Joint UNDP/World Bank Energy Sector Management Program Activity Completion Report No. 007/83 Country: sRi L^AN Activity: POWER SYSTEM LOSS REDUCTION STUDY July 1983 Report of the Joint UNDP/Ubdd Bank Energy Sector Management Progran This document has a restricted distribution. Its contents may not be disclosed without authorization fron the Government, the UNDP or the World Bank. Energy Sector Management Program The Joint UNDP/World Bank Energy Sector Management Program is designed to provide a rapid and flexible response to governments who request assistance in implementing the policy, planning and institutional recommendations of the Energy Assessment Reports produced under another Joinr UNDP/World Bank Program, or in carrying out prefeasibility studies for euergy investments identified in these reports. The Energy Sector Management Program can provide the following types of assistance for countries which have had assessments: o assistance to improve a government's ability to manage its energy sector, for example by defining staffing and work programs, evaluating management information needs, identifying sources of public and private finance, develiping a medium-term investment plan; o prefeasibility work on priority investment plans, especially those which will improve the efficiency of energy use, bring about economic fuel substitution, or provide enough affordable energy to rural areas; o specific short-term assistance in institutional and manpower development, both at the sectoral and agency levels. The Progr-.A aims to supplement, advance and strengthen the impact of bilateral or multilateral resources already av.iilable for technical assistance in the energy sector. * ** * ** Funding of the Program The Program is a major international effort and, while the core finance has been provided jointly by the UNDP and the World Bank, important financial contributions to the Program have been made by the Governments of the United Kingdom, the Netherlands, Denmark, Finland, Norway, Sweden, Australia and New Zealand. -w SRI LANKA POWR SYSTEX LOSS REDUCTION STUDY JULY 1983 CURRENCY EQUIVALENTS Currency Unit - Rupee US$ 1.00 - Rs 22.9 UNITS Thousand circular mills (kcmil) - 1.94 sq.mm. 1 mile - 1.6903 kilometer 1 meter (m) - 3.28 feet 1 kilometer (km) a 1,000 meters 1 kilovolt (kV) - 1,000 volts 1 kilowatt (kW) = 1,000 volts 1 megawatt (MW) - 1,000,000 watts 1 kilovolt ampere (kVA) - 1,000 volt amperes 1 megavolt-amp. (IVA) = 1,000 kilovolt-amperes I kilovolt-ampere reactive (kVAR) - 1,000 volt-amperes reactive I mega volt-ampere reactive (KVAR) - 1,000 kilo-volts ampere reactive (kVAR) 1 Gigawatt-hour (GNh) - 1,000,000 kilowatt-hours (kWh) GLOSSARY AND ABBREVIATIONS CEB Ceylon Electricity Board OA Self-cooled transformer rating FA Fan-cooled transformer rating AAC All aluminum conductor ACSR Aluminum conductor, steel-reinforced TABLE OF CONTENTS Page No. FOREWORD AND SUMMARX OF CONCLUSIONS AND RECOMMENDATIONS .......... i Foreyord .......................oooo.................. i Summary of Conclusions............................ ii Summary of Recommendations.........o................................... lii SECTION I - TRANSMISSION AND DISTRIBUTION SYSTEM REVIEW, CIRGUIT ANALYSIS AND LOSS REDUCTION ............... ........ .................... *............ ....... 1 System Losses ........... ... 1 Long Run Marginal 2................................ . 2 System Voltage Levelvo.el.......................... 3 Transmission System........ ... ............. ...... 4 Distribution Circuit Anyiysis 4 Distribution System Rehabilitation Project...e...... 6 Local Authorities and Small Power Entities.......o.. 7 SECTION II - GENERATING PLANT RRHABIITATION.ooI.............oo. 7 The Kelanitissa Steam Plant (KSP)o S P)............ 8 Gas Turbine and Diesel Plant Maintenance.**4,....... 9 LIST OF TABLES IN TEXT 1.1 Component of Losses................................. 2 1.2 Long Run Marginal Costs at Various Points in the System oo....o.... o ...o....o.o......ooo...o. 3 3.1 3stimated Costs and Benefits in Rehabilitation of KSP...SP........................ 8 ANNEXES 1 Terms of Referencef - Distribution System Rehabi3itation Erogram....... o.... o..... ooo....... 10 2 Terms of Reference - Kelanitissa Steam Plant Rehabilitation............................. 16 SRI LANKA UNITED NATIONS DEVELOPMENT PROGRAMME POWER SYSTEM EFFICIENCY STUDY PROJECT IDENTIFICATION REPORT FOREWORD AND bUMM&RY OF CONCLUS'.ONS AND RECOMMENDATIONS Foreword i. Power system efficiency improvement and loss reduction have been pin- pointed as issues in almost all the country en. Uy assessment reports completed under the joint UNDP/World Bank Energy Assessment Program. Sn response to these findings the UNDP agreed to support a preparatory project (INT/82/009) to develop methodology which could be used to identify improve- ments in electric power systems of the countries studied to reduce technical losses and thereby conserve energy. This report is one of the case studies completed under that preparatory project. ii. In Sri Lanka, the principal power entity is the government-owned Ceylon Electricity Board (CEB) wbich supplies consumers in its own service area and sells power in bulk to 218 "local authqrities" and small entities which, in turn, serve customers in their respective service areas. CEB's sales to the local authorities amount to about 25% of its total sales. This report deals in detail with CEB and makes a recommendation for future review of the local authorities. The report is presented in three sections: (a) The Project Identlfication Report (b) The Generation Technical Supplement (c) The Transmission and Distribution Technical Supplement iii. The study recommends the following: (a) A Distribution System Rehabilitation Project which involves short-term improvements to correct the most urgent needs to improve efficiency, and consultant's studies to prepare a longer-term Distribution System Betterment Program. The proposed Phase I investment required would be $24.7 million, and the Phase II Program would require $56.5 million. The savings in distribution system losses would repay the investment cost in about 3 years. About 25% of these costs would be In local currency. These estimates do not make any provision for work in the local authorities' systems. This would add about 35% to the cost estimates. - ii - (b) A Steam Plant Rehabilitation Program to restore the principal thetmal plant to normal efficiency, operating reliability and capacity. The cost of this rehabilitation program would be about $2 million and fuel savings in 1 year would repay the cost. Summary of Conclusions iv. The CEB power system and its administrative organizatiou have suffered considerable deterioration during recent years because of a severe flight of senior and experienced personnel from the entity due to an inade- quate salary scale. The management structure and technical