The geo-scientific methods used in the 1970’s and 1980’s which preceded the drillings at Hanlé 1 and Hanlé 2 and those of Asal 1 to Asal 6 were electric, magneto-telluric (AMT) or electromagnetic. These methods have penetrative limitations (with distortions beyond a depth of 500 metres experienced and ineffectiveness beyond depths of 1000 metres). In recent years however, new techniques have been implemented successfully on numerous geothermal sites, in particular island volcanic zones or oceanic rift valleys. These new techniques enable a more complete and accurate geo-scientific investigation to be carried out, at greater depths. The Icelandic company ISOR was contracted to execute a MT (magneto-telluric) study on the Asal rift zone, covering the entire area between the Ocean and Lac Assal, and also to extend the resistivity Transient Electro-Magnetic (TEM) study done in 1988 by the Icelandic National Energy Authority (NEA), to validate the presumed existence of a superficial reservoir and to confirm outlines and resistivity characteristics of a reservoir of higher enthalpy situated more in depth. This study was aimed at a preliminary critical analysis of the existing scientific data, followed by a field trip to complete the data (in particular geology and geophysics). A reinterpretation of all the data, including those collected at the end of 2007 during the field trip then lead to the scientifically 7 A concession was granted by the Government of Djibouti in 2007 for the exploration of Asal rift area. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.14 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI substantiated proposal for a reservoir model. This model allowed for the definition of drilling targets for a future exploration program. This model identified the internal zone of the Assal rift in the South East (Fiale caldera and Lava Lake as the key areas (see Figure C.5 below). Figure C.5: Key areas in Lac Assal Fiale Lava Lake Korili/Gale le koma Forages existants For reasons of confidentiality called for by the Djibouti authorities, the detailed results and conclusions 8 of the ISOR study were not disclosed to Parsons Brinckerhoff (PB) in writing . Nevertheless, the findings of the ISOR study have been disclosed to PB and are replicated below: • The existence of a "hydrological barrier" close to the surface was confirmed. This is identified by the higher resistivity levels to the North of this barrier (directed West / North East and running North of the existing drillings) than in the South where lower resistivity levels are observed. This barrier results from a sharp fall in the groundwater level, probably resulting from mineral precipitation of ocean salty sea 8 A power point presentation and a local consultation in the CERD offices were organized during the PB visit to Djibouti (March 2009). Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.15 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI water circulating in the shallow layers from the Ocean in the south towards Lac Assal in the North. Considering the numerous geologic fractures present in the rift valley (directed for the greater part South East / North West but interrupted by directed fractures oriented East / North East), ISOR concludes that it is likely that there are several hydro- geological systems in the rift zone. These "systems" are most probably isolated and not connected and show most probably different thermal and geo-chemical characteristics. • The study confirms that the peripheral zone situated South West of the rift and which was the object of investigations and most of the previous drillings corresponds to an isolated zone not connected to a vaster system. This is also confirmed by the results of well A3 discharge tests. This “isolation� explains the strong salt concentration of previously extracted fluids (concentration by evaporation). • The presence of a combination of good conditions (geology, seismicity, low resistivity levels, surface manifestations such as hot springs and fumaroles, concentrated active seismicity in the same area) indicate that the zone located South East of the rift (Fiale, Lava lake) would be suitable for further field investigations. The presence of a ductile dome under which low resistivity levels allied to important seismic activities (measured continuously by the French seismic observatory) could probably result from deep magmatic activity (heat mining). This hydro-geological system seems also more "opened" than that situated in the South West, in particular by its relativity to the ocean and the presence of the ground water barrier which facilitates the reheating of in-depth sea water through the active faults. Lower levels of salinity are anticipated because of this wider hydrological system which has less stagnant fluids. Furthermore, the superficial aquifer is likely to act as a screen for raising geothermal vapour from great depth and would explain, according to ISOR, the very few fumaroles present at the surface in the zone of Fiale / Lava Lake which is due to dilution in deeper depths (we note that temperatures of hot springs range between 75°C and 91°C in the rift area). • A second promising zone is situated to the North West of the rift (Korili /Gale Kôma - see Figure B.5) which, in view of ISOR study conclusions, also presents favourable characteristics (surface manifestations, hot springs up to 80°C, interesting resistivity Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.16 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI levels) evidenced by the MT study and confirmed by geology and hydrogeology interpretations. • Based on all the data collected during the MT survey, ISOR also demonstrated “a posteriori�, with vertical sections in the plan of existing drillings Asal A1 to Asal A6; why some of them were unproductive or why they did not reach a wider reservoir. The large number of MT stations used for this study has allowed ISOR to be confident on the quality of the results presented in their report. • The MT survey confirms the existence of the superficial aquifer situated at 400 to 600 metres depth, and ISOR states that the increasing resistivity levels right under this aquifer can also result from mineral precipitation reducing the porosity of the rocks. • The proposed exploration drilling program resulting from ISOR study was not disclosed however, we understand from the above conclusions that future drillings will be diverted to target geologic faults in the zone of Fiale /Lava lake in priority, and possible also in a second stager in the North West portion of the rift. Such targets are consistent with the geothermal field model presented in ISOR report. In conclusion, PB notes that significant findings and progress were achieved, due in particular to the Icelandic experience and know-how in geophysics. Only the results of the forthcoming exploratory drillings however, will allow for the validation of the hypotheses raised by ISOR and confirm the real and tangible existence of a resource presenting appropriate conditions. The salinity of the geothermal fluids extracted by these future drillings will also determine and impact greatly upon the technical solutions necessary for the valuation of these fluids. It should be noted that during our review of the exploration activity in Djibouti that in the year 2000 the American Geothermal Development Associates (GDA) company based in Reno (Nevada, USA) with funds raised by the US Trade and Development Agency, undertook a feasibility study to develop a 30 MWe geothermal power plant project at Lac Assal. This study did not lead to any further development either. C.4 Ranking of most promising sites for future development On the basis of the data made available (see Section C.1), we have placed each prospective site in order of preference for development using the following criteria: Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.17 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI • Existence of deep wells showing high temperatures (> 200°C) and good productivity and/or medium range geothermal resource temperatures for possible electricity generation via binary units. • Indication of high resource temperatures from geo-chemical analysis of hot springs or fumaroles with a good connection to the deep reservoir. • A comprehensive geological/hydrological survey that assesses the likelihood of a suitable heat source in the area that shows regional and local faults and geological structures that could be part of a geothermal system. Based on the above, there are several geothermal development opportunities envisaged in Djibouti. The following factors have been considered in our ranking of the geothermal projects: • The present development stage of the most promising prospects, • The additional exploration phase(s) needed to evidence and quantify the resource in terms of thermal power potential and geo-chemical characteristics of the geothermal fluids, • The implementation phase needed subsequently to drill production wells and engineer and construct the steam-field, • Power plant and transmission lines to connect the geothermal plant to the grid. C.4.1 Lac Assal - deep reservoir Considering the recent scientific program and the conclusions of 2007/08 ISOR study, this site is a priority for further development. It must however be large enough in size to justify an interconnection with the existing electricity network, knowing that a high voltage line of about 50 km distance must be built at the same time as a geothermal power plant. Before these constructions, the exploratory drilling program proposed by ISOR must be financed, scheduled and executed and its results will have to meet the minimal conditions required for an industrial development. This will apply first and foremost to the geothermal fluids extracted and tested after drillings, which will have to present sufficient physical-chemical characteristics to be exploited. The geothermal reservoir will have to exhibit sufficient capacity to provide a continuous supply of energy for electricity production. Specific studies on the resource and the reservoir must be financed and executed immediately after the completion of the exploratory drillings. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.18 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Preliminary indications concerning the economy of a 30 MW project development at Lac Assal are given in Section C.7. PB must however clarify that at this stage of development; estimations of power can be only speculative, knowing that only the drillings can determine the capacity of a well, its characteristics and behaviour in continuous discharge. The reservoir studies mentioned above will define the long term capability of the resource. C.4.2 Lac Assal - shallow reservoir (400 - 600 metres) Following the identification of a superficial reservoir at Lac Assal, it is advisable that this resource is validated and its characteristics are determined. The technologies available on the market to exploit a resource of medium temperature (below 180°C) are well managed and cost effective. Such a development is not advisable as a stand alone plant (at the Lac Assal location). This is because it is not likely to justify the significant investment required in a high voltage line to connect to EdD network. A binary power station however, of one or more generating units could be developed as ‘add-ons’ to the greater geothermal development on the Assal rift zone (see Section C.4.1 above). This option could also be developed in parallel to the main plant, if the exploratory drilling program at Lac Assal included one or more drillings at a shallow depth in the most favourable zones. The specific economic data for this kind of binary plant are also provided in section C.7. C.4.3 Arta This site is interesting due to its proximity to the existing electrical grid and the capital Djibouti-Ville. The studies available for this site are, however, rather old and incomplete and do not consider a major deep exploratory program (more than 2000 metres depth announced by Geothermica) without implementing first geophysics investigations of the MT type. The financial risk related to deep “blind� drillings is too significant in comparison with up front “limited� costs for a geophysical survey using the modern means of investigations and interpretation. The program necessary to develop this site is very similar to the one in progress at Lac Assal, namely a geophysics/geology/geochemistry campaign leading to a program of exploratory drillings in order to validate a resource for an industrial exploitation. C.4.4 Obock This site is of real interest to the development of local resources because Obock’s distance from the inter-connected network. Furthermore, the only means of production installed to date at Obock are diesel engines. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.19 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Taking into account the ease of accessibility to the site from Obock town and limited local power needs, a geothermal development program in Obock could be envisaged in the future. The initial steps required to develop a geothermal resource at Obock would be the execution of geo-chemical analysis using modern means of analysis, followed by a synthesis of the geological data and geophysics available in order to define which will be the next stage. This site can also be of considered for use in medical, tourism and leisure industries (balneotherapy, hotel resorts). These secondary uses are not the subject of this study and are not studied further. C.4.5 Gaggadé The plain of Gaggadé, adjacent to Hanlé, was subject to common investigations at the time of the preliminary drillings Hanlé 1 and 2 in 1987. These drillings proved to be disappointing with regard to the temperatures recorded at the depths drilled. It is not unreasonable however, to suggest that with new methods of geophysical investigations, combined with directional techniques of drillings not used at Hanlé in the past; this site could still be of interest. We do note however, that the Gaggadé site is located further away from the network than Lac Assal and is consequently not a priority from this point of view. An isolated local exploitation from a superficial aquifer could however be considered, provided that a preliminary geo-scientific program confirms the existence of such an aquifer. C.4.6 Lak Abhe Although no pertinent geo-scientific data is available, this zone provides an interesting possibility for a small scale geothermal development if the investigations in progress prove to be positive. Taking into account the absence of existing studies on this zone, it is reasonable to assume that any development at this location would not take place for another 5 to 10 years into the future. C.5 Review of possible conversion technologies Having reviewed the geothermal potential in Djibouti in the previous sub-section, we present below an overview of the plant types that could be developed. C.5.1 Flash plants Flash geothermal plants are based upon conventional steam turbines, especially the low pressure section installed in fossil fuelled thermal power stations. However special provisions are needed to deal with the effects of corrosion, scale formation and the wetness in the steam. As the steam contains some non-condensable gases, a gas extraction plant is also required. The gas is largely Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.20 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI carbon dioxide but may also include hydrogen sulphide (H2S). A typical single flash process with a separator (flash tank) installed between the wells and the turbine is shown in Figure C.6 below. Figure C.6: A single flash process Typical unit sizes of conventional geothermal turbines are 55 - 60 MW for central station turbine generators and up to 35 MW for modular units. Generally speaking, at a rating of below 25 MW, a single flow unit will be adopted whilst for a rating of between 30 - 75 MW a single or double flow unit may be used. Currently, the largest geothermal turbines in single cylinder, double-flow configuration are 110 MW at 50 Hz and 77.5 MW at 60 Hz. Resource utilisation efficiency (MWe generated versus steam enthalpy available from the well) is approximately 16 per cent with availabilities in excess of 90 per cent. A second separator may be installed on the brine line in order to separate steam and brine at a lower pressure (double flash plant). Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.21 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Technology for highly mineralised geothermal fluids The salinity levels in the fluids extracted from Lac Assal are been recorded at levels of around 100 mg/l and even up to 150 mg/l in some cases. At such levels of salinity, specific technologies may be necessary to treat these “charged� fluids and ensure a continuous exploitation without unsolvable deposits and scaling problems resulting from precipitation of dissolve solids during the geothermal fluid extraction process. A schematic of a crystalliser reactor clarifier (CRC) plant is shown in Figure C.7. Figure C.7: Crystalliser reactor clarifier (CRC) plant schematic It can be seen from this schematic that the crystalliser tanks (2 in this case) act as supplementary flash tanks and extract additional steam which is fed to the steam turbine. This type of plant is more complex than a standard flash plant, in particular with regard to the balance of plant equipment (crystallisers, clarifiers, filters etc). As a result of this, extra investment and additional maintenance costs are likely to be incurred over and above the flash plant. These additional costs will undoubtedly impact on the overall profitability of the power plant when compared with the same capacity flash plant. For Djibouti it is therefore essential (and as proposed by ISOR after the 2007 geo-scientific studies) to target geothermal fluids with lower concentrations of dissolved solids to avoid or mitigate these extra investment costs. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.22 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI C.5.2 Binary plants Organic Rankine Cycle TECHNOLOGY DESCRIPTION In the Organic Rankine Cycle (ORC) turbine the organic working fluid is vaporised through the application of a heat source in the evaporator. The organic fluid vapour expands in the turbine and is then condensed using a flow of water in a shell-and-tube heat exchanger (alternatively, ambient air can be used for cooling). The condensate is pumped back to the evaporator thus closing the thermodynamic cycle. Heating and cooling sources are not directly in contact with the working fluid or with the turbine. The schematic of an ORC binary simple cycle and combined cycle plant are presented in Figure C.8 below. Figure C.8: ORC binary cycle principle (left) & binary combined cycle plant (right) The geothermal combined cycle provides an important and possibly essential benefit by being conducive to the use of air-cooling, which in turn enables the entire steam condensate mass to be re- injected. This 100 per cent fluid re-injection (except for NCG) preserves the reservoir mass balance, thereby sustaining the reservoir life and providing pressure support. In most water-cooled plants the steam condensate itself is used as the make-up water for cooling towers, and 70 – 80 per cent of the condensate will typically be evaporated through the cooling towers into the atmosphere and will be lost. This issue is particularly relevant for steam-dominated geothermal reservoirs, which are more susceptible to pressure decline and reduction in reservoir capacity over time For a high enthalpy water-dominated resource, the overall efficiency of the power plant can be enhanced by integrating a standard ORC unit on the separated brine line. This is known as the bottoming cycle unit. Such a unit can be installed on existing brine lines with sufficient enthalpy to produce additional electricity. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.23 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI In Binary Combined Cycle plants, we find two major sections. The first is called a Geothermal Combined Cycle Unit (GCCU) and this utilises a conventional geothermal steam turbine to expand the separated steam as with a conventional plant. However, instead of the exhaust steam being condensed in a direct contact condenser, the turbine exhaust steam is directed to a heat exchanger where the geothermal steam heat recovery provides both latent and sensible heat to vaporise an organic fluid in an ORC unit. In addition to the GCCU, which utilises the separated geothermal steam, use is also made of the energy in the separated brine to provide a further increment of power by using an ORC turbine as explained previously. FACTORS LIMITING APPLICATION ORC turbines are limited to approximately 5 MW to 10 MW in a special configuration due to reliance on liquid organic gasses such as iso-butane or pentane. Ormat manufactures standard ORC units of less than 10 MW, and achieves greater plant outputs combining standard units. In the binary combined cycle version, the conventional steam turbine used can be of bigger size, up to 50 MW or more. However, Ormat standardized ORC unit size allow modularity in design and construction as well as easy transportation and installation at site. A picture of an ORC binary unit is shown in Figure C.9 below. Figure C.9: 5MWe ORC binary unit on skid Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.24 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Kalina cycle TECHNOLOGY DESCRIPTION The Kalina cycle differs from the already commercial Organic ORC technology in that an ammonia/water mixture is used as the secondary working fluid instead of a pure organic fluid such as iso-pentane. Both of these technologies are examples of binary cycles, where the heat is transferred to a secondary fluid from which power is extracted. With the Kalina cycle, the ammonia water mixture boils at a variable temperature unlike pure substances which boil at constant temperature. Variable temperature boiling permits the working fluid to better match the temperature gradient of the heat source fluid as it cools in a heat exchanger. Also, by using a different mixture composition in the condensing part of the cycle, condensation can be achieved at slightly above atmospheric pressure, avoiding the need for vacuum conditions and costly gas extraction equipment. A schematic of the Kalina cycle is presented in Figure C.10 below. Figure C.10: Typical Kalina cycle schematic A significant aspect of the Kalina design is that the molecular weight of NH3 is similar to water (H2O) (17 and 18 respectively). Therefore standard steam turbines can be used, and the performance is not affected by changes in mixture ratio (i.e. no special design equipment is required). Turbine, generator, condenser and cooling loop are similar to other thermal power plant applications. Conventional 50 MW or 100 MW steam turbines can be used with ammonia-water. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.25 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI PERFORMANCE With low temperature heat sources, the efficiency of utilisation can vary widely depending on the supply temperature and constraints such as the allowable temperature the source fluid can be cooled to. A typical constraint is the potential for supersaturated silica in geothermal brine to precipitate out and scale heat exchangers if it is cooled too much. Kalina identified performance advantages compared to ORC are: • A 10 to 40 per cent improvement in overall cycle efficiency over existing ORC technology (this figure is still to be confirmed in long run industrial applications). • Initial scale of commercial Kalina cycle plant 5 to 25 MW, but potential to scale up once proven. UTC technology (Small scale binary applications) In 2000, United Technologies Company (UTC) Power engineered a new power system based on ORC technology, known as PureCycle®. To understand the power conversion process, UTC suggests thinking of an air conditioner that uses electricity to generate cooling. The system reverses this process and uses heat to produce electricity. The system is driven by a simple evaporation process and is entirely enclosed (like all binary units), which means it produces virtually zero emissions. After the heat is extracted for power, the water is returned to the earth for reheating. UTC Power’s system, in cold climates, can operate on 75ºC geothermal water and by changing the refrigerant type can use hydrothermal resources up to 150ºC. Previously, experts had assumed that geothermal fluids needed to be in the range of 107ºC to produce economically viable power generation. An example of such technology application is Chena in Alaska, where two units are currently in operation at Hot Springs resort. These are the first units to operate on geothermal energy. A 250 KWe gross (200 KWe net) unit, isolated from the electric grid, uses a standard UTC chiller equipment with just a few modifications. The design incorporates a centrifugal compressor, which runs in reverse as a radial inflow turbine, and heat exchangers originally designed for large chiller applications, to produce electric power. This makes geothermal power generation highly competitive with existing diesel generation in rural Alaska, as fuel costs are virtually eliminated. Maintenance costs for the power plant are reported to be very low, at about 1 USc per kWh. As the air conditioners are standard equipment from the air conditioning industry, the capital costs of such units are known to be lower than the equivalent small ORC units developed by others. A picture of a UTC converter is shown in Figure C.11 below. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.26 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Figure C.11: Picture of a UTC converter C.6 Typical project development steps and associated work scope Geothermal projects are characterised by high risks at the beginning of the development programme (i.e. during exploration phase). This is particularly true in the case of a development in an unknown zone where no or few geo-scientific data or reports are available and where basic investigations must be carried out initially. Figure C.12 below presents the relationship between expenditure and risk during the life cycle of a geothermal development. The curve highlights that risks related to a geothermal resource decrease as exploration progresses and the characteristics of the resource are determined, first by means of scientific investigation methods and then with exploration drillings. The project development phase can be decided only when sufficient tangible and proven elements enable the explorer to define the long term potential of the resource for a profitable investment, along with calculated acceptable risks for the developer and his project financing partners. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.27 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Figure C.12: Expenditure and risk prior to geothermal development Risk Geothermal projects go through several main phases in time. These phases are: 1. Exploration 2. Development 3. Design and Construction 4. Operation and Maintenance Typically, decisions to proceed are made between the exploration and development phases (interpretation of geo-scientific surveys, geochemistry, thermal gradient, legal, environmental or financial issues) and between development and construction phases (resource characteristics, sustainability, project economics, financing issues, etc). Figure C.13 below presents the various stages of a project and gives approximate durations as well as indicative budgets for each one of them. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.28 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Figure C.13: The stages of a geothermal development Table C.3 below gives a short description of the content of each development phase. It should be noted that in the case of a development of an existing field (additional drillings and new generation unit(s) on a zone already in exploitation), some stages can be reduced and the project execution can be faster. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.29 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Table C.3: Detailed project life cycle Exploration Phase Geology, Structure and alteration mapping Geochemistry sampling of key geothermal manifestations Geophysical surveys (MT- TDEM) and interpretation Shallow well drilling and data logging Thermal gradient and soils permeability assessments Data integration and definition of exploration strategy and drilling targets for the development phase Pre-Feasibility Pre-feasibility (conceptual) studies for the steam field (SAGS), power plant and transmission. Comparison of various conversion technologies Evaluation of legal and environmental issues related to geothermal development Development Phase Drilling of first full size discovery well(s) Confirmation of the resource characteristics, size, boundaries Well Tests and reserve estimates. Determine mass flow, steam flow and enthalpy of the resource. Define project design parameters. Preliminary design of the plant, including size, location of wells, separators, power station and energy transmission (own use or export or both), economic appraisal and identification of possible suppliers and their Feasibility timelines Financial modelling of various development alternatives (hybrid systems, direct uses) and decision to proceed in detailed design and construction phase of 1 alternative Solve legal, environmental and financing issues to clear way for construction phase. Detailed Design and Construction Execution of production drillings and re-injection drillings for the specified development. Detailed design of civil works as well as steam field, power plant and Project implementation transmission equipment. Prepare specifications. Outsource plant equipment, control manufacturing, transport erection and commissioning of all systems. Operation & Maintenance Wells work-over and acidising Wells, Steam field and Power Routine and planned maintenance of all equipment Plant Safe Operation and Monitoring of resource and power station characteristics and adjustment of preventive maintenance operating parameters Optimisation of resources and O&M costs Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.30 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI C.7 Possible development alternatives at Lac Assal and related economics C.7.1 30 MW flash plant - deep drillings without transmission Capex The assumed capital cost breakdown for a 30 MW flash plant is shown in Table C.4 below. These costs are provided for reference only, whereby the following assumptions have been made: • Drilling depth 2,500 metres and well output of 3.35 MW per well. • Site preparation costs and short distance electrical transmission costs (for a Greenfield development) of 10 per cent of project direct costs. • Developer indirect costs of 14 per cent of all direct costs including site preparation and nearby transmission costs. • Electrical transmission is limited to a step transformer and connection to a nearby switchyard. As stated earlier in this report, only a complete and successful exploration drilling programme at Lac Assal will give more precise information on the project size and related economics. Table C.4: Capital cost estimate for a 30 MW flash plant Cost Breakdown Wells (US$ millions) 41.2 Steam field (US$ millions) 5.0 Power plant (US$ millions) 45.0 Site preparation & transmission (US$ millions) 8.6 Indirect costs (US$ millions) 13.2 Transmission - Total Investment (US$ millions) 113.0 Investment (US$ / kW) 3,767 O&M costs Typical O&M costs to be considered for preliminary economic analysis for this size of geothermal plant are given in Table C.5 below. It should be noted however, that these figures are not site specific at this stage and that the figures presented are generic and for preliminary analysis purposes only. This is especially true with regard to the well costs, which are resource and site specific and can only be determined at a later stage of the project implementation when the well(s) output decline is known or estimated. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.31 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Table C.5: Annual O&M costs for a 30 MW flash plant US$ per annum Steam Field Staff costs 448,000 Variable costs 160,000 Power Station Staff costs 510,400 Variable costs 256,000 Contract overhauls 153,600 Off-site costs 224,000 Insurance 582,400 Make-up well cost 700,000 Transmission Total Cost 3,034,400 The annual O&M costs presented above can be converted into a cost (US cents per kWh) charge and these costs are presented in Table C.6 below. The costs assume an annual generation at base load with a capacity factor of 90 per cent. Table C.6: Operating cost (per kWh) for a 30 MW flash plant O&M USc/KWh Steamfield 0.26 PowerPlant 0.39 Off site &Insurance 0.34 Make up wells 0.30 Transmission TOTAL 1.28 Project implementation timeframe By taking into account the project development phases detailed in Section C.6 of this report and the present development status at Lac Assal, it is possible to estimate the possible project development timeframe and capital expenditure disbursement. It is believed that the flash plant could begin full (30 MW) commercial operation in year 2016 (earliest) with a spread of capital expenditure (CAPEX) expenditures over the 4 years prior to its commercial operation (i.e. 2012 to 2015). The capital disbursement for this project is presented below in Table C.7. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.32 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Table C.7: Capital cost disbursement for a 30 MW flash plant Year n-4 n-3 n-2 n-1 n Plant Capex Disbursement (%) 0.7 % 1.8 % 47.6 % 50 % In the event of a successful development of stage 1 and promising resource assessments showing room for a bigger development, a second unit of 30 MW could begin commercial operation in year 2021 (earliest). The second unit would have the same capital cost spread as presented in Table C.7. C.7.2 Flash plant deep drilling and 63 kV transmission line Capex The assumed capital cost breakdown for a 30 MW flash plant (with associated transmission) is shown in Table C.8 below. These costs are provided for reference only, whereby the following assumptions have been made: • Drilling depth 2,500 m and well output of 3.35 MW per well. • Site preparation costs and short distance electrical transmission costs (for a Greenfield development) of 10 per cent of project direct costs. • Developer indirect costs of 14 per cent of all direct costs including site preparation and nearby transmission costs. • Assumed that the geothermal plant would be connected into the 63 kV circuits from Pk12 to Ali Sabieh at PK51. Therefore transmission is assumed as a 50 km 63 kV double circuit line with associated switchgear. As stated earlier, only a complete and successful exploration drilling programme at Lac Assal will give more precise information on the project size and related economics. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.33 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Table C.8: Capital cost estimate for a 30 MW flash plant with transmission line Cost Breakdown Wells (US$ millions) 41.20 Steam field (US$ millions) 5.00 Power plant (US$ millions) 45.00 Site preparation & transmission (US$ millions) 8.60 Indirect costs (US$ millions) 13.20 Transmission 9.00 Total Investment (US$ millions) 122.00 Investment (US$ / kW) 4,066.67 It should be noted the cost per KW installed has increased by 15 per cent compared to the case without a transmission line. O&M costs Typical O&M costs to be considered for preliminary economic analysis for this size of geothermal plant are given in Table C.9 below. This table differs from Table C.5 in only that the O&M costs of a transmission line are also included for (we assume 2 per cent of the transmission capital cost). Again, it should be noted however, that these figures are not site specific at this stage and that the figures presented are generic and for preliminary analysis purposes only. This is especially true with regard to the well costs, which are resource and site specific and can only be determined at a later stage of the project implementation when the well(s) output decline is known or estimated. Table C.9: Annual O&M costs for a 30 MW flash plant with transmission line US$ per annum Steam Field Staff costs 448,000 Variable costs 160,000 Power Station Staff costs 510,400 Variable costs 256,000 Contract overhauls 153,600 Off-site costs 224,000 Insurance 582,400 Make-up well cost 700,000 Transmission 180,000 Total Cost 3,214,400 The annual O&M costs presented above can be converted into a cost (US cents per kWh) charge and these costs are presented in Table C.10 below. The costs assume an annual generation at base load with a capacity factor of 90 per cent. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.34 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Table C.10: Operating cost (per kWh) for a 30 MW flash plant with transmission line O&M USc/KWh Steamfield 0.26 PowerPlant 0.39 Off site &Insurance 0.34 Make up wells 0.30 Transmission 0.08 TOTAL 1.36 Project implementation timeframe By taking into account the project development phases detailed in Section C.6 of this report and the present development status at Lac Assal, it is possible to estimate the possible project development timeframe and capital expenditure disbursement. It is believed that the flash plant could begin full (30 MW) commercial operation in year 2016 (earliest) with a spread of Capex expenditures over the 4 years prior to its commercial operation (i.e. 2012 to 2015). It is believed that capital disbursement and timeframe for development would be the same as for case 1 (assuming land clearances, permit acquisition and the subsequent engineering, manufacturing and construction of the Transmission line were carried out in parallel to the power plant development). The capital disbursement for this project is presented below in Table C.11. Table C.11: Capital cost disbursement for a 30 MW flash plant with transmission line Year n-4 n-3 n-2 n-1 n Plant Capex Disbursement (%) 0.7 % 1.8 % 47.6 % 50 % Line Capex Disbursement (%) 5% 50 % 45 % In the event of a successful development of stage 1 and promising resource assessments showing room for a bigger development, a second unit of 30 MW could begin commercial operation in year 2021 (earliest). This second unit would have the same capital cost spread as presented in Table C.7 above. C.7.3 6 MW Binary plant shallow drilling Existing drillings at Lac Assal show that there is a superficial reservoir with good permeability and flow rates with temperatures between 140ºC to 180ºC at depths between 400 metres and 600 metres. The Lac Assal site has significant potential for the development of binary units; however, due to the limitation of smaller sizes, such developments must be seen as being in addition to the main flash Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.35 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI plant development. In order for these smaller binary units to become economical viable, the transmission line must already exist. The economic details of a generic Binary unit are given below in Table C.12 to Table C.15. Table C.12: Capital cost estimate for a 6 MW binary plant Cost Breakdown Wells (US$ millions) 6.00 Steam field (US$ millions) 1.50 Power plant (US$ millions) 13.00 Site preparation & transmission (US$ millions) 2.00 Indirect costs (US$ millions) 3.50 Transmission - Total Investment (US$ millions) 26.00 Investment (US$ / kW) 4,333 Table C.13: Annual O&M costs for a 6 MW binary plant US$ per annum Steam Field Staff costs 45,000 Variable costs 25,000 Power Station Staff costs 65,000 Variable costs 50,000 Contract overhauls 100,000 Off-site costs 40,000 Insurance 45,000 Make-up well cost - Transmission - Total Cost 370,000 Table C.14: Operating cost (per kWh) for a 6 MW binary plant O&M USc/KWh Steamfield 0.13 PowerPlant 0.38 Off site &Insurance 0.17 Make up wells 0.00 TOTAL 0.68 If binary units are to be built in parallel with the main plant, geo-scientific investigations and exploration drilling program must start at the same time as the main development. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.36 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI However, we recommend to plan this development with some time stagger (for example, starting exploration after the exploration phase of the main plant), and for the plant to be handled as an “add on� project. On the basis of this recommendation, the earliest commercial operation of a 6 MW binary unit could therefore be 2018, followed by 1 or 2 additional units in 2022 and 2026. The assumed capital expenditure programme is detailed in Table C.15 below. Table C.15: Capital cost disbursement for a 6 MW binary unit Year n-4 n-3 n-2 n-1 n Plant Capex Disbursement (%) 4.8 % 45.2 % 50 % C.8 Other possible Geothermal developments in Djibouti C.8.1 Lake Abhe No specific details are available to assess the potential of this location for geothermal power plant development. A company from India however, is due to perform a geo-scientific exploration program, 9 starting in 2009 . Drillings at Lake Abhe may be deeper than at Lac Assal and the project economics are likely to be more expensive. Table C.16 and Table C.17 below present the capital cost and O&M assumptions for a geothermal plant located at this site. Again, it should be noted that all estimates are approximate at this stage. Table C.16: Capital cost of a geothermal power development at Lake Abhe Cost Breakdown 3 MW 6 MW Wells (US$ millions) 6.0 9.0 Steam field (US$ millions) 1.0 1.5 Power plant (US$ millions) 8.0 13.0 Site preparation & transmission (US$ millions) 1.0 2.0 Indirect costs (US$ millions) 2.0 3.5 Total Investment (US$ millions) 18.0 29.0 Investment (US$ / kW) 6,000 4,833 9 To be confirmed. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.37 APPENDIX C REVIEW OF GEOTHERMAL RESOURCES IN DJIBOUTI Table C.17: Operating cost (per kWh) of a geothermal power development at Lake Abhe USc/kWh 3 MW 6 MW Steam field 0.18 0.15 Power Plant 0.55 0.45 Off-site & Insurance 0.37 0.30 Total cost 1.10 0.90 It should be noted that Lake Abhe is an isolated area in the south west of the country and as such, a project in this location would likely be for local use only and the development size adjusted to local needs and resource capability. C.8.2 Obock There is no evidence of a potential resource at Obock and therefore an exploration program is needed to confirm the existence of a resource (financing to be arranged first). C.8.3 Arta There is no evidence of a potential resource at Arta and therefore an exploration program is needed to confirm the existence of a resource (financing to be arranged first). Drillings are likely to be deep at more than 2,000 meters which may be dissuasive if the resource is confirmed to be small in size. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 C.38 APPENDIX D REVIEW OF WIND RESOURCES IN DJIBOUTI APPENDIX D REVIEW OF WIND RESOURCES IN DJIBOUTI D. REVIEW OF WIND RESOURCES IN DJIBOUTI PB wind specialists performed a desk-top review of all available data and reports to confirm whether the case for wind development put forward in Djibouti is reasonable. The following list of data has been used in our analysis of the wind resource and investment costs in Djibouti: • Lahmeyer International (2006) – Développement de l’Energie Eolienne – Etude de Préfaisabilité (Final report) • Lahmeyer International (2006) – Global Environment Facility (GEF) Development of Wind Energy and Energy Efficiency (Final report) • Said Ismael Awaleh – Wind Resource Assessment in Djibouti (CERD report) • Djibouti Least Cost Program – Possible Contribution of Indigenous Wind Energies The reports detailed above have been reviewed to identify any potential barriers to wind farm development and to assess the capacity for large scale wind projects in Djibouti. The sites of Ghoubet, Ali Sabieh, Djibouti Ville, Egralyta and Bada Wein have been subject to a high level evaluation. Furthermore, investments and O&M costs from previous reports have been analysed and included in the financial analysis. The unit cost of generation for each wind development opportunity has been estimated from the short-list of options using data in existing reports and from our extensive knowledge of O&M costs. D.1 Wind Resources in Djibouti and identified sites The World Wind Atlas (WWA) has been used to gain an understanding of the areas within Djibouti which would have the most favourable wind regimes to develop wind power projects. The WWA is a database developed by Sander and Partner GmbH that provides wind maps and wind-statistics for specified points around the globe. This utilizes a mathematical model using many sources of low level and atmospheric wind speed data. Figure D.1 shows a wind map of the study area, using WWA data. The wind speed data has been exported from WWA for a height of 50 mAGL. It should be noted that this is a very high level assessment of the wind resource in the area and takes no account of local effects which would alter wind conditions for a specific site. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 D.3 APPENDIX D REVIEW OF WIND RESOURCES IN DJIBOUTI Figure D.1: WWA map of Djibouti   It appears that there is no significant wind speed variation across the country and the difference in energy output for different sites will mainly depend on site specificities such as local topography, vegetation and thermal effects. Additionally, a wind monitoring program, managed by CERD, included the commissioning of several wind monitoring masts across the country which are detailed in Table D.1 and Figure D.2 below. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 D.4 APPENDIX D REVIEW OF WIND RESOURCES IN DJIBOUTI Table D.1: Location of metrological masts in Djibouti Description Easting Northing Altitude Height (m) Gali  Mahaba (a) 42.5556° 11.6206° 500 40 Gali  Ma'aba (b)   42.374959°   11.567182° 4489 40 Ghoubet (a) 42.4889° 11.5355° 150 40 Ghoubbet (b)   42.489864°   11.422726° 230 40 Day 42.6256° 11.7658° 1210 40 Bada Wein (a) 42.6508° 11.238° 550 40 Bada Wein (b)   42.650736°   11.239530° 583 40 Gediah Alleh 42.9522° 11.5402° 230 40 Egralyta 42.5485° 11.4671° 450 40 Yoboki 42.1175° 11.4619° 210 40 Garad Dahol 42.895° 11.3183° 530 40 D Reyaleh   42.665702°   11.425806° 678 40 PK 30 42.8969° 11.5355° 360 40 Arta (a) 42.8971° 11.5378° 370 40 Attar (b)   43.207717°   11.482595° 31 40 Kalaf 42.793° 11.7927° 120 40 PK23   42.958042°   11.538891° 185 40 Daba Riyadle 42.6652° 11.4244° 600 40 Ras Bir   43.309068°   11.996849° 68 40 Ras Ali   42.933388°   11.774251° 49 40 Figure D.2: Location map of metrological Masts in Djibouti. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 D.5 APPENDIX D REVIEW OF WIND RESOURCES IN DJIBOUTI The sites of Ghoubet, Ali Sabieh, Djibouti Ville, Egralyta and Bada Wein have been selected by Lahmeyer and CERD as most promising sites for large scale wind turbines development in terms of location and wind resource. These identified sites would have an overall capacity of 45.9 MW. The energy yield estimations produced by Lahmeyer for these sites (see below) have been reviewed and it is the opinion of PB that they seem realistic. However, the calculations method has not been detailed in the report and it should be noted that these energy yields may be highly uncertain. A complete energy yield assessment including anemometer calibration, long term correction and wind flow modelling should be carried out at a later stage should the sites being investigated any further. The energy yields obtained by Lahmeyer for the sites of Ghoubet, Ali Sabieh, Djibouti Ville, Egralyta and Bada Wein used in the financial analysis are detailed in Table D.2. Table D.2: Energy yield of wind power projects. Sites Wind Farm Capacity (MW) AEP (MWh) Capacity Factor (%) Ghoubet #1 10.2 42,347 47.4 Ali Sabieh #2 10.2 34,611 38.7 Djibouti Ville #3 10.2 27,233 30.5 Egralyta #4 10.2 32,173 36.0 Bada Wein #5 5.1 13,637 30.5 Total 45.9 150,001 D.2 Wind Turbine Generator selection: At this stage of the development, Lahmeyer recommended the use of medium scale wind turbines (850 kW – Gamesas G52, Vestas V52 or Repower R48)) which are well suited for a wide range of terrain and for a broad spectrum of medium and high winds. Furthermore, the ease of transportation for those turbines would be an advantage as roads in Djibouti may not be suitable for large scale wind turbine transportation. In our opinion any development should not be limited to the use of these turbines and that other models should also be considered during feasibility studies. Nevertheless investment and O&M costs for Vestas V 52 have been used in the financial model. D.3 Costs Estimations: The cost estimations for procurement and commissioning of wind turbines (including crane hire, transportation, foundations, access road and consultancy fees) produced by Lahmeyer have been reviewed and seem realistic. It is assumed that these values represent 2005 price levels. We have assumed a 10.2 per cent increase in price to represent 2008 price levels. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 D.6 APPENDIX D REVIEW OF WIND RESOURCES IN DJIBOUTI Furthermore, grid connection costs have been updated based on present contract prices from other recent projects in Djibouti. The grid connection costs as used in the financial model are set out in Table D.3. Table D.3: Grid connection costs Part Unit cost ($000) 63 kV swgr bay 700 63/20 kV 15 MVA tx 600 20 kV swgr panel 50 63 kV Single Circuit Line (per km) 65 20 kV Cable (per km) 60 20 kV Line (per km) 45 Two scenarios were presented in the report for the sites of Ghoubet and Egralyta as the transmission line might be shared with geothermal plant located in the vicinity of the wind park. According to the IEA, the total investment cost, including turbine, foundation and grid connection typically varies from US$ 1,394 to US$1,881 per kW installed capacity (1 EUR = 1.39373 USD – 25 10 June 2009) . The investment cost defined for the sites of Ghoubet (with 10 km 63 kV line), Ali Sabieh, Djibouti Ville, and Egralyta (with 10 km 63 kV line) are within this range of investment cots. However, the sites of Ghoubet (with 45 km 63 kV line), Egralyta (with 35 km 63 kV line) and Bada Wein are well above this range which could be explained by the high grid connection costs. Operation and Maintenance costs have been estimated to be 3 per cent of the total investment cost, 11 ranging between 1.4 and 2.7 USc per kWh and in line with the industry O&M costs . D.4 Identified sites D.4.1 Ghoubet According to the Lahmeyer report, the site of Ghoubet presents good wind characteristics with a mean recorded wind speed of 8.9 m/s at 40 mAGL. It has been assumed that the Ghoubet wind project can be connected to the 63 kV network. Two grid connection scenarios have been considered, an independent connection or a shared connection with the Lac Assal geothermal plant. Lifetime electricity generation costs have been estimated at 7.43 and 6.64 USc per kWh respectively. CAPEX and Operational Expenditure (OPEX) costs are detailed in Table D.4 below. 10 €1,000 to €1,350/kW for Italy, UK, Netherlands, Portugal, Germany, Japan, Greece, Spain, Canada, Denmark, US and Norway - The Economics of Wind Energy, EWEA (March 2009). 11 Less than €c1/kWh for new 600 kW turbines - The Economics of Wind Energy, EWEA (March 2009). Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 D.7 APPENDIX D REVIEW OF WIND RESOURCES IN DJIBOUTI Table D.4: CAPEX and OPEX for Ghoubet wind power project. Total Capacity (kW) 10,200 Inflation Rate (Average 2005-2008) 10.2% Production (Mwh) 42,347 Capacity Factor 47.4% Cost for 2005 Cost for 2008 Procurement Commissioning Total Total 1000 USD 1000 USD 1000 USD 1000 USD WTG + International Shipping 9,523.2 742.4 10,265.6 11,312.7 National Transportation 0.0 524.8 524.8 578.3 Crane Hire 0.0 166.4 166.4 183.4 Crane Transportation 0.0 70.4 70.4 77.6 Foundations 409.6 147.2 556.8 613.6 Access Road 0.0 460.8 460.8 507.8 Consultancy 0.0 192.0 192.0 211.6 Total for CAPEX (Excl Grid Connection) 9,932.8 2,304.0 12,236.8 13,485.0 Procurement Commissioning Total Total 1000 USD 1000 USD 1000 USD 1000 USD Physical Contingency 0.0 1,254.0 1,254.0 1,381.9 Contingency on costs 0.0 371.2 371.2 409.1 Total for Contingency 0.0 1,625.2 1,625.2 1,791.0 Unit Price Quantity Total 45 km Grid Connection 1000 USD 1000 USD 20 kV Cable 60.0 4.0 240.0 20 kV swgr panel 50.0 4.0 200.0 63/20 kV 15 MVA tx 600.0 1.0 600.0 63 kV Single Circuit Line 65.0 45.0 2,925.0 63 kV swgr bay 700.0 3.0 2,100.0 Total for Grid Connection 6,065.0 CAPEX (1000 USD) 21,340.9 Specific Costs (USD/kW) 2,092.2 Unit Price Quantity Total 10 km Grid Connection 1000 USD 1000 USD 20 kV Cable 60.0 4.0 240.0 20 kV swgr panel 50.0 4.0 200.0 63/20 kV 15 MVA tx 600.0 1.0 600.0 63 kV Single Circuit Line 65.0 10.0 650.0 63 kV swgr bay 700.0 3.0 2,100.0 Total for Grid Connection 3,790.0 CAPEX (1000 USD) 19,065.9 Specific Costs (USD/kW) 1,869.2 45 km line 10 km line Availability (%) 97 97 O&M costs (3% of CAPEX) 640,227.7 571,977.7 O&M Cost (USD / kW) 62.8 56.1 O&M Cost (USD / kWh) 0.015 0.014 Build Period (Years) 1 1 Life of Plant (Years) 20 20 Layout 12 Vestas V52 12 Vestas V52 Total Capacity (kW) 10,200 10,200 Production (Mwh) 42,347 42,347 Capacity Factor 47% 47% O&M costs (3% of CAPEX) 640227.72 571977.72 O&M Cost (USD / kW) 62.77 56.08 O&M Cost (USD / kWh) 0.015 0.014 CAPEX (1000 USD) 21,340.9 19,065.9 Specific Costs (USD/kW) 2,092.2 1,869.2 Lifetime Cost (USc / kWh) Discount Rate 10% 7.43 6.64 Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 D.8 APPENDIX D REVIEW OF WIND RESOURCES IN DJIBOUTI D.4.2 Ali Sabieh It has been assumed that the Ali Sabieh wind project can be connected directly to the 20 kV network which lower the grid connection costs. CAPEX and OPEX costs are detailed in Table D.5. Table D.5: CAPEX and OPEX costs for Ali Sabieh wind power project Total Capacity (kW) 10,200 Inflation Rate (Average 2005-2008) 10.2% Production (Mwh) 34,611 Capacity Factor 38.7% Cost for 2005 Cost for 2008 Procurement Commissioning Total Total 1000 USD 1000 USD 1000 USD 1000 USD WTG + International Shipping 9,523.2 742.4 10,265.6 11,312.7 National Transportation 0.0 460.8 460.8 507.8 Crane Hire 0.0 166.4 166.4 183.4 Crane Transportation 0.0 70.4 70.4 77.6 Foundations 409.6 147.2 556.8 613.6 Access Road 0.0 409.6 409.6 451.4 Consultancy 0.0 192.0 192.0 211.6 Total for CAPEX (Excl Grid Connection) 9,932.8 2,188.8 12,121.6 13,358.0 Procurement Commissioning Total Total 1000 USD 1000 USD 1000 USD 1000 USD Physical Contingency 0.0 1,254.4 1,254.4 1,382.3 Contingency on costs 0.0 371.2 371.2 409.1 Total for Contingency 0.0 1,625.6 1,625.6 1,791.4 Unit Price Quantity Total 1000 USD 1000 USD 20 kV Cable 60.0 4.0 240.0 20 kV swgr panel 50.0 4.0 200.0 Total for Grid Connection 440.0 CAPEX (1000 USD) 15,589.4 Specific Costs (USD/kW) 1,528.4 Availablilty (%) 97 O&M costs (3% of CAPEX) 467,682.4 O&M Cost (USD / kW) 45.9 O&M Cost (USD / kWh) 0.014 Build Period (Years) 1 Life of Plant (Years) 20 Layout 12 Vestas V52 Total Capacity (kW) 10,200 Production (Mwh) 34,611 Capacity Factor 39% O&M Cost (USD / kW) 45.85 O&M Cost (USD / kWh) 0.014 Build Period (Years) 1.00 CAPEX (1000 USD) 15,589.4 Specific Costs (USD/kW) 1,528.4 Lifetime Cost (USc / kWh) Discount Rate 10% 6. 65 Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 D.9 APPENDIX D REVIEW OF WIND RESOURCES IN DJIBOUTI D.4.3 Djibouti Ville It has been assumed that the Djibouti Ville wind project can be connected directly to the 20 kV network which lowers the grid connection costs. CAPEX and OPEX costs are shown in Table D.6. Table D.6: CAPEX and OPEX costs for Djibouti Ville wind power project. Total Capacity (kW) 10,200 Inflation Rate (Average 2005-2008) 10.2% Production (Mwh) 27,233 Capacity Factor 30.5% Cost for 2005 Cost for 2008 Procurement Commissioning Total Total 1000 USD 1000 USD 1000 USD 1000 USD WTG + International Shipping 9,523.2 453.1 9,976.3 10,993.9 National Transportation 0.0 265.6 265.6 292.7 Crane Hire 0.0 101.6 101.6 111.9 Crane Transportation 0.0 43.0 43.0 47.4 Foundations 409.6 89.8 499.4 550.4 Access Road 0.0 250.0 250.0 275.5 Consultancy 0.0 117.2 117.2 129.1 Total for CAPEX (Excl Grid Connection) 9,932.8 1,320.3 11,253.1 12,400.9 Procurement Commissioning Total Total 1000 USD 1000 USD 1000 USD 1000 USD Physical Contingency 0.0 765.6 765.6 843.7 Contingency on costs 0.0 226.6 226.6 249.7 Total for Contingency 0.0 992.2 992.2 1,093.4 Unit Price Quantity Total 1000 USD 1000 USD 20 kV Cable 60.0 4.0 240.0 20 kV swgr panel 50.0 4.0 200.0 20 kV Line 45.0 10.0 450.0 20 kV swgr panel 50.0 1.0 50.0 Total for Grid Connection 940.0 CAPEX (1000 USD) 14,434.3 Specific Costs (USD/kW) 1,415.1 Availablilty (%) 97 O&M costs (3% of CAPEX) 433,029.6 O&M Cost (USD / kW) 42.5 O&M Cost (USD / kWh) 0.016 Build Period (Years) 1 Life of Plant (Years) 20 Layout 12 Vestas V52 Total Capacity (kW) 10,200 Production (Mwh) 27,233 Capacity Factor 30% O&M Cost (USD / kW) 42.45 O&M Cost (USD / kWh) 0.016 Build Period (Years) 1.00 CAPEX (1000 USD) 14,434.3 Specific Costs (USD/kW) 1,415.1 Lifetime Cost (USc / kWh) Discount Rate 10% 7.81 Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 D.10 APPENDIX D REVIEW OF WIND RESOURCES IN DJIBOUTI D.4.4 Egralyta It has been assumed that the Egralyta wind project can be connected to the 63 kV network. Two scenarios have been considered connecting to an existing grid connection. CAPEX and OPEX costs are shown in Table D.7. Table D.7: CAPEX and OPEX costs for Egralyta wind power project. Total Capacity (kW) 10,200 Inflation Rate (Average 2005-2008) 10.2% Production (Mwh) 32,173 Capacity Factor 36.0% Cost for 2005 Cost for 2008 Procurement Commissioning Total Total 1000 USD 1000 USD 1000 USD 1000 USD WTG + International Shipping 9,523.2 742.4 10,265.6 11,312.7 National Transportation 0.0 524.8 524.8 578.3 Crane Hire 0.0 166.4 166.4 183.4 Crane Transportation 0.0 70.4 70.4 77.6 Foundations 409.6 142.2 551.8 608.1 Access Road 0.0 460.8 460.8 507.8 Consultancy 0.0 192.0 192.0 211.6 Total for CAPEX (Excl Grid Connection) 9,932.8 2,299.0 12,231.8 13,479.4 Procurement Commissioning Total Total 1000 USD 1000 USD 1000 USD 1000 USD Physical Contingency 0.0 1,254.4 1,254.4 1,382.3 Contingency on costs 0.0 371.2 371.2 409.1 Total for Contingency 0.0 1,625.6 1,625.6 1,791.4 Unit Price Quantity Total 1000 USD 1000 USD 20 kV Cable 60.0 4.0 240.0 20 kV swgr panel 50.0 4.0 200.0 63/20 kV 15 MVA tx 600.0 1.0 600.0 63 kV Single Circuit Line 65.0 35.0 2,275.0 63 kV swgr bay 700.0 3.0 2,100.0 Total for Grid Connection 5,415.0 CAPEX (1000 USD) 20,685.9 Specific Costs (USD/kW) 2,028.0 Unit Price Quantity Total 1000 USD 1000 USD 20 kV Cable 60.0 4.0 240.0 20 kV swgr panel 50.0 4.0 200.0 63/20 kV 15 MVA tx 600.0 1.0 600.0 63 kV Single Circuit Line 65.0 10.0 650.0 63 kV swgr bay 700.0 3.0 2,100.0 Total for Grid Connection 3,790.0 CAPEX (1000 USD) 19,060.9 Specific Costs (USD/kW) 1,868.7 Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 D.11 APPENDIX D REVIEW OF WIND RESOURCES IN DJIBOUTI 35 km line 10 km line Availablilty (%) 97 97 O&M costs (3% of CAPEX) 620,575.6 571,825.6 O&M Cost (USD / kW) 60.8 56.1 O&M Cost (USD / kWh) 0.019 0.018 Build Period (Years) 1 1 Life of Plant (Years) 20 20 Layout 12 Vestas V52 12 Vestas V52 Total Capacity (kW) 10,200 10,200 Production (Mwh) 32,173 32,173 Capacity Factor 36% 36% O&M costs (3% of CAPEX) 620575.64 571825.64 O&M Cost (USD / kW) 60.84 56.06 O&M Cost (USD / kWh) 0.02 0.02 CAPEX (1000 USD) 20,685.9 19,060.9 Specific Costs (USD/kW) 2,028.0 1,868.7 Lifetime Cost (USc / kWh) Discount Rate 10% 9.48 8.74 D.4.5 Bada Wein It has been assumed that the Bada Wein wind project can be connected to the 63 kV network. CAPEX and OPEX costs are shown in Table D.8. Table D.8: CAPEX and OPEX costs for Bada Wein wind power project Total Capacity (kW) 5,100 Inflation Rate (Average 2005-2008) 10.2% Production (Mwh) 13,637 Capacity Factor 30.5% Cost for 2005 Cost for 2008 Procurement Commissioning Total Total 1000 USD 1000 USD 1000 USD 1000 USD WTG + International Shipping 4,889.6 371.2 5,260.8 5,797.4 National Transportation 0.0 307.2 307.2 338.5 Crane Hire 0.0 96.0 96.0 105.8 Crane Transportation 0.0 70.4 70.4 77.6 Foundations 199.7 74.2 273.9 301.9 Access Road 0.0 153.6 153.6 169.3 Consultancy 0.0 192.0 192.0 211.6 Total for CAPEX (Excl Grid Connection) 5,089.3 1,264.6 6,353.9 7,002.0 Procurement Commissioning Total Total 1000 USD 1000 USD 1000 USD 1000 USD Physical Contingency 0.0 934.4 934.4 1,029.7 Contingency on costs 0.0 204.8 204.8 225.7 Total for Contingency 0.0 1,139.2 1,139.2 1,255.4 Unit Price Quantity Total 1000 USD 1000 USD 20 kV Cable 60.0 4.0 240.0 20 kV swgr panel 50.0 4.0 200.0 63/20 kV 15 MVA tx 600.0 1.0 600.0 63 kV Single Circuit Line 65.0 15.0 975.0 63 kV swgr bay 700.0 3.0 2,100.0 Total for Grid Connection 4,115.0 CAPEX (1000 USD) 12,372.4 Specific Costs (USD/kW) 2,426.0 Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 D.12 APPENDIX D REVIEW OF WIND RESOURCES IN DJIBOUTI Availablilty (%) 97 O&M costs (3% of CAPEX) 371,172.5 O&M Cost (USD / kW) 72.8 O&M Cost (USD / kWh) 0.027 Build Period (Years) 1 Life of Plant (Years) 20 Layout 12 Vestas V52 Total Capacity (kW) 5,100 Production (Mwh) 13,637 Capacity Factor 31% O&M costs (3% of CAPEX) 371172.55 O&M Cost (USD / kW) 72.78 O&M Cost (USD / kWh) 0.03 CAPEX (1000 USD) 12,372.4 Specific Costs (USD/kW) 2,426.0 Lifetime Cost (USc / kWh) Discount Rate 10% 13.39 D.5 Conclusions The desktop analysis based on the supplied information shows that there is a potential for wind farm development in Djibouti. The sites of Ghoubet, Ali Sabieh, Djibouti Ville, Egralyta and Bada Wein can generate electricity at cots ranging between 6.64 and 13.39 USc per kWh. Further analysis is required to investigate the possibility of installing bigger wind turbines. A breakdown of the cost of energy from these projects is shown in Table D.9. For onshore wind power projects, the European Wind Energy Association (EWEA) indicates costs should be between 0.07 -0.09 USD per kWh at windy sites and between 0.1 – 0.14 USD per kWh for low average wind 12 speed sites. Our analysis indicates that all of the projects lie within this range and that most are at the favourable end of the scale. Table D.9: Cost of energy from wind power projects. Project Cost per kWh (USc) Ghoubet (connection option 1) 7.43 Ghoubet (connection option 2) 6.64 Ali Sabieh 6.65 Djibouti Ville 7.81 Egralyta (connection option 1) 9.48 Egralyta (connection option 2) 8.74 Bada Wein 13.39 The wind monitoring program, managed by CERD, included the commissioning of several wind monitoring mast across the country and there may be a need to carry additional data analysis on this data. Furthermore, the energy yield predicted high uncertainty as some anemometers were not calibrated and the calculation approach has not been detailed. 12 EWEA, The Economics of Wind Energy, March 2009. Using exchange rate of 1 EUR = 1.39373 USD – 25 June 2009. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 D.13 APPENDIX E DESCRIPTION OF ASPLAN LEAST COST GENERATION PLANNING SOFTWARE APPENDIX E DESCRIPTION OF ASPLAN LEAST COST GENERATION PLANNING SOFTWARE E. DESCRIPTION OF ASPLAN LEAST-COST GENERATION PLANNING SOFTWARE E.1 Approach to modelling – the ASPLAN model ASPLAN is a pc based generation expansion planning model licensed by Analytical Solutions of the USA. ASPLAN has been developed from WASP, the best known and most widely used generation expansion planning model. It uses the same Booth-Baleriaux algorithm for probabilistic production simulation using equivalent load duration curves (ELDCs) to model energy despatch and system security (i.e. loss of load probability and energy not served), and the same optimality techniques using Bellman's Principle of Optimality to select the least cost plan from thousands of possible options. The model comprises of three main functions, the load forecast, cost production simulation, and generation plan optimisation. These are described in turn below. E.1.1 Load forecast The starting point for the generation expansion planning is the demand forecast. The cost production simulations are based on the manipulation of load duration curves representing the demand of the system. The annual forecast power and energy demands are also entered allowing for changes in the system annual load factor. E.1.2 Cost production simulation The program incorporates a probabilistic production costing model which calculates the expected amount of energy generated by each set on the system together with the fuel and O&M cost associated with plant operation. Simulations may be carried out on an annual, seasonal or monthly basis. For this study, to better incorporate the seasonal impact on planned maintenance, simulations of despatch are performed for each month and aggregated for the year. Data requirements for simulation include: • cost and operating assumptions for existing plant; • technical and economic assumptions regarding the modelling methodology (price basis, time horizon, etc) and the system as a whole {system security criteria, etc); • fuel price forecasts; • cost and operating assumptions for potential new (candidate) plant. The simulation process works as follows, taking into account merit order despatch and planned and forced outage rates: Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 E.3 APPENDIX E DESCRIPTION OF ASPLAN LEAST COST GENERATION PLANNING SOFTWARE • the simulation is performed for each month, with plant on planned maintenance in that month removed from consideration. Planned maintenance is scheduled systematically into months with the highest monthly reserve margins, i.e. predominantly into the Spring and Autumn months which have lower peak demands; • the lowest variable cost generating unit is despatched first under the load duration curve for the month. This determines it's expected despatch when it is available; it's own forced outage rate is factored in to determine its overall level of despatch; • the second lowest cost unit is then despatched. The despatch of this unit is dependent on whether the first is available, and to reflect this, an ELDC is derived to take account of the forced outage of the first unit. The second unit is then despatched under the ELDC, after which account of its own forced outage is made; • subsequent units are despatched following least cost merit order despatch. Each time, a new ELDC is derived which takes account of the forced outages of plant despatched previously; • the above process results in the calculation of the amount of generation despatched by each plant in each month. This is summed to give annual totals. The calculation is extended to determine total fuel costs O&M costs for the year; • after all plant has been despatched there will remain a small amount of generation which, in theory, cannot be met. This is the energy not served (ENS) which is also calculated as part of the simulation process along with the loss of load probability (LOLP); The above simulation process is performed for different combinations of existing and new (candidate) plant in different years as directed under the optimisation process (see below). For each combination, the key results retained by the model is the annual values for total fuel and O&M costs, reserve margin, ENS and LOLP. E.1.3 Generation plan optimisation The simulations may be carried out either for pre-specified generation plant programmes, or more commonly the model may be used in its optimisation mode. The latter uses a dynamic programming algorithm to determine the least cost generation expansion plan which meets acceptable levels of system security given a range of thermal candidate plants for possible inclusion in the programmes. Once an optimum expansion programme has been determined, a re-simulation may be carried for the optimised plan to provide detailed findings for further analysis. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 E.4 APPENDIX E DESCRIPTION OF ASPLAN LEAST COST GENERATION PLANNING SOFTWARE Within the optimisation process, the range of plant combinations to be considered must be defined. This range must be limited to avoid excessive run times though thousands of possible generation plans can still be considered each time. The range must be adjusted for each run until an unconstrained solution is found. The optimisation balances lowest cost with acceptable system security. The planning criteria may be specified either in terms of minimum or annual reserve margin, ENS or LOLP. If criteria are defined, plant combinations that fail the criteria are removed from further consideration. Alternatively an unconstrained solution may be obtained where the cost of ENS to the system is part of the least cost calculation and provides the incentive to add new capacity. The cost of each generation plan is calculated as a net present value (at a pre-specified discount rate) of all costs including the annual fuel and O&M costs as well as the investment costs of bringing new plant on line. From the plans which meet the system security criteria, the generation plan with the lowest cost is identified. Though the model effectively considers thousands of possible generation plans, the optimisation algorithm allows the numbers of simulations, which take up most of the processing time, to be minimised. The net result remains the least-cost generation plan that still meets the system security criteria. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 E.5 APPENDIX F GENERATION PLANNING RESULTS APPENDIX F GENERATION PLANNING RESULTS F. GENERATION PLANNING RESULTS F.1 Investment and operating cost schedules for base demand forecast scenarios In this section we present the cost schedules for each base demand forecast development scenario (Scenarios 1 to 12) carried out as part of this study. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.3 APPENDIX F GENERATION PLANNING RESULTS Table F.1: Scenario 1 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.4 APPENDIX F GENERATION PLANNING RESULTS Table F.2: Scenario 2 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.5 APPENDIX F GENERATION PLANNING RESULTS Table F.3: Scenario 3 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.6 APPENDIX F GENERATION PLANNING RESULTS Table F.4: Scenario 4 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.7 APPENDIX F GENERATION PLANNING RESULTS Table F.5: Scenario 5 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.8 APPENDIX F GENERATION PLANNING RESULTS Table F.6: Scenario 6 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.9 APPENDIX F GENERATION PLANNING RESULTS Table F.7: Scenario 7 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.10 APPENDIX F GENERATION PLANNING RESULTS Table F.8: Scenario 8 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.11 APPENDIX F GENERATION PLANNING RESULTS Table F.9: Scenario 9 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.12 APPENDIX F GENERATION PLANNING RESULTS Table F.10: Scenario 10 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.13 APPENDIX F GENERATION PLANNING RESULTS Table F.11: Scenario 11 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.14 APPENDIX F GENERATION PLANNING RESULTS Table F.12: Scenario 12 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.15 APPENDIX F GENERATION PLANNING RESULTS F.2 High demand forecast results Similar to the base case demand forecast, we have investigated a number of high demand forecast developments comprising combinations of liquid fuel only, energy imports over the interconnector and the development of geothermal generation. The high case demand forecast scenarios undertaken in this study are summarised in Figure F.1 and discussed in the sub-sections below. Figure F.1: High demand forecast scenarios High Case  Demand  Forecast Reference  Case  Reference  Case with Geothermal  Development  Only Zero Imports Zero Imports (Scenario  13) (Scenario  16) Up to 700 GWh  180 GWh Imports Imports (Scenario  14) (Scenario  17) 700 GWh Imports (Scenario  15) F.2.1 Reference case Similar to the analysis previously presented, we have derived a least cost generation expansion plan for the high demand forecast scenario assuming a liquid fuel development relying purely of HFO and diesel-fired generating units. The least cost generation expansion plan that satisfies the planning criteria is set out in Table F.13 below and in the capacity chart presented in Figure F.2. The table indicates that 24 MW of capacity is required in 2013 and a total of 218 MW would be required by the end of the planning period. The development over the planning period would comprise eleven 12 MW and seven 7 MW HFO-fired diesel units, one 15 MW and one 7 MW open cycle gas turbine generating sets. These gas turbine generating sets would replace the retired units at the Marabout power station. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.16 APPENDIX F GENERATION PLANNING RESULTS On the basis of the planting programme identified below, we have simulated the reference case system assuming 3 alternate energy import scenarios; zero energy imports, 180 GWh of energy imports and 700 GWh of energy imports over the interconnector. The results of these scenarios are discussed below. Table F.13: Reference case planting programme Year Existing Capacity Capacity Added Total Capacity Peak Demand Reserve Margin MW Net MW Net Type MW Net MW % 2009 97 0 97 68 42 2010 97 0 97 75 30 2011 112 0 112 85 32 2012 112 0 112 101 11 2013 107 24 2xDIESEL HFO 12MW 131 118 11 2014 113 31 1xDIESEL HFO 7MW 144 130 11 2015 98 56 2xDIESEL HFO 12MW 154 140 10 2016 119 56 174 152 15 2017 119 56 174 157 11 2018 119 68 1xDIESEL HFO 12MW 187 162 15 2019 116 68 184 167 10 2020 114 76 1xGAS TURBINE 7MW 190 171 11 2021 111 88 1xDIESEL HFO 12MW 199 175 14 2022 109 88 197 179 10 2023 104 100 1xDIESEL HFO 12MW 204 182 12 2024 104 112 1xDIESEL HFO 12MW 217 186 16 2025 104 112 217 191 13 2026 104 112 217 196 10 2027 104 125 1xDIESEL HFO 12MW 229 201 14 2028 104 125 229 206 11 2029 104 132 1xDIESEL HFO 7MW 236 212 11 2030 82 159 1xDIESEL HFO 12MW 1xGAS TURBINE 15M 241 217 11 2031 73 173 2xDIESEL HFO 7MW 246 223 10 2032 73 180 1xDIESEL HFO 7MW 253 228 11 2033 67 192 1xDIESEL HFO 12MW 259 234 10 2034 55 211 1xDIESEL HFO 7MW 1xDIESEL HFO 12MW 266 240 11 2035 55 218 1xDIESEL HFO 7MW 273 247 11 Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.17 APPENDIX F GENERATION PLANNING RESULTS Figure F.2: Reference case capacity chart Wind Interconnector New Gas Turbines Marabout Boulaos New Diesel Geothrmal Demand Forecast Reserve Margin 300 45% 40% 250 35% Capacity / Demand (MW) 200 30% Reserve Margin (%) 25% 150 20% 100 15% 10% 50 5% 0 0% 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 Zero energy imports (Scenario 13) The level of generation by plant type, assuming that there are no energy imports over the interconnector throughout the forecast period, is presented in Figure F.3 below. The results of this scenario are similar to the equivalent base demand forecast scenario (Scenario 1), such that the new diesel capacity (more efficient and reliable than the existing diesel units) occupies the base load duty within the merit order. The new diesel capacity would be followed by the units at the Boulaos power station and the Marabout units (until retired) and the open cycle gas turbines (thereafter) would undertake the peaking duties. The net present value (NPV) calculated for the capital and operating cost streams for this development plan amounts to US$ 1,138 million over the planning period for this study. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.18 APPENDIX F GENERATION PLANNING RESULTS Figure F.3: Generation by plant type (zero energy imports) New Gas Turbines Marabout Boulaos New Diesel Generation (GWh) 1600 1400 1200 1000 Generation (GWh) 800 600 400 200 0 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 180 GWh energy imports (Scenario 14) Similar to the equivalent base case demand forecast results (Scenario 2), the introduction of energy imports over the interconnector displaces the expensive peaking and mid merit order plant resulting in a significant reduction to the NPV. At an energy import level of 180 GWh per annum, the NPV calculated for the capital and operating cost streams for this development plan amounts to US$ 984 million over the planning period for this study. This represents a saving of US$ 154 million (14 per cent) over the planning period (compared with scenario 13). 700 GWh energy imports (Scenario 15) As with the base demand forecast analysis, the scale of the NPV reduction increases with increased levels of imports. At an energy import level of up to 700 GWh per annum, the NPV calculated for the capital and operating cost streams for this development plan amounts to US$ 960 million over the planning period for this study. This represents a saving of US$ 471 million (41 per cent) over the planning period (compared with scenario 13). Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.19 APPENDIX F GENERATION PLANNING RESULTS F.2.2 Reference case with geothermal development only We have derived a generation expansion plan for the high demand forecast scenario assuming a liquid fuel development relying on HFO and diesel-fired generating units together with the development of up to 60 MW of geothermal power. The least cost generation expansion plan that satisfies the planning criteria is set out in Table F.14. The table indicates that 24 MW of capacity is required in 2013 and a total of 218 MW would be required by the end of the planning period. Table F.14: Least cost generation plan with 60 MW of geothermal plant High demand forecast Year Existing Capacity Capacity Added Total Capacity Peak Demand Reserve Margin MW Net MW Net Type MW Net MW % 2009 97 0 97 68 42 2010 97 0 97 75 30 2011 112 0 112 85 32 2012 112 0 112 101 11 2013 107 24 2xDIESEL HFO 12MW 131 118 11 2014 113 31 1xDIESEL HFO 7MW 144 130 11 2015 98 58 HFO 7MW 1xDIESEL HFO 12MW 1xGAS TUR 157 140 12 2016 119 78 1xGEOTHERMAL 20MW 197 152 30 2017 119 78 197 157 25 2018 119 78 197 162 21 2019 116 98 1xGEOTHERMAL 20MW 215 167 29 2020 114 98 212 171 24 2021 111 98 210 175 20 2022 109 118 1xGEOTHERMAL 20MW 227 179 27 2023 104 118 223 182 22 2024 104 118 223 186 19 2025 104 118 223 191 16 2026 104 118 223 196 14 2027 104 126 1xGAS TURBINE 7MW 230 201 15 2028 104 126 230 206 12 2029 104 133 1xDIESEL HFO 7MW 237 212 12 2030 82 158 2xDIESEL HFO 12MW 240 217 10 2031 73 172 1xDIESEL HFO 7MW 1xGAS TURBINE 7MW 245 223 10 2032 73 179 1xDIESEL HFO 7MW 252 228 10 2033 67 192 1xDIESEL HFO 12MW 258 234 10 2034 55 211 1xDIESEL HFO 7MW 1xDIESEL HFO 12MW 266 240 10 2035 55 218 1xDIESEL HFO 7MW 273 247 10 Zero energy imports (scenario 16) The level of generation by plant type and the plant utilisation factors assuming zero energy imports over the interconnector are presented in Figure F.4. This figure highlights the significant displacement of thermal generation throughout the forecast period. The NPV calculated for the capital and operating cost streams for this development plan amounts to US$ 960 million over the planning period for this study, US$ 178 million (16 per cent) lower than the reference case (scenario 13). This confirms that, for the cost and performance parameters assumed in this study, geothermal generation is part of the least cost plan for EdD. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.20 APPENDIX F GENERATION PLANNING RESULTS Figure F.4: Energy allocation with 60 MW geothermal plant High demand forecast New Gas Turbines Wind Marabout Boulaos New Diesel Interconnector Geothrmal Total Generation (GWh) 1,600 1,400 1,200 1,000 Generation (GWh) 800 600 400 200 0 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 Up to 700 GWh energy imports (Scenario 17) The introduction of energy imports over the interconnector (up to 700 GWh) in addition to the development of the geothermal plant, maximises the displacement of diesel units giving the lowest net present value. The NPV calculated for the capital and operating cost streams for this development plan amounts to US$ 600 million over the planning period for this study, US$ 538 million (48 per cent) lower than the reference case (scenario 13). This confirms that, for the cost and performance parameters assumed in this study, geothermal generation is part of the least cost plan for EdD. It should be noted that the operational requirement to take all generated geothermal energy and maintaining some 15 MW of diesel generation on line at all times, results in restricting energy imports in the medium term during the development phase of the geothermal generating plant as shown in Figure F.5. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.21 APPENDIX F GENERATION PLANNING RESULTS Figure F.5: Energy allocation with 60 MW of geothermal and Up to 700 GWh of imports – High demand forecast Interconnector New Gas Turbines Marabout Boulaos New Diesel New Geothermal Generation (GWh) 1600 1400 1200 687 654 620 591 564 1000 532 503 Generation (GWh) 477 450 425 400 376 354 335 800 472 452 430 565 540 511 600 531 479 414 304 0 181 400 0 200 0 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 F.2.3 High demand forecast development plans - comparison Table F.15 sets out a summary of the NPVs calculated for the derived high demand forecast development plans. Table F.15: Summary table for NPV values for scenarios considered High demand forecast scenario Scenario NPV Import Scenario Geothermal Scenario Wind Scenario No. (US$m) 17 Geo Import Geothermal (60 MW) No Wind 600 15 Up to 700 GWh Import No Geothermal No Wind 667 16 No Imports Geothermal (60 MW) No Wind 960 14 180 GWh Import No Geothermal No Wind 984 13 No Imports No Geothermal No Wind 1138 The investment and operating cost schedules for each scenario discussed above are presented in the Tables below. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.22 APPENDIX F GENERATION PLANNING RESULTS Table F.16: Scenario 13 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.23 APPENDIX F GENERATION PLANNING RESULTS Table F.17: Scenario 14 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.24 APPENDIX F GENERATION PLANNING RESULTS Table F.18: Scenario 15 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.25 APPENDIX F GENERATION PLANNING RESULTS Table F.19: Scenario 16 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.26 APPENDIX F GENERATION PLANNING RESULTS Table F.20: Scenario 17 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.27 APPENDIX F GENERATION PLANNING RESULTS F.3 Low demand forecast results The electricity system of Djibouti is subject to several key vulnerabilities. Specific and minor events can a have a significant impact on development plans, derailing/delaying investment programmes and plans such as those laid out in the base and high demand forecast scenarios of this master plan. These events could be things like a delay in the commissioning of the interconnector, delays in construction of other thermal plant, delays in geothermal development, recession, and various other demand affecting factors. The result (or consequence) of any of these events occurring would be a fall in demand and/or a delay in commissioning of the next best plant. As such, the low demand forecast scenarios detailed here provide a pessimistic expansion plan outlook. Similar to the base case demand forecast, we have investigated a number of developments comprising combinations of liquid fuel only, energy imports over the interconnector and the development of geothermal generation. The low case demand forecast scenarios undertaken as part of this study are summarised in Figure F.6 below. Figure F.6: Low demand forecast scenarios Low Case  Demand  Forecast Reference  Case  Reference  Case with Geothermal  Development  Only Zero Imports Zero Imports (Scenario  18) (Scenario  21) Up to 700 GWh  180 GWh Imports Imports (Scenario  19) (Scenario  22) 700 GWh Imports (Scenario  20) Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.28 APPENDIX F GENERATION PLANNING RESULTS F.3.1 Reference case Similar to the analysis previously presented, we have derived a least cost generation expansion plan for the low demand forecast scenario assuming a liquid fuel development relying on HFO and diesel- fired generating units. The low demand forecast scenario is significantly lower than the base case, requiring one 12 MW diesel unit in 2015 and a total of 125 MW of new generating units by the end of the planning period. The planting programme comprises a total of nine 12 MW diesel sets and two 7 MW open cycle gas turbine generating sets. The least cost generation expansion plan is presented in the planting programme shown in Table F.21 and in the capacity chart presented in Figure F.7 below. Table F.21: Least cost generation plan – reference case Year Existing Capacity Capacity Added Total Capacity Peak Demand Reserve Margin MW Net MW Net Type MW Net MW % 2009 97 0 97 65 49 2010 97 0 97 67 45 2011 112 0 112 69 61 2012 112 0 112 80 39 2013 107 0 107 84 28 2014 113 0 113 92 22 2015 98 12 1xDIESEL HFO 12MW 111 99 12 2016 119 12 131 113 15 2017 119 24 1xDIESEL HFO 12MW 143 119 20 2018 119 24 143 122 17 2019 116 24 141 123 14 2020 114 37 1xDIESEL HFO 12MW 150 127 19 2021 111 37 148 128 15 2022 109 49 1xDIESEL HFO 12MW 158 130 21 2023 104 49 153 132 16 2024 104 57 1xGAS TURBINE 7MW 161 134 20 2025 104 57 161 137 17 2026 104 57 161 139 16 2027 104 64 1xGAS TURBINE 7MW 169 142 19 2028 104 64 169 144 17 2029 104 64 169 147 15 2030 82 89 2xDIESEL HFO 12MW 171 149 15 2031 73 101 1xDIESEL HFO 12MW 174 152 15 2032 73 101 174 154 13 2033 67 113 1xDIESEL HFO 12MW 180 156 15 2034 55 125 1xDIESEL HFO 12MW 180 159 13 2035 55 125 180 162 11 Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.29 APPENDIX F GENERATION PLANNING RESULTS Figure F.7: Least cost generation plan reference case – capacity chart Low demand forecast scenario Wind Interconnector New Gas Turbines Marabout Boulaos New Diesel Geothrmal Demand Forecast Reserve Margin 200 70% 180 60% 160 50% 140 Capacity / Demand (MW) Reserve Margin (%) 120 40% 100 30% 80 60 20% 40 10% 20 0 0% 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 Zero energy imports (Scenario 18) Similar to the base demand forecast results, in this scenario thermal plant will operate to meet all demand into the future. The net present value (NPV) calculated for the capital and operating cost streams for this development plan amounts to US$ 808 million over the planning period for this study 180 GWh energy imports (Scenario 19) Similar to the equivalent base case demand forecast results (Scenario 2), the introduction of energy imports over the interconnector displaces the expensive peaking and mid merit order plant resulting in a significant reduction to the NPV. At an energy import level of 180 GWh per annum, the NPV calculated for the capital and operating cost streams for this development plan amounts to US$ 658 million over the planning period for this study. This represents a saving of US$ 150 million (19 per cent) over the planning period (in comparison to scenario 18). Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.30 APPENDIX F GENERATION PLANNING RESULTS 700 GWh energy imports (Scenario 20) Whilst the EdD system under the low demand forecast scenario can absorb the full 180 GWh per annum, Figure F.8 shows that with a minimum generation of 25 MW by EdD, the system would not be able to absorb 700 GWh annually (if made available). Nevertheless, the scale of the NPV reduction increases with increased levels of imports. The NPV calculated for the capital and operating cost streams for this development plan amounts to US$ 478 million over the planning period for this study. This represents a saving of US$ 330 million (41 per cent) over the planning period (in comparison to scenario 18). Figure F.8: Energy allocation with up to 700 GWh Low demand forecast Interconnector New Gas Turbines Marabout Boulaos New Diesel Generation (GWh) 1000 900 800 700 644 628 Generation (GWh) 600 615 602 589 579 564 551 537 525 514 498 489 481 470 461 500 442 433 419 389 316 400 281 0 235 0 217 162 300 200 100 0 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 F.3.2 Reference case with geothermal development only We have derived a generation expansion plan for the low demand forecast scenario assuming a liquid fuel development relying purely of HFO and diesel-fired generating units together with the development of up to 40 MW of geothermal power. The reduced geothermal capacity arose from the lower demand growth rate which impacts the minimum load level as well as the peak demand. For the low demand forecast scenario, minimum load does not reach 60 MW. Due to the ‘must take’ nature of the geothermal energy generated, Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.31 APPENDIX F GENERATION PLANNING RESULTS development of geothermal plant was restricted to 40 MW of capacity to enable absorbing the full output from the geothermal development. The earliest commissioning date for the first 20 MW unit is assumed to be 2016. 2019 is assumed to be the earliest commissioning date for the second unit. The least cost generation expansion plan that satisfies the planning criteria is set out in Table F.22. For this scenario, 132 MW of generating capacity is required comprising four 12 MW diesel engines, four 7 MW of diesel engines, two 7 MW of open cycle gas turbine capacity and two 20 MW geothermal generating units. Geothermal units where again, selected as soon as they were made available. The immediate selection is primarily for energy cost savings, maximising the displacement of liquid fuel-fired generating units. Table F.22: Least cost generation plan with 60 MW of geothermal plant Low demand forecast Year Existing Capacity Capacity Added Total Capacity Peak Demand Reserve Margin MW Net MW Net Type MW Net MW % 2009 97 0 97 65 49 2010 97 0 97 67 45 2011 112 0 112 69 61 2012 112 0 112 80 39 2013 107 0 107 84 28 2014 113 0 113 92 22 2015 98 12 1xDIESEL HFO 12MW 111 99 12 2016 119 32 1xGEOTHERMAL 20MW 151 113 33 2017 119 32 151 119 27 2018 119 32 151 122 24 2019 116 52 1xGEOTHERMAL 20MW 168 123 36 2020 114 52 166 127 31 2021 111 52 164 128 27 2022 109 52 161 130 24 2023 104 52 156 132 18 2024 104 52 156 134 16 2025 104 60 1xGAS TURBINE 7MW 164 137 20 2026 104 60 164 139 18 2027 104 60 164 142 16 2028 104 67 1xDIESEL HFO 7MW 171 144 19 2029 104 67 171 147 17 2030 82 91 2xDIESEL HFO 12MW 174 149 16 2031 73 104 1xDIESEL HFO 12MW 176 152 16 2032 73 104 176 154 14 2033 67 111 1xGAS TURBINE 7MW 178 156 14 2034 55 125 2xDIESEL HFO 7MW 180 159 13 2035 55 132 1xDIESEL HFO 7MW 187 162 16 Zero energy imports (Scenario 21) Assuming no energy import over the interconnector resulted in a NPV calculated for the capital and operating cost streams for this development plan amounts to US$ 665 million over the planning period for this study, US$ 143 million (18 per cent) lower than the reference case (Scenario 18). Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.32 APPENDIX F GENERATION PLANNING RESULTS Up to 700 GWh energy imports (Scenario 22) The introduction of energy imports over the interconnector (up to 700 GWh) in addition to the development of the geothermal plant maximises the displacement of diesel units giving the lowest net present value. The NPV calculated for the capital and operating cost streams for this development plan amounts to US$ 443 million over the planning period, US$ 365 (45 per cent) lower than the reference case (Scenario 18). It should be noted that the operational requirement to take all generated geothermal energy and maintaining some 15 MW of diesel generation on line at all times, results in restricting energy imports in the medium term during the development phase of the geothermal generating plant as shown in Figure F.9. Figure F.9: Energy allocation with 60 MW of geothermal and up to 700 GWh of imports – high demand forecast Interconnector New Gas Turbines Marabout Boulaos New Diesel New Geothermal Generation (GWh) 1000 900 800 401 389 376 700 363 350 339 325 312 300 287 275 262 252 243 234 225 210 Generation (GWh) 600 354 340 310 500 316 400 281 0 235 0 217 161 300 200 100 0 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 F.3.3 Low demand forecast scenarios Table F.23 sets out a summary of the NPVs calculated for the derived low demand forecast development plans. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.33 APPENDIX F GENERATION PLANNING RESULTS Table F.23: Summary table for NPV values for scenarios considered Low demand forecast scenario Scenario NPV Import Scenario Geothermal Scenario Wind Scenario No. (US$m) 22 Geo Import Geothermal (40 MW) No Wind 443 20 Up to 700 GWh Import No Geothermal No Wind 478 19 180 GWh Import No Geothermal No Wind 658 21 No Imports Geothermal (40 MW) No Wind 665 18 No Imports No Geothermal No Wind 808 The investment and operating cost schedules for each scenario discussed above are presented in the Tables below. Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.34 APPENDIX F GENERATION PLANNING RESULTS Table F.24: Scenario 18 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.35 APPENDIX F GENERATION PLANNING RESULTS Table F.25: Scenario 19 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.36 APPENDIX F GENERATION PLANNING RESULTS Table F.26: Scenario 20 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.37 APPENDIX F GENERATION PLANNING RESULTS Table F.27: Scenario 21 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.38 APPENDIX F GENERATION PLANNING RESULTS Table F.28: Scenario 22 investment and operating cost schedule Least Cost Electricity Master Plan (Volume 2) Final Report November 2009 Page F.39 APPENDIX G LOAD FLOW RESULTS Figure G.1 - 2009 Load flow DIgSILENT 27.64 -27.51 17.62 -18.77 93.17 93.17 BOULAOS 63 62.92 MARABOUT 63 62.56 1.00 0.99 25.89 25.78 4.83 4.83 -8.97 -15.76 -12.56 27.51 0.00 6.00 6.00 -9.43 -10.98 -9.21 18.77 -0.00 21.42 21.42 71.27 90.87 85.32 93.17 0.00 9.00 15.82 12.61 10.66 12.81 10.80 71.27 90.87 85.32 -4.82 -4.82 -27.30 -0.00 -5.72 -5.72 5.66 -15.22 -0.00 21.42 21.42 1.03 93.17 0.00 -0.74 9.00 15.82 12.61 MARABOUT 15 10.66 12.81 10.80 BOULAOS 20 20.36 73.18 95.37 92.24 MARABOUT 20 20.17 14.90 1.02 1.01 0.99 24.60 20.62 55.78 0.00 -5.80 -4.36 -4.36 -4.36 -4.84 -4.43 -3.22 0.00 0.00 41.00 0.00 0.00 27.30 0.00 -2.88 -1.61 -1.61 -2.22 -1.56 -3.00 -3.09 -0.00 0.00 27.41 -1.53 -1.53 18.28 0.00 73.40 62.21 62.21 65.51 68.05 60.61 50.62 0.00 0.00 G G G ~ ~ ~ G21 G22 G25 -0.00 5.82 4.37 4.37 4.37 4.85 4.44 3.23 -0.00 -0.00 G G G G G G -0.00 3.28 1.96 1.96 2.52 1.98 3.27 3.28 0.00 -0.00 ~ ~ ~ ~ ~ ~ 0.00 73.40 62.21 62.21 65.51 68.05 60.61 50.62 0.00 0.00 BOULAOSLOAD M1 M2 M3 M4 M5 M6 MARABOUTLOAD 5.82 4.37 4.37 4.37 4.85 4.44 3.23 3.28 1.96 1.96 2.52 1.98 3.27 3.28 73.70 63.87 63.87 67.26 69.86 60.87 50.83 G G G G G G G G G G ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ G1 1 G12 G13 G14 G15 G16 G17 G18 G23 G24 Load Flow Balanced Nodes Branches Project: Line-Line Voltage, Magnitude [kV] Active Power [MW] Graphic: Djibouti Transmis Voltage, Magnitude [p.u.] Reactive Power [Mvar] Date: 7/17/2009 Voltage, Angle [deg] Loading [%] PowerFactory 14.0.508 Annex: Figure G.2 - 2011 Dry Season Load Flow DIgSILENT -24.89 -16.94 17.47 17.47 -0.64 0.65 -18.65 -12.39 -12.56 13.20 13.20 -2.37 -0.63 -6.60 -0.63 2.40 1.00 8.63 0.63 8.63 6.60 8.75 0.64 8.75 PK12 20 19.78 0.99 24.09 