staff in place today has been so depleted that they are unable to handle the problems they face without outside help. CEB has management consultants reviewing the management organization. v. The distribution system and the principal steam plant have deterio- rated to the point where quality of service is poor and severe power shortages are being experienced. The new and planned future additions to generating capacity will alleviate the power shortage in about 2 years' time and last until about 1987-89, depending on future growth in demand. But major rehabilitation of the existing steam plant is required and appropriate mainte- nance programs must be put in place to prevent the new plant from quickly deteriorating. The transmission and distribution networks are in need of major rehabilitation and expansion, both on the CEB system and even more so on the local authorities' systems. Neither the organization nor the trained personnel needed to deal with these problems is available. vi. The principal oil fired steam plant, Kelanitissa (50 MW), has deteriorated so that it has been derated to 35 MW and its heat rate has increased from 11,000 Btu/kWh, which would normally be expected from a plant of this age, to 14,250 Btu/kWh. The direct cost of rehabilitating this plant would be about $1.5 million and the annual savings in fuel cost would be about $2.8 million. vii. The hydro plants, which supply about 80% of CEB's power requirements, are generally satisfactorily maintained. However, due to the annual dry and wet seasonal rainfall cycle, these hydro plants depend on a thermal plant backup to compensate for low water periods. Opportunities appear to exist for cost-effective modification and improvements to the hydro plants, consisting of retrofitted turbine runners, improvements to trash racks and penstocks, and possible civil work modification, such as tail-race realignment. The Phase I Program includes an engineering survey of these prospects. viii. Sri Lanka has been suffering from an unasually severe drought since 1981 which reduced hydro plant reservoirs to such low levels that the thermal plants were unable to supply the deficit in energy and power supply had to be restricted. This situation and the deteriorated physical condition of the generation, transmission and distribution systems created unacceptable low voltage conditions throughout the transmission and distrilution network. The addition of new hydro and thermal capacity now scheduled will alleviate this - iii - situation in about 2 years' time. However, there remains the need to review all system planning criteria to avoid a repetition of this situation. The long-range geueration expaasion program to supply forecast demand about 1990- 1992 has still not been adequately defined, so that less-than-optimum- solutions may have to be adopted. In turn, the uncertainty of the generation program makes it difficult to plan transmission facilities, where needed reinforcements are late in being carried out. ix. The existing transmission system is loaded beyond its capacity in sections and has higher losses than would normally be expected. Total losses on the CEB system in 1982 were about 18.1% of gross generation; the non- technical portion of these losses - those due to billing errors and illegal diversion of current -- were estimated at about 3.5% of total losses. In other words, total system losses could be reduced to about 14.6% by eliminating non-technical losses. If technical losses could be reduced by 30% by 1990, total system losses could be reduced to about 10.2%. This would seem to be an attainable target and would represent an average annual fuel savings of about $11 million from 1985 to 1990. X. Distribution and transmission system computer-based loss analysis has not been done by CIB so far, but an engineer has recently been assigned to this task along with other duties. Proposals for the acquisition of computer programs for loss reduction analysis and system design are being reviewed by CEB. The analysis made by the team on selected test circuits to show the effect of simulating the addition of capacitors gave benefit to cost ratios of 10:1 or more. The average benefit to cost ratio when reconductoring and adding capacitors was 3:1 because of the much higher cost of reconductoring. The use of a computer-based analysis substantially enhances results obtained by using traditional design methods. xi. The Distribution System engineering, planning and construction are separated under several departments and regions so that coordination is lacking. The design criteria needs revision to bring practices into line with today's system requirements and economic conditions. xii. The quickest means of accomplishing the restoration of CEB and initiating the loss reduction activities would be to engage outside consul- tants and contractors for about 3 years to plan, organize and start the activities while, at the same time, training CEB personnel to take over and continue the activities into the future. In parallel with this program to restore the system to adequate operating efficiency, steps should be taken to correct the salary and manpower problems. The government is aware of this and is reviewing its options to find a solution. xiii. Summary of Recommendations a (1) In order to provide coordination for distribution activity and to improve system efficiency in the area where the highest percentage of system losses occur, it is recom- mended that a Distribution Engineering Department be created. - iv - (2) A computer-based distribution circuit analysis program is needed as a permanent, continuing activity to identify distribution system losses and keep them at acceptable levels and optimize circuit performance to meet changing system conditions. In order to ensure a satisfactory computer analysis, it is essential to establish an adequate data base. For example, distribution system maps should be completed and kept up-to-date; substation feeder metering should be improved and data should be recorded and analyzed on a systematic basis. CEB has invited offers from consul- tants to supply computer programs but additions to those programs may be needed to accommodate the rehabilitation program recommended in this report. (3) Transmission and distributiun system design criteria and standards should be revised to reflect current economic costs. (4) Metering, billing and collection practices should be reviewed to ensure that all appropriate measures are being taken to reduce losses and unaccounted-for usage. (5) A Distribution System Long Range Program should be developed under the proposed new Distribution Engineering Department when this is formed. (6) The Kelanitissa Steam Plant should be rehabilitated to improve heat rate, extend the plant's life expectancy and increase firm capacity. (7) A comprehensive maintenance program for all thermal plants should be organized and implemented. (8) The dispatch system for generating facilities should be improved to ensure adequate system voltage control, optimize maintenance and maximize the economic usage of each type of plant to achieve the lowest cost of power generation for the system as a whole. (9) In view ot the technological advances made in hydro turbine design in recent years, it is recommended that CEB should contract a qualified consultant to make a comprehensive audit of the operating efficiency and maintenance proce- dures of its hydro plants to ascertain whether significant improvements in efficiency can be identified and justified. (10) While the local authorities' and other small entities' systems were not examined in detail, it was concluded, from some on-site inspection and discusrions, that losses are probably high and the physical installations have deteriorated. Therefore, a review should be made of this situation to determine the best proce-dure to improve performance on these small systems. This matter is being reviewed by the govermnent. SRI LANKA POWER SYSTEM EFFICIENCY STUDY PROJECT IDENTIFICATION REPORT SECTION I TRANSMISSION AND DISTRIBUTION SYSTEM REVIEW, CIRCUIT ANALYSIS AND LOSS REDUCTION General 1.01 This section of the report reviews the condition and problems of the Transmission and Distribution System, describes briefly the Loss Reduction Study and presents recommendations for the rehabilitation of the CEB system. Further deteil on the transmission and distribution system is given in the Transmission and Distribution System Technical Supplement. System Losses 1.02 Before examining the loss situation, it is important to understand that all information on system performance such as sales, gross generation, substation circuit load data, power factor, etc., -is to varying degrees inac- curate and in a number of instances missing altogether. Some of the problem is due to a lack of instrumentation and in other cases because of poor mainte- nance of equipment and a lack of systematic supervision. Thus, while the conclusions reached in this report are reasonably good indicators of the problems, it was not possible to comp.ete a >recise analysis. 1.03. A review of system losses shows thai: during the 1970's average losses were about 15% of gross generation. It appears that those losses were practically all technical losses. In 1981 and 1982, losses jumped to 18.8% and 18.1 % of gross generation. The only plausible explanation for this sudden increase is the deterioration of the system and a substantial increase in non-technical losses. In view of this background, it is the team's opinion that non-technical losses could be substantially reduced or eliminiated by applying a vigorous campaign to remove the causes of these losses, e.g., billing and meter reading errors, Improperly metered or unmetered customers ard illegal usage. It is also evident, as shown by the circuit analysis, that t,achnical losses can be reduced substantially by rehabilitating the distribu- tion system and improving transmission system design. 1.04 A rough analysis of total system losses gives the following propor- tions, as a percentage of gross generation, for the principal components of losses for the year 1982: 2- TPble 1.1 Percentage of Coar-onent of Losses Gross Generation Transdission system losses 3.4% Substation power transformer losses 4.6% Distribution system technical losses 6.6% Distribution system non-technical losses 3.5% Total system 1 qses, both technical and non-technicall' 18.1% 1/ As a point of reference, it might be interesting to note that in industrialized countries, total power system losses average about 7.5%. 1.05 The conclusions to be drawn from the above are: (a) Transmission system and power transformer losses are consider- ably higher than would be expected for 'he CEB system, indicating that the transmission system design and loss levels should be carefully reviewed. (b) Distribution system losses could be reduced materially. (c) Non-technical losses could be reduced very substantially or practically eliminated. (d) Total system losses in 1982 cost CEB about $41 million based on gas turbine fuel cost, or $38 million if calculated on the long run marginal cost of energy. Long Run Marginal Costs 1.06 The long run marginal cost (LRMC) may be defined as the present value of the economic cost of supplying an incremental unit of demand on the power system. 1.07 For the purposes of valuing technical losses in the electrical net- work, the long run marginal cost is estimated as a two-part cost, an incremental capacity related cost and an incremental energy cost. A 1.08 The peak kilowatt charge is based on the investment and fixed opera- tion and maintenance cost of the incremental capacity related facilities that the system planners would adopt as the least cost means of providing the - 3 - incremental peak demand kilowatt. These facilities are the powerhouse and associated eloctro-mechanical equipment of the Rantembe/Randenigala hydro scheme and the incremental additions to the network included in the least cost investment program from 1983 to 1993. 1.09 The marginal energy cost is derived on the basis of incremental fuel and variable operation and maintenance costs of the generating facilities that would be used to provide the incremental kilowatt-hour. The marginal energy cost is calculated as a weighted average incremental cost of peak and off-peak energy. The generating system simulation indicates that the existing gas turbines would be needed to provide the energy at peak times in the dry season and heavy oil-fired diesels in the wet season. Off-peak energy in the wet and dry seasons would respectively be provided by hydro and steam plants. 1.10 Table 1.2 shows the estimated LRMC's at 11Z discount rate, as the opportunity cost of capital to Sri Lanka, and at the international border price of fuel at Rs 5.76/litre for diesel 2 and Rs 3.88/litre for furnace oil. Table 1.2: Long Run Marginal Costs at Various Points in the System I. Marginal Capacity Cost US $/kW/Year At High Voltage Sub. At Medium Voltage At Low Voltage Generation at Bus (132-66 kV) (33 - 11 kV) (400 V) (Distribution Transfer) (Customrr SVC) 57 93 124 136 II. Marginal Energy Cost US S/WH At High Voltage Sub. At Medium Voltage At Low Voltage Generation at Bus (132 - 66 kV) (33 - 11 kV) (400 V) (Distribution Transfer) (Customer SVC) 0.0550 0.0594 0.0624 0.0633 System Voltage Levels 1.11 The late completion of new hydro and thermal generating plants, the derating of existing thermal plants (Section II), and the unusually severe drought added to the deterioration of the transmission and distribution systems have resulted in unacceptably low voltage conditions throughout the system in varying degrees. The normally accepted voltage variation of *6% has long been exceeded and now is frequently 10% on the low voltage