7.90 -7.90 2.23 -2.23 14.49 19.67 -0.00 228.04 228.72 0.00 -16.90 -16.91 0.99 0.99 20.75 PK-52 230 ADIGALA 230 0.00 9.83 PK12 LOAD 14.49 228.07 0.00 -16.86 0.99 7.91 2.45 20.75 PK12 230 226.09 -16.63 0.98 PK12 63 62.83 1.00 25.51 DIRE DAWA 230 0.95 0.95 -7.39 -2.42 -0.17 -0.17 -3.49 1.37 2.92 2.92 12.65 4.39 7.41 2.42 3.44 -1.47 12.65 4.39 -7.41 -2.42 -3.44 1.47 12.54 7.47 7.41 30.27 -30.14 2.43 1.24 13.56 -14.72 -4.19 12.54 93.49 93.49 7.47 BOULAOS 63 MARABOUT 63 63.13 62.79 1.00 1.00 25.80 25.65 -0.12 -0.12 -8.97 -16.16 -12.30 27.72 0.00 7.20 7.20 -9.00 -10.94 -9.26 18.91 0.00 -0.95 -0.95 19.95 19.95 69.37 92.00 84.06 93.52 0.00 -0.65 -0.65 2.92 2.92 ALI SABIEH 63 9.00 16.23 12.35 62.48 10.16 12.81 10.80 0.99 69.37 92.00 84.06 25.17 1.90 1.30 19.35 0.13 0.13 -27.50 -0.00 -6.95 -6.95 5.66 -15.34 -0.00 19.95 19.95 1.03 93.52 0.00 Ali Sabieh T1 -0.83 MARABOUT 15 9.00 16.23 12.35 10.16 12.81 10.80 BOULAOS 20 71.23 96.84 91.15 MARABOUT 20 20.33 20.25 14.95 1.02 1.01 1.00 25.91 20.49 55.65 -4.09 -4.09 0.00 -5.80 -4.26 -4.26 -4.26 -4.74 -5.80 -0.60 -5.64 0.00 43.30 0.00 0.00 27.50 1.50 1.50 0.00 -3.05 -1.71 -1.71 -2.36 -1.67 -3.05 -3.42 -1.07 0.00 28.94 -1.54 -1.54 18.41 -1.90 55.58 55.58 0.00 74.40 61.65 61.65 65.37 67.36 74.40 39.43 87.59 0.00 G G G -1.26 ~ ~ ~ 19.35 G21 G22 G25 ALI SABIEH 20 -0.00 5.82 4.28 4.28 4.28 4.75 5.82 0.60 5.68 -0.00 G G G G G G 19.66 0.98 -0.00 3.46 2.06 2.06 2.66 2.08 3.46 3.54 1.49 0.00 ~ ~ ~ ~ ~ ~ 4.10 4.10 0.00 74.40 61.65 61.65 65.37 67.36 74.40 39.43 87.59 0.00 24.48 -1.22 -1.22 BOULAOS LOAD M1 M2 M3 M4 M5 M6 1.90 1.26 55.58 55.58 MARABOUT LOAD 4.10 4.10 5.82 4.28 4.28 4.28 4.75 5.82 0.60 5.68 -1.22 -1.22 3.46 2.06 2.06 2.66 2.08 3.46 3.54 1.49 57.07 57.07 74.71 63.29 63.29 67.12 69.15 74.71 39.60 87.59 ALI SABIEH LOAD G G G G G G G G G G G G ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ G32 G31 G11 G12 G13 G14 G15 G16 G17 G18 G23 G24 Load Flow Balanced Nodes Branches Project: Line-Line Voltage, Magnitude [kV] Active Power [MW] Graphic: Djibouti Transmis Voltage, Magnitude [p.u.] Reactive Power [Mvar] Date: 7/20/2009 Voltage, Angle [deg] Loading [%] PowerFactory 14.0.508 Annex: Figure G.3 - 2011 Wet Season Load Flow DIgSILENT -24.35 -28.43 -16.42 28.93 26.08 26.08 -17.82 -30.25 -12.08 -27.05 -26.98 -12.16 30.50 24.36 24.36 27.24 20.54 20.54 27.05 20.59 20.59 -6.42 0.51 6.42 PK12 20 19.93 1.00 -27.03 8.50 -8.50 5.71 -5.71 10.01 27.71 14.29 49.93 230.17 229.13 0.00 -20.67 -22.21 1.00 1.00 25.69 PK-52 230 ADIGALA 230 27.71 14.29 49.93 PK12 LOAD 227.96 -22.64 0.99 -27.65 -27.65 8.51 -12.52 -12.52 6.04 49.93 49.93 25.69 PK12 230 230.00 -18.96 1.00 PK12 63 64.00 1.02 -25.56 DIRE DAWA 230 1.06 1.06 22.08 22.60 -0.13 -0.13 8.82 10.43 3.17 3.17 36.16 37.87 -21.96 -22.47 -8.51 -10.08 36.16 37.87 21.96 22.47 8.51 10.08 37.40 39.66 Voltage Levels 400. kV 230. kV 132. kV 66. kV 63. kV 45. kV 33. kV 20. kV 15. kV -21.93 7.43 -7.42 -22.44 13.8 kV -10.68 6.85 -8.15 -12.78 37.40 30.59 30.59 39.66 11. kV 10.5 kV BOULAOS 63 MARABOUT 63 5.5 kV 63.18 63.07 1.00 1.00 3. kV -26.44 -26.45 20.89 20.89 -1.22 -13.77 -12.30 29.85 0.00 17.25 17.25 -10.24 -11.25 -9.17 20.94 0.00 -1.05 -1.05 75.04 75.04 56.23 83.74 83.70 101.18 0.00 -0.72 -0.72 3.17 3.17 ALI SABIEH 63 1.24 13.82 12.35 63.60 11.00 12.80 10.70 1.01 56.23 83.74 83.70 -25.92 2.10 1.44 21.02 -20.77 -20.77 -29.60 -0.00 -13.80 -13.80 5.70 -16.75 0.00 75.04 75.04 1.04 101.18 0.00 Ali Sabieh T1 -56.08 MARABOUT 15 1.24 13.82 12.35 11.00 12.80 10.70 BOULAOS 20 58.08 88.25 90.78 MARABOUT 20 19.39 20.22 15.02 0.97 1.01 1.00 -32.37 -31.98 3.55 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 -4.96 0.00 46.50 -0.00 -0.00 29.60 -0.00 -0.00 -0.00 0.00 -0.00 -0.00 0.00 -0.00 0.00 0.00 -3.48 0.00 31.08 -1.53 -1.53 19.82 -2.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 96.88 0.00 G G G -1.40 ~ ~ ~ 21.02 G21 G22 G25 ALI SABIEH 20 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 5.00 -0.00 G G G G G G 20.00 1.00 0.00 -0.00 0.00 0.00 -0.00 0.00 -0.00 -0.00 4.00 -0.00 ~ ~ ~ ~ ~ ~ -0.00 -0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 96.88 0.00 -26.65 0.00 0.00 BOULAOS LOAD M1 M2 M3 M4 M5 M6 2.10 1.40 0.00 0.00 MARABOUT LOAD 5.00 4.00 95.57 ALI SABIEH LOAD G G G G G G G G G G G G ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ G32 G31 G11 G12 G13 G14 G15 G16 G17 G18 G23 G24 Load Flow Balanced Nodes Branches Project: Line-Line Voltage, Magnitude [kV] Active Power [MW] Graphic: Djibouti Transmis Voltage, Magnitude [p.u.] Reactive Power [Mvar] Date: 7/28/2009 Voltage, Angle [deg] Loading [%] PowerFactory 14.0.508 Annex: Figure G.4 - 2015 Dry Season Load Flow DIgSILENT -25.82 -17.55 17.80 17.80 -0.64 0.65 -19.39 -12.84 -13.04 13.46 13.46 -2.37 -0.63 -6.86 -0.63 2.40 1.09 8.78 0.63 8.78 6.86 8.91 0.64 8.91 PK12 20 19.92 1.00 24.67 26.00 -13.00 -13.00 17.47 -8.74 -8.74 10.19 20.39 -0.00 232.22 232.92 0.00 -16.91 -16.92 1.01 1.01 40.31 40.31 PK-52 230 ADIGALA 230 10.02 10.20 0.00 0.00 PK12 LOAD 15.03 232.25 0.00 -16.86 1.01 13.02 13.02 9.51 9.51 40.31 40.31 PK12 230 230.22 -16.64 1.00 PK12 63 63.00 1.00 26.90 DIRE DAWA 230 1.61 1.61 -13.55 -38.36 10.35 12.27 0.31 0.31 -1.88 -28.92 4.47 6.68 4.97 4.97 71.97 63.20 17.46 21.63 13.58 38.46 3.13 32.40 71.97 63.20 13.58 38.46 3.13 32.40 76.89 93.13 -10.33 -12.23 -4.47 -6.63 17.46 21.63 G ~ HFO 7 MW G ~ 10.33 4.47 12.23 6.63 HFO 12 MW 18.98 23.83 Voltage Levels 400. kV 230. kV 132. kV 66. kV 63. kV 45. kV 33. kV 20. kV 15. kV -10.32 13.55 13.55 -13.52 -13.52 -12.22 13.8 kV -6.63 6.90 6.90 -8.16 -8.16 -9.31 18.98 44.26 44.26 44.26 44.26 23.83 11. kV 10.5 kV BOULAOS 63 MARABOUT 63 5.5 kV 62.59 62.43 0.99 0.99 3. kV 26.49 26.43 4.65 4.65 0.00 -13.77 -12.30 13.08 13.08 13.08 0.00 6.60 6.60 0.00 -11.23 -9.15 8.54 8.54 8.54 -0.00 -1.60 -1.60 22.58 22.58 0.00 84.47 84.40 43.80 43.80 43.80 0.00 -1.11 -1.11 4.97 4.97 ALI SABIEH 63 -0.00 13.82 12.35 62.26 -0.00 12.81 10.70 0.99 0.00 84.47 84.40 26.39 3.20 2.23 32.90 -4.64 -4.64 -13.03 -13.03 -13.03 -0.00 -6.29 -6.29 5.26 -7.71 -7.71 -7.71 0.00 22.58 22.58 0.96 43.80 43.80 43.80 0.00 Ali Sabieh T1 -3.51 MARABOUT 15 13.82 12.35 12.81 10.70 BOULAOS 20 88.27 90.78 MARABOUT 20 20.19 20.17 14.86 1.01 1.01 0.99 25.24 23.89 56.43 -4.09 -4.10 -5.68 -5.80 -4.26 -4.26 -4.26 -4.74 -5.80 -5.68 -5.67 -1.49 65.10 0.00 0.00 39.10 -3.91 1.05 -2.71 -3.73 -2.10 -2.10 -2.88 -2.06 -3.73 -3.75 -1.86 -3.13 43.52 -1.53 -1.53 26.18 -3.20 72.63 54.23 84.99 78.82 64.21 64.21 69.49 69.74 78.82 77.75 81.83 48.08 G G G -2.13 ~ ~ ~ 32.90 G21 G22 G25 ALI SABIEH 20 5.70 5.82 4.28 4.28 4.28 4.75 5.82 5.70 5.70 1.50 G G G G G G 19.47 0.97 3.21 4.19 2.48 2.48 3.22 2.50 4.19 4.19 2.28 3.28 ~ ~ ~ ~ ~ ~ 4.10 4.10 84.99 78.82 64.21 64.21 69.49 69.74 78.82 77.75 81.83 48.08 25.22 4.40 -0.77 BOULAOS LOAD M1 M2 M3 M4 M5 M6 3.20 2.13 72.63 54.23 MARABOUT LOAD 4.10 4.10 5.70 5.82 4.28 4.28 4.28 4.75 5.82 5.70 5.70 1.50 4.40 -0.77 3.21 4.19 2.48 2.48 3.22 2.50 4.19 4.19 2.28 3.28 80.22 55.67 86.11 79.15 65.93 65.93 71.34 71.60 79.15 78.07 81.83 48.08 ALI SABIEH LOAD G G G G G G G G G G G G ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ G32 G31 G11 G12 G13 G14 G15 G16 G17 G18 G23 G24 Load Flow Balanced Nodes Branches Project: Line-Line Voltage, Magnitude [kV] Active Power [MW] Graphic: Djibouti Transmis Voltage, Magnitude [p.u.] Reactive Power [Mvar] Date: 7/28/2009 Voltage, Angle [deg] Loading [%] PowerFactory 14.0.508 Annex: Figure G.5 - 2015 Wet Season Load Flow DIgSILENT -12.72 -60.04 -19.80 62.37 45.73 45.73 39.57 13.23 69.48 -11.90 -63.02 -59.02 -15.55 -58.68 -19.88 64.05 44.93 44.93 60.02 43.66 43.66 59.02 15.55 44.82 44.82 39.57 13.23 69.48 -2.38 0.81 PK12 20 19.41 0.97 -32.64 28.00 -14.00 -14.00 18.82 -9.41 -9.41 39.57 13.23 69.48 227.31 221.70 -21.27 -24.56 0.99 0.96 44.57 44.57 PK-52 230 ADIGALA 230 PK12 LOAD 15.00 219.25 0.00 -25.58 0.95 -39.46 -39.46 -39.46 14.03 14.03 -9.80 -9.80 -9.80 10.35 10.35 69.48 69.48 69.48 44.57 44.57 PK12 230 230.00 -17.86 1.00 PK12 63 61.61 0.98 22.22 22.22 21.23 21.23 -30.12 1.32 1.32 2.60 2.60 DIRE DAWA 230 1.72 1.72 0.00 0.00 35.13 35.13 33.76 33.76 0.43 0.43 -0.00 -0.00 5.41 5.41 0.00 0.00 -0.00 -0.00 0.00 0.00 0.00 0.00 -22.11 -22.11 -21.12 -21.12 -1.02 -1.02 -2.33 -2.33 35.13 35.13 33.76 33.76 G ~ 22.11 22.11 21.12 21.12 HFO 7 MW G 1.02 1.02 2.33 2.33 ~ 35.31 35.31 34.33 34.33 HFO 12 MW Voltage Levels 400. kV 230. kV 132. kV 66. kV 63. kV 45. kV 33. kV 20. kV 15. kV 13.8 kV 11. kV 10.5 kV 5.5 kV 3. kV -1.70 -1.70 -1.19 -1.19 5.41 5.41 ALI SABIEH 63 60.78 0.96 -30.67 3.40 2.37 35.84 Ali Sabieh T1 -22.08 0.00 0.00 0.00 0.00 -21.09 -3.06 7.26 7.26 -8.49 -8.49 -4.88 35.31 -22.08 24.31 24.31 24.31 24.31 -21.09 34.33 -3.06 MARABOUT 63 -4.88 35.31 34.33 BOULAOS 63 61.16 61.09 0.97 0.97 -3.40 -31.19 -31.13 -2.26 21.53 21.53 21.53 0.00 -13.76 -6.67 14.06 14.06 14.06 0.00 35.84 4.20 4.20 4.20 0.00 -11.16 -9.82 8.91 8.91 8.91 -0.00 62.77 62.77 62.77 0.00 86.23 66.91 47.68 47.68 47.68 0.00 ALI SABIEH 20 Boulaos T3 19.48 0.97 -31.92 -0.00 13.82 6.70 3.40 -0.00 12.81 10.80 2.26 0.00 86.23 66.91 -21.45 -21.45 -21.45 -14.00 -14.00 -14.00 -0.00 -1.79 -1.79 -1.79 5.14 -7.92 -7.92 -7.92 0.00 62.77 62.77 62.77 0.93 47.68 47.68 47.68 0.00 -61.19 MARABOUT 15 13.82 6.70 MARABOUT 20 12.81 10.80 ALI SABIEH LOAD BOULAOS 20 88.27 70.61 20.00 19.68 14.54 1.00 0.98 0.97 -37.45 -33.99 -1.13 -0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 -5.66 0.00 70.00 -0.00 42.00 -38.79 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 -2.64 0.00 46.79 -2.90 28.12 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 86.58 0.00 G G G ~ ~ ~ G21 G22 G25 SVS -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 5.70 -0.00 SVS G G G G G G -0.00 -0.00 -0.00 -0.00 0.00 -0.00 -0.00 -0.00 3.11 -0.00 ~ ~ ~ ~ ~ ~ -0.00 -0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 86.58 0.00 Static Va.. -0.00 -0.00 BOULAOS LOAD Static Va.. M1 M2 M3 M4 M5 M6 0.00 0.00 MARABOUT LOAD 5.70 3.11 86.58 G G G G G G G G G G G G ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Load Flow Balanced G32 G31 G11 G12 G13 G14 G15 G16 G17 G18 G23 G24 Nodes Branches Project: Line-Line Voltage, Magnitude [kV] Active Power [MW] Graphic: Djibouti Transmis Voltage, Magnitude [p.u.] Reactive Power [Mvar] Date: 7/28/2009 Voltage, Angle [deg] Loading [%] PowerFactory 14.0.508 Annex: Figure G.6 - 2020 Dry Season Load Flow DIgSILENT -27.03 -15.01 18.95 18.95 -0.64 0.67 PK12 20 -26.53 -10.48 -14.81 18.68 18.68 10.25 10.25 -2.37 -0.63 -8.79 -0.63 2.45 8.92 7.24 0.63 7.24 8.79 0.64 19.56 0.98 24.05 47.80 -15.93 -15.93 -15.93 32.12 -10.71 -10.71 -10.71 14.52 19.88 -0.00 229.89 230.12 0.00 -17.05 -17.02 1.00 1.00 50.33 50.33 50.33 PK-52 230 ADIGALA 230 0.00 9.68 0.00 9.94 PK12 LOAD 14.52 229.31 0.00 -16.96 1.00 15.97 15.97 15.97 11.91 11.91 11.91 50.33 50.33 50.33 PK12 230 226.28 -16.66 0.98 PK12 63 62.34 0.99 9.67 26.87 2.16 DIRE DAWA 230 -16.70 -16.70 -13.55 -39.64 15.48 9.67 9.67 9.67 Lac Assal 63 6.78 6.78 -9.91 -48.02 2.16 2.16 2.16 46.26 46.26 44.65 82.80 15.48 15.48 15.48 13.58 39.82 10.87 54.00 44.65 82.80 13.58 39.82 10.87 54.00 95.99 74.55 63.53 37.17 1.01 G Lac Assal G1 Lac Assal GTX1 ~ -19.94 20.00 80.40 20.00 80.40 80.40 -2.01 -2.01 3.63 G G HFO 7 MW ~ ~ HFO 12 MW -19.39 19.94 31.01 31.01 -3.63 4.91 Lac Assal GTX2 -9.65 -9.65 -9.65 -9.65 Lac Assal G2 -2.18 -2.18 -2.18 -2.18 15.48 15.48 15.48 15.48 -19.94 20.00 80.40 20.00 80.40 80.40 -2.01 -2.01 3.63 Palmeraie 63 G ~ 62.10 1.69 1.69 13.61 13.61 8.00 -19.39 19.94 31.01 31.01 0.99 -3.63 4.91 26.47 -3.24 -3.24 4.91 4.91 5.38 5.70 5.70 24.29 24.29 Voltage Levels 17.43 46.26 -5.92 1.96 1.01 6.04 400. kV Palmeraie 230. kV 132. kV 17.43 46.26 -5.92 1.96 1.01 6.04 66. kV 63. kV 45. kV 33. kV 63.10 31.54 1.00 20. kV 15. kV 13.8 kV PK-51 63 11. kV 10.5 kV 5.5 kV 3. kV -1.95 -1.95 -1.37 -1.37 6.04 6.04 ALI SABIEH 63 62.63 0.99 31.29 3.90 2.73 39.95 Ali Sabieh T1 -1.69 -1.69 6.39 6.39 -6.38 -6.38 -13.60 -13.60 1.11 1.11 3.51 3.51 -4.78 -4.78 -7.55 -7.55 5.70 5.70 22.50 22.50 22.50 22.50 24.29 24.29 MARABOUT 63 BOULAOS 63 62.11 62.02 0.99 0.98 -3.90 26.45 26.43 -2.60 10.05 10.05 10.05 -13.74 -13.76 -12.04 13.32 13.32 13.32 0.00 39.95 6.87 6.87 6.87 -9.38 -11.21 -9.25 8.22 8.22 8.22 -0.00 34.30 34.30 34.30 92.33 85.06 84.27 44.16 44.16 44.16 0.00 ALI SABIEH 20 20.05 1.00 29.94 13.80 13.82 12.09 3.90 11.44 12.81 10.80 2.60 92.33 85.06 84.27 -10.03 -10.03 -10.03 -13.27 -13.27 -13.27 -0.00 -6.15 -6.15 -6.15 5.62 -7.37 -7.37 -7.37 0.00 34.30 34.30 34.30 1.02 44.16 44.16 44.16 0.00 1.79 MARABOUT 15 13.80 13.82 12.09 11.44 12.81 10.80 MARABOUT 20 ALI SABIEH LOAD BOULAOS 20 94.03 88.27 90.06 20.00 20.05 14.77 1.00 1.00 0.98 23.63 23.81 56.43 -0.00 -4.09 -4.10 0.00 -5.80 -4.26 -4.26 -4.26 -1.20 0.00 0.00 -5.66 0.00 63.70 -0.00 39.80 -1.56 -3.91 0.41 0.00 -4.68 -2.65 -2.65 -3.60 -2.87 -0.00 -0.00 -2.64 -0.00 42.58 -3.01 26.65 73.25 53.32 0.00 85.94 68.41 68.41 76.07 42.38 0.00 0.00 86.58 0.00 G G G ~ ~ ~ G21 G22 G25 SVS -0.00 5.82 4.28 4.28 4.28 1.20 -0.00 -0.00 5.70 -0.00 SVS G G G G G G 0.00 5.22 3.08 3.08 4.00 3.03 0.00 0.00 3.11 0.00 ~ ~ ~ ~ ~ ~ 4.10 4.10 0.00 85.94 68.41 68.41 76.07 42.38 0.00 0.00 86.58 0.00 Static Va.. 4.40 -0.15 BOULAOS LOAD Static Va.. M1 M2 M3 M4 M5 M6 73.25 53.32 MARABOUT LOAD 4.10 4.10 5.82 4.28 4.28 4.28 1.20 5.70 4.40 -0.15 5.22 3.08 3.08 4.00 3.03 3.11 80.22 54.74 86.30 70.23 70.23 78.10 43.51 86.58 G G G G G G G G G G G G ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Load Flow Balanced G32 G31 G11 G12 G13 G14 G15 G16 G17 G18 G23 G24 Nodes Branches Project: Line-Line Voltage, Magnitude [kV] Active Power [MW] Graphic: Djibouti Transmis Voltage, Magnitude [p.u.] Reactive Power [Mvar] Date: 7/28/2009 Voltage, Angle [deg] Loading [%] PowerFactory 14.0.508 Annex: Figure G.7 - 2020 Wet Season Load Flow DIgSILENT -13.22 -60.86 -19.12 63.25 46.11 46.11 40.12 12.77 70.05 PK12 20 -12.40 -63.88 -59.85 -14.88 -59.50 -19.21 64.93 45.59 45.59 60.87 44.09 44.09 59.85 14.88 45.19 45.19 40.12 12.77 70.05 -1.78 0.22 20.00 1.00 -49.39 51.50 -17.17 -17.17 -17.17 -0.00 34.61 -0.56 -0.56 -0.56 -32.92 15.00 40.12 12.77 70.05 227.37 221.86 0.00 -37.36 -40.71 0.99 0.96 44.04 44.04 44.04 PK-52 230 ADIGALA 230 SVS PK12 LOAD Static Va.. 15.00 219.44 0.00 -41.74 0.95 -40.01 -40.01 -40.01 17.20 17.20 17.20 -9.29 -9.29 -9.29 1.48 1.48 1.48 70.05 70.05 70.05 44.04 44.04 44.04 PK12 230 230.00 -33.90 1.00 PK12 63 61.72 0.98 25.40 -46.33 2.82 DIRE DAWA 230 -16.57 -16.57 0.00 0.00 40.25 25.40 25.40 25.40 Lac Assal 63 6.06 6.06 0.00 0.00 2.82 2.82 2.82 45.69 45.69 0.00 0.00 40.25 40.25 40.25 -0.00 -0.00 0.00 -0.00 0.00 0.00 -36.08 63.33 1.01 G Lac Assal G1 ~ -19.94 20.00 80.14 20.00 80.14 80.14 -1.20 -1.20 2.81 G G HFO 7 MW ~ ~ HFO 12 MW -19.39 19.94 30.91 30.91 -2.81 4.09 -25.25 -25.25 -25.25 -25.25 Lac Assal G2 -2.40 -2.40 -2.40 -2.40 40.25 40.25 40.25 40.25 -19.94 20.00 80.14 20.00 80.14 80.14 -1.20 -1.20 2.81 Palmeraie 63 G ~ 61.24 25.16 25.16 21.03 21.03 8.60 -19.39 19.94 30.91 30.91 0.97 -2.81 4.09 -47.45 -0.40 -0.40 2.31 2.31 5.78 39.89 39.89 34.16 34.16 Voltage Levels 17.28 45.69 -5.21 2.11 1.13 6.56 400. kV Palmeraie 230. kV 132. kV 17.28 45.69 -5.21 2.11 1.13 6.56 66. kV 63. kV 45. kV 33. kV -41.70 62.65 0.99 20. kV 15. kV 13.8 kV PK-51 63 11. kV 10.5 kV 5.5 kV 3. kV -2.10 -2.10 -1.48 -1.48 6.56 6.56 ALI SABIEH 63 62.13 0.99 -41.97 4.21 2.96 43.44 Ali Sabieh T1 -25.13 -25.13 0.47 0.47 -0.47 -0.47 -21.00 -21.00 -1.63 -1.63 0.99 0.99 -2.23 -2.23 -4.86 -4.86 39.89 39.89 6.52 6.52 6.52 6.52 34.16 34.16 MARABOUT 63 BOULAOS 63 61.16 61.14 0.97 0.97 -4.20 -47.54 -47.53 -2.80 21.03 21.03 21.03 0.00 -13.76 0.00 14.32 14.32 14.32 0.00 43.44 4.15 4.15 4.15 0.00 -11.16 0.00 4.73 4.73 4.73 0.00 61.33 61.33 61.33 0.00 86.24 0.00 43.15 43.15 43.15 0.00 ALI SABIEH 20 19.85 0.99 -43.45 -0.00 13.82 -0.00 4.20 -0.00 12.81 -0.00 2.80 0.00 86.24 0.00 -20.95 -20.95 -20.95 -14.27 -14.27 -14.27 -0.00 -1.85 -1.85 -1.85 5.14 -3.92 -3.92 -3.92 -0.00 61.33 61.33 61.33 0.93 43.15 43.15 43.15 0.00 -77.54 MARABOUT 15 13.82 12.81 MARABOUT 20 ALI SABIEH LOAD BOULAOS 20 88.27 20.00 20.00 14.56 1.00 1.00 0.97 -53.66 -50.45 -17.53 -0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 -5.66 0.00 68.50 -0.00 -0.00 42.80 -37.59 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -2.64 0.00 45.79 -12.40 -3.00 28.65 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 86.58 0.00 G G G ~ ~ ~ G21 G22 G25 SVS -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 5.70 -0.00 SVS G G G G G G 0.00 -0.00 0.00 0.00 0.00 0.00 -0.00 -0.00 3.11 0.00 ~ ~ ~ ~ ~ ~ -0.00 -0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 86.58 0.00 Static Va.. 0.00 0.00 BOULAOS LOAD Static Va.. M1 M2 M3 M4 M5 M6 0.00 0.00 MARABOUT LOAD 5.70 3.11 86.58 G G G G G G G G G G G G ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Load Flow Balanced G32 G31 G11 G12 G13 G14 G15 G16 G17 G18 G23 G24 Nodes Branches Project: Line-Line Voltage, Magnitude [kV] Active Power [MW] Graphic: Djibouti Transmis Voltage, Magnitude [p.u.] Reactive Power [Mvar] Date: 7/28/2009 Voltage, Angle [deg] Loading [%] PowerFactory 14.0.508 Annex: Figure G.8 - 2035 Dry Season Load Flow DIgSILENT -24.89 -16.94 17.47 17.47 -0.64 0.65 PK12 20 -18.65 -12.39 -12.56 13.20 13.20 -2.37 -0.63 -6.60 -0.63 2.40 1.00 8.63 0.63 8.63 6.60 8.75 0.64 8.75 20.00 1.00 22.36 91.80 -22.95 -22.95 -22.95 -22.95 0.00 61.69 -6.70 -6.70 -6.70 -6.70 -34.87 14.49 19.67 -0.00 228.04 228.72 0.00 0.00 9.83 -16.90 -16.91 0.99 0.99 61.31 61.31 61.31 61.31 PK-52 230 ADIGALA 230 0.00 9.83 SVS PK12 LOAD 14.49 228.07 0.00 -16.86 0.99 23.01 23.01 23.01 23.01 8.49 8.49 8.49 8.49 61.31 61.31 61.31 61.31 PK12 230 226.09 -16.63 0.98 PK12 63 63.00 1.00 24.46 26.33 8.73 DIRE DAWA 230 -12.50 -12.50 -12.50 -12.50 -47.46 -92.39 40.12 24.46 24.46 24.46 5.86 5.86 5.86 5.86 -16.70 -75.64 8.73 8.73 8.73 35.23 35.23 35.23 35.23 37.83 78.55 40.12 40.12 40.12 63.59 38.34 1.01 47.53 92.72 Lac Assal G3 Lac Assal GTX 3 19.12 86.40 37.83 78.55 -19.94 20.00 80.49 20.00 80.49 80.49 -2.22 -2.22 3.85 G ~ Lac Assal 63 47.53 92.72 19.12 86.40 80.76 88.01 G Lac Assal G1 ~ -19.94 20.00 80.49 20.00 80.49 80.49 -2.22 -2.22 3.85 G G HFO 7 MW ~ ~ HFO 12 MW -28.66 29.91 46.57 46.57 -5.78 9.45 -24.31 -24.31 -24.31 -24.31 Lac Assal G2 -8.32 -8.32 -8.32 -8.32 40.12 40.12 40.12 40.12 -19.94 20.00 80.49 20.00 80.49 80.49 12.92 35.23 -2.22 -2.22 -5.56 3.85 Palmeraie 63 G ~ 62.25 17.26 19.91 19.91 22.90 -28.66 29.91 46.57 46.57 0.99 -5.78 9.45 25.36 1.53 7.41 7.41 15.39 17.26 27.46 34.73 34.73 1.53 12.92 35.23 -5.56 27.46 Voltage Levels 12.92 35.23 -5.56 2.83 1.67 8.74 400. kV Palmeraie 230. kV 132. kV 12.92 35.23 -5.56 2.83 1.67 8.74 66. kV 63. kV 45. kV 33. kV 63.32 29.88 1.01 20. kV 15. kV 13.8 kV PK-51 63 11. kV 10.5 kV 5.5 kV 3. kV -2.80 -2.80 -2.01 -2.01 8.74 8.74 ALI SABIEH 63 62.61 0.99 29.54 5.61 4.02 57.84 Ali Sabieh T1 -17.25 -17.25 3.18 3.18 -3.17 -3.17 -19.88 -19.88 -3.65 -3.65 2.83 2.83 -4.10 -4.10 -10.05 -10.05 27.46 27.46 14.61 14.61 14.61 14.61 34.73 34.73 MARABOUT 63 BOULAOS 63 62.19 62.14 0.99 0.99 -5.60 25.31 25.31 -3.73 18.39 18.39 18.39 -13.25 -13.76 0.00 15.37 15.37 15.37 0.00 57.84 7.44 7.44 7.44 -9.47 -11.21 -0.00 9.43 9.43 9.43 -0.00 55.82 55.82 55.82 90.23 84.96 0.00 50.79 50.79 50.79 0.00 ALI SABIEH 20 19.88 0.99 27.58 13.30 13.82 -0.00 5.60 11.44 12.81 0.00 3.73 90.23 84.96 0.00 -18.32 -18.32 -18.32 -15.30 -15.30 -15.30 -0.00 -5.54 -5.54 -5.54 5.63 -8.31 -8.31 -8.31 0.00 55.82 55.82 55.82 1.02 50.79 50.79 50.79 0.00 0.43 MARABOUT 15 13.30 13.82 11.44 12.81 MARABOUT 20 ALI SABIEH LOAD BOULAOS 20 92.03 88.27 20.00 20.00 14.79 1.00 1.00 0.99 20.09 22.28 55.31 -0.00 -4.09 -4.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 -5.66 0.00 68.80 -0.00 -0.00 45.90 -23.24 -3.91 0.41 0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -2.64 0.00 45.99 -1.29 -3.00 30.73 73.25 53.32 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 86.58 0.00 G G G ~ ~ ~ G21 G22 G25 SVS -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 5.70 -0.00 SVS G G G G G G -0.00 0.00 -0.00 -0.00 0.00 -0.00 0.00 0.00 3.11 0.00 ~ ~ ~ ~ ~ ~ 4.10 4.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 86.58 0.00 Static Va.. 4.40 -0.15 BOULAOS LOAD Static Va.. M1 M2 M3 M4 M5 M6 73.25 53.32 MARABOUT LOAD 4.10 4.10 5.70 4.40 -0.15 3.11 80.22 54.74 86.58 G G G G G G G G G G G G ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Load Flow Balanced G32 G31 G11 G12 G13 G14 G15 G16 G17 G18 G23 G24 Nodes Branches Project: Line-Line Voltage, Magnitude [kV] Active Power [MW] Graphic: Djibouti Transmis Voltage, Magnitude [p.u.] Reactive Power [Mvar] Date: 7/28/2009 Voltage, Angle [deg] Loading [%] PowerFactory 14.0.508 Annex: Figure G.9 - 2035 Wet Season Load Flow DIgSILENT -20.50 -94.07 99.95 70.36 70.36 46.70 76.98 -0.15 0.44 PK12 20 101.59 -19.48 -98.99 -12.48 -93.49 -92.72 71.33 10.91 71.33 95.98 67.52 67.52 93.49 66.47 66.47 46.70 76.98 -3.02 -0.15 3.02 0.15 20.00 1.00 -44.14 98.70 -24.67 -24.67 -24.67 -24.67 -0.00 66.33 -3.03 -3.03 -3.03 -3.03 -54.19 46.70 76.98 -0.15 227.37 223.20 -27.08 -32.63 0.99 0.97 63.75 63.75 63.75 63.75 PK-52 230 ADIGALA 230 46.70 76.98 -0.15 SVS PK12 LOAD 221.47 -34.36 0.96 -46.56 -46.56 -46.56 -46.56 24.74 24.74 24.74 24.74 4.58 4.58 4.58 4.58 4.97 4.97 4.97 4.97 76.98 76.98 76.98 76.98 63.75 63.75 63.75 63.75 PK12 230 230.00 -21.60 1.00 PK12 63 62.33 0.99 -39.80 DIRE DAWA 230 0.00 -11.75 -11.75 -11.75 -11.75 0.00 0.00 33.57 33.57 33.57 33.57 -88.11 4.97 4.97 4.97 4.97 0.00 0.00 7.51 7.51 7.51 7.51 32.88 32.88 32.88 32.88 0.00 0.00 53.67 53.67 53.67 53.67 -28.42 63.39 1.01 -0.00 -0.00 Lac Assal G3 Lac Assal GTX 3 SVS -0.00 -0.00 0.00 0.00 -18.95 19.00 76.23 19.00 76.23 76.23 -1.47 -1.47 2.93 G ~ Static Var System(5) Lac Assal 63 G Lac Assal G1 ~ -18.95 19.00 76.23 19.00 76.23 76.23 -1.47 -1.47 2.93 G G HFO 7 MW ~ ~ HFO 12 MW -33.30 -33.30 -33.30 -33.30 -6.69 -6.69 -6.69 -6.69 -27.30 28.42 44.10 44.10 -4.39 7.63 53.67 53.67 53.67 53.67 Lac Assal G2 -18.95 19.00 76.23 19.00 76.23 76.23 12.11 32.88 -1.47 -1.47 -4.75 2.93 Palmeraie 63 G ~ 61.52 24.70 -27.30 28.42 44.10 44.10 0.98 -4.39 7.63 -41.21 16.60 29.80 29.80 24.46 24.46 1.16 1.16 3.92 3.92 12.11 32.88 -4.75 47.20 47.20 39.86 39.86 Voltage Levels 12.11 32.88 -4.75 3.08 1.88 9.64 400. kV Palmeraie 230. kV 132. kV 12.11 32.88 -4.75 3.08 1.88 9.64 66. kV 63. kV 45. kV 33. kV -36.47 62.76 1.00 20. kV 15. kV 13.8 kV PK-51 63 11. kV 10.5 kV 5.5 kV 3. kV -3.06 -3.06 -2.21 -2.21 9.64 9.64 ALI SABIEH 63 61.98 0.98 -36.84 6.11 4.41 63.85 Ali Sabieh T1 -29.75 -29.75 0.33 0.33 -0.33 -0.33 -24.42 -24.42 -3.19 -3.19 1.17 1.17 -2.42 -2.42 -6.48 -6.48 47.20 47.20 6.97 6.97 6.97 6.97 39.86 39.86 MARABOUT 63 BOULAOS 63 61.42 61.40 0.97 0.97 -6.10 -41.31 -41.30 -4.06 23.21 23.21 23.21 0.00 -10.78 0.00 16.50 16.50 16.50 0.00 63.85 5.19 5.19 5.19 0.00 -11.52 0.00 5.93 5.93 5.93 -0.00 67.76 67.76 67.76 0.00 76.42 0.00 49.97 49.97 49.97 0.00 ALI SABIEH 20 19.62 0.98 -39.02 -0.00 10.82 -0.00 6.10 -0.00 12.81 -0.00 4.06 0.00 76.42 0.00 -23.11 -23.11 -23.11 -16.43 -16.43 -16.43 -0.00 -2.38 -2.38 -2.38 5.16 -4.85 -4.85 -4.85 0.00 67.76 67.76 67.76 0.94 49.97 49.97 49.97 0.00 -71.31 MARABOUT 15 10.82 12.81 MARABOUT 20 ALI SABIEH LOAD BOULAOS 20 78.55 20.00 20.00 14.62 1.00 1.00 0.97 -48.03 -44.64 -11.30 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 -4.67 0.00 74.00 -0.00 -0.00 -0.00 49.30 0.00 0.00 0.00 -0.00 -0.00 -0.00 0.00 -0.00 -0.00 -0.00 -2.78 0.00 49.47 -39.55 -13.96 -3.00 33.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 75.33 0.00 G G G ~ ~ ~ SVS G21 G22 G25 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 -0.00 4.70 -0.00 Static Va.. SVS G G G G G G 0.00 0.00 -0.00 -0.00 -0.00 -0.00 0.00 0.00 3.14 -0.00 ~ ~ ~ ~ ~ ~ -0.00 -0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 75.33 0.00 -0.00 -0.00 Static Va.. M1 M2 M3 M4 M5 M6 0.00 0.00 BOULAOS LOAD MARABOUT LOAD 4.70 3.14 75.33 G G G G G G G G G G G G ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Load Flow Balanced G32 G31 G11 G12 G13 G14 G15 G16 G17 G18 G23 G24 Nodes Branches Project: Line-Line Voltage, Magnitude [kV] Active Power [MW] Graphic: Djibouti Transmis Voltage, Magnitude [p.u.] Reactive Power [Mvar] Date: 7/28/2009 Voltage, Angle [deg] Loading [%] PowerFactory 14.0.508 Annex: APPENDIX H TRANSIENT STABILITY RESULTS Figure H.1 - 2011 Dry Season: Short-circuit Boulaos 63 kV cleared by tripping Boulaos-PK-12 circuit DIgSILENT 20.00 16.00 12.00 8.00 4.00 0.00 -4.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Total Active Power in MW G23: Total Active Power in MW G16: Total Active Power in MW 25.00 20.00 15.00 10.00 5.00 0.00 -5.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Rotor angle with reference to reference machine angle in deg G23: Rotor angle with reference to reference machine angle in deg G16: Rotor angle with reference to reference machine angle in deg 2011 rotor angles Date: 7/27/2009 Annex: /1 Figure H.2 - 2011 Dry Season: Short-circuit PK-12 63 kV cleared by tripping Boulaos-PK-12 circuit DIgSILENT 18.00 15.00 12.00 9.00 6.00 3.00 0.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Total Active Power in MW G23: Total Active Power in MW G16: Total Active Power in MW 25.00 20.00 15.00 10.00 5.00 0.00 -5.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Rotor angle with reference to reference machine angle in deg G23: Rotor angle with reference to reference machine angle in deg G16: Rotor angle with reference to reference machine angle in deg 2011 rotor angles Date: 7/27/2009 Annex: /1 Figure H.3 - 2011 Dry Season: Short-circuit Boulaos 20 kV cleared by tripping Boulaos 63/20 kV tfmr. DIgSILENT 20.00 16.00 12.00 8.00 4.00 0.00 -4.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Total Active Power in MW G23: Total Active Power in MW G16: Total Active Power in MW 100.00 75.00 50.00 25.00 0.00 -25.00 -50.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Rotor angle with reference to reference machine angle in deg G23: Rotor angle with reference to reference machine angle in deg G16: Rotor angle with reference to reference machine angle in deg 2011 rotor angles Date: 7/27/2009 Annex: /1 Figure H.4 - 2011 Wet Season: Short-circuit Boulaos 63 kV cleared by tripping Boulaos-PK-12 circuit DIgSILENT 60.00 40.00 20.00 0.00 -20.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Total Active Power in MW G23: Total Active Power in MW G16: m:Psum:bus1 100.00 50.00 0.00 -50.00 -100.00 -150.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Rotor angle with reference to reference machine angle in deg G23: Rotor angle with reference to reference machine angle in deg sym_5_1: Rotor angle with reference to reference machine angle in deg 2011 rotor angles Date: 7/27/2009 Annex: /1 Figure H.5 - 2011 Wet Season: Short-circuit PK-12 63 kV cleared by tripping Boulaos-PK-12 circuit DIgSILENT 50.00 37.50 25.00 12.50 0.00 -12.50 -25.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Total Active Power in MW G23: Total Active Power in MW G16: m:Psum:bus1 100.00 50.00 0.00 -50.00 -100.00 -150.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Rotor angle with reference to reference machine angle in deg G23: Rotor angle with reference to reference machine angle in deg sym_5_1: Rotor angle with reference to reference machine angle in deg 2011 rotor angles Date: 7/27/2009 Annex: /1 Figure H.6 - 2011 Wet Season: Short-circuit Boulaos 20 kV cleared by tripping Boulaos 63/20 kV tfmr. DIgSILENT 40.00 30.00 20.00 10.00 0.00 -10.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Total Active Power in MW G23: Total Active Power in MW G16: m:Psum:bus1 100.00 50.00 0.00 -50.00 -100.00 -150.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Rotor angle with reference to reference machine angle in deg G23: Rotor angle with reference to reference machine angle in deg sym_5_1: Rotor angle with reference to reference machine angle in deg 2011 rotor angles Date: 7/27/2009 Annex: /1 Figure H.7 - 2035 Dry Season: Short-circuit Boulaos 63 kV cleared by tripping Boulaos-PK-12 circuit DIgSILENT 125.00 100.00 75.00 50.00 25.00 0.00 -25.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Total Active Power in MW G23: Total Active Power in MW Lac Assal G1: Total Active Power in MW HFO 12 MW: Total Active Power in MW 50.00 37.50 25.00 12.50 0.00 -12.50 -25.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Rotor angle with reference to reference machine angle in deg G23: Rotor angle with reference to reference machine angle in deg HFO 12 MW: Rotor angle with reference to reference machine angle in deg Lac Assal G1: Rotor angle with reference to reference machine angle in deg 2011 rotor angles Date: 7/27/2009 Annex: /1 Figure H.8 - 2035 Wet Season: Short-circuit PK-12 63 kV cleared by tripping Boulaos-PK-12 circuit DIgSILENT 125.00 100.00 75.00 50.00 25.00 0.00 -25.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Total Active Power in MW G23: Total Active Power in MW Lac Assal G1: Total Active Power in MW HFO 12 MW: Total Active Power in MW 50.00 37.50 25.00 12.50 0.00 -12.50 -25.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Rotor angle with reference to reference machine angle in deg G23: Rotor angle with reference to reference machine angle in deg HFO 12 MW: Rotor angle with reference to reference machine angle in deg Lac Assal G1: Rotor angle with reference to reference machine angle in deg 2011 rotor angles Date: 7/27/2009 Annex: /1 Figure H.9 - 2035 Dry Season: Short-circuit Boulaos 20 kV cleared by tripping Lac Assal-PK-51 circuit DIgSILENT 150.00 120.00 90.00 60.00 30.00 0.00 -30.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Total Active Power in MW G23: Total Active Power in MW Lac Assal G1: Total Active Power in MW HFO 12 MW: Total Active Power in MW 200.00 100.00 0.00 -100.00 -200.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Rotor angle with reference to reference machine angle in deg G23: Rotor angle with reference to reference machine angle in deg HFO 12 MW: Rotor angle with reference to reference machine angle in deg Lac Assal G1: Rotor angle with reference to reference machine angle in deg 2011 rotor angles Date: 7/27/2009 Annex: /1 Figure H.10 - 2035 Wet Season: Short-circuit Boulaos 63 kV cleared by tripping Boulaos-PK-12 circuit DIgSILENT 50.00 37.50 25.00 12.50 0.00 -12.50 -25.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Total Active Power in MW G23: Total Active Power in MW Lac Assal G1: Total Active Power in MW 100.00 50.00 0.00 -50.00 -100.00 -150.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Rotor angle with reference to reference machine angle in deg G23: Rotor angle with reference to reference machine angle in deg sym_5_1: Rotor angle with reference to reference machine angle in deg Lac Assal G1: Rotor angle with reference to reference machine angle in deg 2011 rotor angles Date: 7/27/2009 Annex: /1 Figure H.11 - 2035 Wet Season: Short-circuit PK-12 63 kV cleared by tripping Boulaos-PK-12 circuit DIgSILENT 50.00 37.50 25.00 12.50 0.00 -12.50 0.00 1.25 2.50 3.75 [s] 5.00 G22: Total Active Power in MW G23: Total Active Power in MW Lac Assal G1: Total Active Power in MW 200.00 100.00 0.00 -100.00 -200.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Rotor angle with reference to reference machine angle in deg G23: Rotor angle with reference to reference machine angle in deg sym_5_1: Rotor angle with reference to reference machine angle in deg Lac Assal G1: Rotor angle with reference to reference machine angle in deg 2011 rotor angles Date: 7/27/2009 Annex: /1 Figure H.12 - 2035 Wet Season: Short-circuit Lac Assal 63 kV cleared by tripping Lac Assal-PK-51 circuit DIgSILENT 40.00 30.00 20.00 10.00 0.00 -10.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Total Active Power in MW G23: Total Active Power in MW Lac Assal G1: Total Active Power in MW 200.00 100.00 0.00 -100.00 -200.00 0.00 1.25 2.50 3.75 [s] 5.00 G22: Rotor angle with reference to reference machine angle in deg G23: Rotor angle with reference to reference machine angle in deg sym_5_1: Rotor angle with reference to reference machine angle in deg Lac Assal G1: Rotor angle with reference to reference machine angle in deg 2011 rotor angles Date: 7/27/2009 Annex: /1 APPENDIX I DISTRIBUTION NETWORK