system with areas where, during peak hours, because voltage on the 33 kV system drops to 24 kV, domestic lighting is reduced to a reddish glow. This condition not only gives extremely poor quality service but substantially increases system losses and represents an ecouomic loss to the community. In many hotels and industries, standby diesel units, aggregating 22 MW, have been installed by the owners to assure electrical supply when the CEB system provides inadequate service. Transmission System 1.12 Transmission system losses are higher than expected. The following points were noted by the team: (a) Existing transmission system conductor size has been selected on the basis of using a standard conductor size rather than on the most economic size. Size for line now under construction was selected on the conductor thermal rating. These design criteria should be reviewed since substantial loss reduction could be achieved by accelerating construction of planned transmission lines or reconductoring existing lines where feasible. (b) Transmission system losses could also be reduced by operating lines in parallel instead of the present radial pattern. This also would improve voltage levels. (c) The 66 kV system is old and has high losses in both the lines and transformers. A study should be made to determine the best procedure for upgrading or eliminating this voltage level. 1.13 The Transmission and Distribution Technical Supplement gives details of the above. Distribution Circuit Analysis 1.14 The analysis of distribution circuits was made by the team by using a computer-based program which, on the basis of the physical and electrical characteristics of the circuit, is able to calculate the technical losses and can then calculate the reduction in losses resulting from the addition of capacitors or changing conductor size. It can also give the optimum location of capacitors. 1.15 The limitation to this analysis is the quality of the data base from which the input to the computer is derived. Unfortunately, as already stated, in Sri Lanka the data base is not good. 1.16 In order to obtain a representative cross section of the system, two 11-kV underground cable circuits in Colombo and four 33-kV rural overhead line circuits were chosen as the test circuits. The results showed that losses could be reduced economically by adding capacitors and/or increasing conductor sizes. The average reduction in losses derived from the proposed improvements to the circuits, when translated into fuel savings, gave benefit to cost ratios of 3 to 1, which clearly shows that a loss reduction program is urgently needed and would pay excellent dividends. Where reconductoring was not essential and only capacitors were added, the benefit to cost ratios were - 5 - above 10 to 1, because of the comparatively low cost of capacitors. Computer- based analysis of distribution circuits is not being made by CEB at the present time. However, offers have been invited for the supply of computer equipment, programs and consulting assistance to establish a distribution circuit analysis program. Changes and extensions to the system have been made using traditional design procedures based on past standards and criteria. In general, these results can be substantially enhanced by using a computer-based circuit analysis. It is urgently necessary for CEB to acquire the equipment to institute a distribution system circuit analysis program on a continuing basis but to make such a program fully effective, a parallel project should be organized to bring the data base up to the required level to facilitate mean- ingful analysis of operating data. These recommendations have been included in the Terms of Reference for the distribution consultant (Annex 1). Distribution System Organization 1.17 The operations, maintenance and construction of the Distribution System has recently been divided into two geographic regions. CEB's manage- ment consultants are now proposing that standards and material specifications be transferred to the jurisdiction of the Commercial Department. There are apparently compelling reasons for these changes, however, the team feels that some other important factors are being left unaddressed. 1.18 The first factor is that this fragmentation of control over distri- bution activity must be coordinated at some point in the organization by a group that specializes in and which understands distribution engineering. At one time, distribution was a secondary activity and a relatively simple func- tion compared to generating plant and transmission lines. This is no longer true. Today, the Distribution System, in most cases, accounts for over 50% of total system losses and 25% or more of the power entity's annual capital investment. It is a major activity. In addition, distribution engineering now involves specialized techniques; design standards and criteria have changed radically in response to changed economic conditions. These new factors impact forcefully on the investment, maintenance and operating costs of the system. It is estimated that CEB could easily reduce distribution system losses to save the equivalent of $11 million annually in fuel costs by applying modern techniques and design criteria. 1.19 It is therefore recommended that a Distribution Engineering Depart- ment be created to coordinate distribution activity; establish uniform criteria, standards and procedures for distribution work; utilize modern computer-based analytical techniques; centralize distribution system overall planning and engineering procedures; collect and analyze system information to monitor performance and identify needed changes. Details are given in Annex 1. The revised organizational arrangements should provide for extending these services and design criteria to the local authorities and independent systems. Consideration should also be given to coordinating procurement and stores to ensure that uniform materials and criteria are used throughout the country. - 6- Distribution System Rehabilitation Project 1.20 In order to rehabilitate the Distribution S-stem and implement the various recommendations made in this report, it is proposed that CEB should consider undertaking a two-phase Distribution System Rehabilitation Program described below. 1.21 In order to introduce new techniques, methods and criteria, it is recommended that consultants be engaged to organize and implement the project and train CEB personnel to take over the activity when it has been established. 1.22 The first phase of the Program would be a 3-year Distribution System Rehabilitation Project which would include the following: (a) Organize and place in operation a Distribution Engineering Department which would include the responsibilities detailed In Annex 1, Terms of Reference for the Consultant. (b) Organize and initiate a Distribution Rehabilitation Project to make physical changes and improvements to the Distribution System to correct the most urgent deficiencies, reduce losses and improve service quality as quickly as possible. Where possible and with CEB's approval, parts of the project, such as installing capacitors, should be implemented while the full study of the system is being completed. (c) The making of various studies, such as: -i) Benefits and costs of converting the existing 33-kv overhead distribution system from 3-phase, 3-wire or 4- wire with a common primary and secondary neutral. This would include the use of single-phase transformers and 230/460 volt secondary, a review of existing substation grounding (now Petersen coil), and upgrading switchgear interrupting capacity; (ii) benefits and costs of converting existing 11-kV overhead to 33-kV as part of system rehabilitation: (iii) other detailed studies as given in the Distribution Technical Supplement. 1.23 The estimated work under the Project (Phase I) will consist of engineering the installation of 233 MVAR of capacitors, reconductoring/rebuilding 189 km of 33 and II-kV line, 1,700 km of new line, and miscellaneous materials. 1.24 The second phase will cover the next 2-year period and will be a continuation of Phase I. It would include completion of the loss reduction and rehabilitation project and include the requirements for materials and - 7 - equipment needed for normal expansion to meet forecast demand and such modifi- cations to the transmission system determined by studies made in Phase I. This should then become part of the permanent function of the Distribution Engineering Department's permanent activities. It may be necessary to engage further consulting assistance beyond that suggested for Phase I. That decision would be taken by CEB in due course, after the results of Phase I can be judged. 1.25 As part of Phase II, the estimated requirements are 152 MVAR of capacitors, reconductoring an additional 78 km of line, 1,350 km of new line, 280 MVA of diatribution transformers and 190 MVA of additional asubstation capacity. Local Authorities and Small Power Entities 1.26 In making the Sri Lanka study, the CEB system was studied in detail while the other power entities were only given a relatively brief visual inspection. The complexity and magnitude of CEB's problems as well as the fact that it alone represents about 75% of the power sector, were the reasons for this decision. It was considered more productive to concentrate on the major entity first. 1.27 The brief visit to some of the local authorities resulted in the following observations and recommendations: (a) In general, these smaller entities tend to copy CEB procedures and standards but frequently are less well managed, suffer higher system losses and are in need of more assistance. (b) The basic problem is that the multiplicity of entities, some very small, needs restructuring so that a single entity could direct the engineering, planning and construction for these entities. (c) It would be useful to have a study made to ascertain the condi- tion of the small entities and the approximate cost to rehabilitate their systems. The consultant's Terms of Reference include this as part of the Phase I work. SECTION II GENERATING PLANT REHABILITATIION Overview 2.01 This section of the report reviews the condition of the generating plants and presents a rehabilitation project for the Kelanitissa Steam Plant and comments on other matters associated with the generation system. - 8 - 2.02 In general, the hydro plants are satisfactorily maintained. On the other hand, the principal steam plant needs major rehabilitation, and mainte- nance procedures need improvement to avoid future problems. System operations and planning should be reviewed to avoid a further recurrence of present difficulties. Further detail on the generating system is given in the Generating Plant Technical Supplement. The Kelanitissa Steam Plant (1SP) 2.03 The steam plant contains two 25 MW units which are about 20 years old. In 1979, the units had become so deteriorated that they were derated to 40 MW and CEB called in consultants and the manufacturers of the boiler and turbine-generator to recommend a project to restore the plant to acceptable efficiency and reliability. The recommendations have still only been partially carried out and the plant is in even worse condition than it was in 1979. It is now rated at 35 MW capacity. The plant heat rate is about 30% higher than it would be if it were in good working order. 2.04 The rehabilitation of KSP would be very attractive because the needed improvements, if properly handled, could be completed in 12-18 months at a cost of about $1.5 million and would give fuel savings of about $2.8 million annually. In addition, by restoring the plant to rated capacity, the system would gain 10 MW of reliable capacity. 2.05 Table 2.1 shows the estimated costs and. benefits of the principal items involved in the rehabilitation of KSP. These are foreign exchange costs. Local costs would add about 10%. However, it is to be noted that this does not include the cost or benefits which would result from completing the repair work recommended in 1979, which has been budgeted separately by CEB. This work is, however, included in the terms of reference for the consultant recommended for the KSP rehabilitation. Table 2.1 Item Cost S Benefit S/Yr 1. Consulting Services 250,000 2. Contractors 150,000 - 3. Burner Replacement 270,000 200,000 4. Control Rehabilitation 210,000 110,000 5. Chlorination System 60,000 250,000 6. Fuel Oil Analyzer 25,000 - 7. Air Preheater/ I.D. Fan Repair 26,000 66,000 8. Variable Speed Fan Drives 175,000 60,000 9. Valve and Line Replacement 60,000 - 10. Various Improvements 30,000 - 11. Restoration of Unit to Rated Capacity and Efficiency - 2,125,000 12. Contingencies 144,000 - TOTAL 1,400,000 2,811,000 -9- 2.06 The rehabilltation of KSP involves specialized work which is usually performed by the equlpment manufacturer or contractors who are specialized in the work. In CEB'a case, the KSP maintenance and operations crew are not sufficiently experienced to handle this work which should be done under the management of a qualified outside consultant. 2.07 In addition to the work summarized in Table 2.1, the consultant would have to do the following, which have not been included in the costs irn Table 2.1 because they are not directly related to the improvement in efficiency of the units: (a) Prepare instructions and detailed procedures for placing KSP in cold standby during the annual wet season. This is very important because of the need to maintain all thermal plants constantly in good condition to back up the hydro plants. (b) Describe and organize a maintenance program to keep KSP in good condition after it has been rehabilitated. (c) Train CEB personnel in the performance of items (a) and (b). 2.08 Terms of Reference for the consultant are gi',n in Annex 2. Gas Turbine and Diesel Plant Maintenance 2.09 CEB haw 120 MW of gas turbines and 12.5 MW of diesel plant in service. It is also installing 80 MW of new diesel units and plans a further addition of 40 MW of diesel units in 1988. These units all burn diesel oil and all are susceptible to damage from contaminants in the fuel. CEB should immediately purchase a fuel oil analyzer and establish a procedure to make sure that oil quality is within acceptable limits at all times. The manufacturers of generating plants usually stipulate the acceptable levels of various fuel contaminants and recommend corrective action should the limits be exceeded. 2.10 It is also essential to institute proper maintenance procedures to ensure that this equipment is kept in good condition. 2.11 In the case of the gas turbines, which require attention by experienced and trained specialists, CEB should consider giving a maintenance and inspection contract to the manufacturer. This procedure is followed by many power entities and is considered mDre efficient than doing the work with in-house personnel. 2.12 In the case of the diesel units, the manufacturer could also be contracted to supervise scheduled overhaul and maintenance or train CEB staff to do it. In any case, complete manufacturer's manuals should be available and carefully followed. It is false economy to neglect maintenance as should be realized after the KSP experience. - 10 - Annex I Page 1 of 5 ANNEX 1 TERMS OF REFERENCE DISTRIBUTION SYSTEM REHABILITATION PROGRAM CEYLON ELECTRICITY BOARD Foreword 1. This project has as its purpose the rehabilitation of the Distribution System, the reduction of losses in the Distribution and Transmission System, improvements in the overall efficiency of the system, and the establishment of a proposed Department of Distribution Engineering and the training of staff. It is the intent that the work will be organized and managed by consultants. The work is proposed in two phases, Phase I, to last for a 3-year period, to cover initial rehabilitation and loss reduction, department organization and several studies, and Phase II, to cover a 2-year period to complete rehabilitation and loss reduction projects, and preparation of a long-ra4ge distribution and transmission system plan. The following are proposed Terms of Reference and principal activities of the consultants. Phase I 2. The consultant shall: (a) Aid in the establishment of a new Distribution Engineering Department which will be responsible for overadl distribution system planning, design standards and engineering. The Consultant will recommend a new organizational structure for the proposed department for CEB's approval. Preliminary organiza- tion and functions are shown in Attachment 1. In agreement with CEB, the Consultant will provide key personnel where requested to establish the new department and train local counterpart staff to take over the various responsibilities when the new organization is functioning. (b) In parallel with the establishment of the new department, the Consultant should start a preliminary analysis of the distribution system to make a feasibility report for a first- phase distribution system rehabilitation and loss reduction program. This would include a project description, cost estimate in foreign and local currency, construction schedule, benefit-to-cost ratio and justification. The purpose of this project would be to correct the worst deficiencies in voltage level, service quality and loss levels as quickly as possible. The study shall consider both distribution and transmission. - ll - Annex 1 -~~~~~ ~Page 2 of 5 The project would have to be coordinated with and be complimentary to any distribution system expansion planned by CEB. The report should cover a three-year period and be submitted in two stages--a preliminary recommendation in two months and a final report in four months. (c) Where possible, with CEB's approval, some elements of the projects, such as the engineering and purchase of materials for the installation and capacitors on a limited basis could be started before the full study is completed. This would be an important feature of the implementation of the Consultant's recommendations, so that improvements to CEB's organization and system performance would not have to wait a final report but that partial implementation could be initiated as the full scheme was being developed by the Consultant. (d) A second feasibility study shall be made of the distribution systems owned and operated by the local authorities. The study shall review the need for loss reduction and rehabilitation in order to improve voltage levels, reduce losses, and generally improve the efficiency of operation of the local authority systems. The work would include project descriptions, cost estimates in local and foreign currency ratios, schedules and justifications. The report shall be submitted in two steps, the first in preliminary form for review .and comment in three months after reviews are completed. Although this report is to be prepared under the auspices of CEB, the report shall be directed to the Government of Sri Lanka. (e) Implement plan formulated in Phase (b) and (c) above. (i) Note the work shall be coordinated with current CEB projects. Work shall be separated into that which can and should be done by construction contractors and work which must be done by CEB. The work shall take into account the results of studies in (f). (ii) Prepare specifications, including drawings, for work that can be done by contract, and specifications for material to be purchased and quantity estimates. (iii) Assist CEB in issuing requests for proposals, evaluation and award for contractors and materials. (iv) Establish mutually agreeable long-run marginal costs for the economic evaluation of improvements. (v) Provide a Construction Manager and Inspector(s) to coordinate construction and insure that work is built in accordance with proper standards, drawings and specifications. -12- Annex 1 (f) Make studies as follows: (i) Advantages, disadvantages and benefits, if any, of converting the existing 33-kV, 3-wire system with Petersen coil grounding to a grounded neutral system, utilizing a common primary and secondary neutral. The study shall consider the possibility of utilizing 125- to 150-kV BIL equipment using single-phase distribution, and the optimum transformer sizes and secondary distribution circuit length and conductor size for the three-phase and single-phase circuits. (ii) Advantages, disadvantages and benefits of expanding or replacing the existing 11-kV overhead system with a 33-kV system. (iii) Review 66-kV transmission system with respect to losses, age and capacity and whether or not system should be retained or replaced with 132-kV. (iv) Distribution system protection, including use of reclosers, relay modifications, and switchgear replacement, if necessary. (v) Permissible minimum and maximum voltages and regulation under normal and emergency system operations for low and medium voltage customers. The study shall include an analysis of effective energy savings that may be obtained by using a narrow band of voltage regulation, recommended transformer secondary voltage and whether or not distribution transformer taps can be eliminated. The study shall consider the use of heavier line conductors that may be required for circuit loss reduction. (vi) Application guides for economic selection of primary and secondary conductor sizes. (viii) Improvements in existing transmission system operation that will minimize losses and maintain proper voltage levels. (ix) Application guide for transformers, including initial size for given and forecasted loads and optimum loading. (g) Prepare a complete set of distribution system construction standards, both overhead and underground. New standards shall 4 incorporate any recommendations above. New standards shall include the use of current state-of-the-art materials, where economically justified and adaptable to local conditions. Standards shall cover general line construction, transformers, switches, capacitors, regulators, street lighting where installed on CEB's poles, services and meters. Annex 1 - 13 - Page 4 of 5 (h) Provide, through lectures, seminars and on-the-job training, instruction in distribution system design. It is the intent that all CEB engineers, both in the new Distribution Department and in the districts, shall be provided this training. (i) Utilize computer programs and equipment now being purchased by CEB for load flow studies and distribution system analysis. If these are not available or if they are inadequate, the Consultant shall provide recommendations regarding improvements that CEB should make, including additional programs that may be useful. Phase II 3. Consultant shall: (a) Continue with work necessary to complete system rehabilitation and loss reduction. It is the intent that all detail work would be accomplished by CEB and the Consultant shall review proposed work. (b) Continue with training and advice as necessary to the Distribution Engineering Department. (c) Prepare a system-wide expansion and rehabilitation plan as a continuation of the above for the Distribution and Transmission System. The plan shall provide a minimum cost system adequate to handle several stages of load growth, such as a 500-MW, 750 MW and a 1,000-MW system or other mutually agreed values. The plan shall take into account the results and recommendations of special studies in other parts of the Terms of Reference. The plan shall provide an orderly program for expansion and rehabilitation, improvements in distribution system protection, modifications to existing substations and general requirements and tentative locations of future substations. (d) Review and make recommendations for a uniform mapping procedure for transmission and distribution system, including secondary distribution and customer services. Establish procedures for maintaining maps either at the district engineer's office or at CEB's central engineering office. Maps shall be prepared in such a way that they can be used for a computerized circuit analysis and for transformer load management. (e) Review and make recommendations for improvements in transmission and distribution system operating and maintenance methods and safety procedures. (f) Establish a material identification system and suitable computer programs to improve material purchasing, forecast materials, inventory control and material distribution for all distribution Annex 1 - 14 - Page 5 of 5 systems, substation, transmission construction materials and general maintenance materials for distribution, substations, transmission, power plants and offices. (g) Review and make recommendations for space requirements and equipment for stores, operation and maintenance to cover the needs of the next 10 years. Miscellaneous 4. (a) It is the intent of the Terms of Reference for the Consultant to establish a local office duaring the life of the project. Key personnel shall preferably stay with the project during its life. (b) Consultant will submit monthly reports, outline work accomplished, work to be done and problems. (c) Consultant will prepare a schedule of his activities. (d) CEB will make available to the consultant all available data, records and drawings. (e) Consultant will be given access to all CEB facilities subject to safety and operating requirements. (f) CEB will appoint specific individuals to be responsible for liaison between CEB and the Consultant. ATTACHMENT 1 Page 1 of 1 - 15 - ATTACHMENT 1 SRI LANKA PROPOSED DISTRIBUTION ENGINEERING DEPARTMENT The Chief Distribution Engineer 1.0 Responsible for standards, operating procedures, performance, special trojects. To head four departments. (a) Design Engineer Responsible for: o Line Construction Standards, up to 33-kV, and including undergound * Equipment Installation Standards (transformers, capacitors, regulators, reclosers, switches) 3 Service and Meter Standards o Material Specifications (for Distribution System materials) (b) Performance (Ouality Control) Engineer Responsible for: * Analysis of outages and equipment failures to provide recommendations for corrective action where outage or failure rates are excessive • Equipment and Material Testing (C) Procedures Engineer Responsible for preparation of system-wide: e Operating Procedures • Maintenance Procedures and Schedules e Safety Requirements o Construction Procedures o Inventory Control (in conjunction with Commercial Department) (d) Senior Distribution Planning Engineer Responsible for: o Master Distribution System Planning * Economic Studies o Load Characteristic Studies o Distribution System Protection o Computer Studies Related to the Above ANNEX 2 Page 1 of 5 -16 - ANNEX 2 CEYLON ELECTRICITY BOARD KELANITISSA STEAM PLANT REHABILITATION TERMS OF REFERENCE FOR CONSULTING ENGINEERING AND PROJECT MANAGEMENT SERVICES 1. The Ceylon Electricity Board (CEB) is soliciting proposals to organize and implement the rehabilitation of the Kelanitissa Power Station which consists of 2 x 25 MW oil-fired steam units, which are approximately 20 years old, for the purpose 6f restoring them tc their rated output, normal heat rate and acceptable reliability. This work will be referred to as 'The Project." Terms of Reference 2. The following are the terms of reference for the consulting engineer and project manager: (a) Make a field inspection of the Relanitissa Steam Plant and make an offer to undertake the work described in these terms of reference. (b) Prepare engineering designs and job descriptions, where required, for specific tasks. (c) Issue specifications for the supply and installation of equipment, material and systems required for The Project. (d) Assist CEB in the evaluation of tenders as required. (e) Undertake on CEB's behalf the procurement, expediting and scheduling of work incltding the fabrication and timely delivery of the equipment, material and/or systems required to complete the specified tasks. (f) Manage and supervise the installation of all the work elements specified herein. (g) Organize and carryout a training program for CEB personnel at all levels, so that they will be competent to take over the operations and maintenance of the Kelanitissa Steam Plant after it has been rehabilitated and before the consultant leaves the site. This training program should function in parallel with the rehabilitation work so that CEB personnel should be completely familiar with the work being done. ANNEX 2 - 17 - Page 2-of S 3. The following paragraphs 4 to 11 list the work elements to be included in the Engineers scope of services. 4. An earlier rehabilitation report was submitted to CEB in December, 1979 outlining in detail specific rehabilitation work necessary to restore the Kelanitissa units to reasonable working condition. The report described the major repairs which were recommended to achieve an ircreased level of unit reliability. These tasks have been partially completed. The tasks and their status as of April 1983 are as follows: Wbrk Sdidule - 1982 tbit Nmiber Wbrk Mewait 1983 Status Nb. 1 Turbine Rplace last tb rows of IP bIading QGpleted Rqjfr throttle valve Not completed Replace H.P. heater tubes Nbt cpUleted (ineral OaebAl of awdliaries Completed No. 2 Boller Ppafir fumnse rear wall Gleted Replac econmizer coils Copleted Replace ecamlzer inlet header Cmpleted and distribton pipes teplace air prbeater tubes CqLeteI RecndLtion shot cladig plant Nbt completed Getural ovebaul of a£ dliaries Cpaleted No. 2 Turbine Peplace 1 bladis To be done as rered Saer1 ovdiail To be done No. 1 Boiler Replace air preter btbes Partially caopleted (1982) Replace eommaxzer cals Tb be doe Rplace econmdzer inlet header To be done Reonditin shot cleardng plant To be done Gneral ove*baul of adUiaries To be done ANNEX 2 - 18 - 1984 No. 1 Turbine Gemil avezbha lb be doe Boilers Thapection ard rqaixs of dimeys lb be done akmral ovdhai and repairs lb be done uring turbine wto e 5. Part of the Engineer's service shall include an assessment of the status of the current material procurement program for the above, and to reconfirm the need to perform the above tasks, with special emphasis on the No. 2 Turbine L.P. blading replacement requirement. Equipment and materials procurement required to complete the above work shall be included In the scope of services rendered by the Engineer. The referenced report can be examined by the Engineer in the CEB Kelanitissa office upon request. 6. The Engineer shall have the boiler interior inspected to assess the deterioration of the drum, header and tube walls due to corrosion created by the previous lack of proper feedwater chemical control and improper cold storage procedures (December, 1979 report). The Engineer is to recommend corrective action and if approved by CEB, will implement this recommendation. 7. The Engineer shall implement the installation of the deaeration start-up shut-down steam blanketing system as outlined in the December 1979 report. 8. Additional repair and rehabilitation work as currently planned by CEB is also to be included in the Engineer's tender. These work elements include: (a) the repair of the boiler feedwater system to overcome pump -overload problem. (b) other items added by CEB. 9. The Engineer shall include in his proposal the work required to perform the following supplementary serrices in addition to that described in the previous paragraphs. (a) Analyze and implement the work necessary to restore the Kelanitissa units to full generating capacity (25 KW each). (b) Analyze and implement the work necessary to reduce the apparent excessively high plant auxiliary power demand. (c) Analyze and implement the work necessary to reduce the plant heat rate to a more reasonable value. ANNEX 2 - 19 - Page 4 of 5 10. The following work elements are also to be included in the Engineers' proposals: (a) Prepare specifications and analyze costs for the replacement of the Peabody International fuel burners with modern design, high efficiency, low excess air, low nitrogen oxide emission burners. It is assumed that Peabod7 has the original details required to offer these new burner units. The burners are to be furnished with new atomizers and piston operated light oil ignitors. Appropriate service personnel for installation and setting of the burners shall be included in the Peabody contract. New flame scanning monitoring systems with control room alarms shall also be included in the Peabody supply. (b) The Engineer shall assess and then secure the services of an instrument and controls finm required to supply and install equipment, materials and parts necessary to repair, rehabilitate, revamp and/or install new, modern boiler control systems to permit operating these units to the same degree of automatic control as the original installation. The service also includes the recalibration and resetting of all instruments and controls requiring such adjustments. Suitable warranties and guarantees shall be obtained by the Engineer for CB to insure the proper operation and control of these units. This work shall include furnishing a field engineer to supervize the installation and commissioning of these control systems. (c) The Engineer shall analyze and implement all the work necessary to bring the air preheater/induced draft fan systems to an efficient level of operation. It is recorded here that fan drives of both units operate at full amperage at a unit load of 60 to 80%. (d) To prepare specifications and designs, evaluate tenders to assist CEB in the procurement and field manage the installation of variable speed forced draft and induced draft fan drives. This work includes the addition of new or additional switchgear if required and the design of appropriate foundations and controls, also as required. A field service engineer shall be included in the supplier's contract. (e) To inspect the units and implement the replacement or repair of all leaking valves and lines. This task includes the removal of the two boilers' superheater safety valves and their isolating valves. New safety valves shall be installed without isolating valves. In addition, a new demineralizer water supply line to the units shall be furnished. -20 - ANNEX 2 Page 5 of 5 (f) Any other energy loss reduction tasks as identified by the Engineer's field inspections should be tendered to CEB as additional work elements. These tasks should demonstrate that improved operation, efficiency, performance and/or unit reliability will result from the proposed improvements. 11. A benefit-to-cost calculation shall be submitted by the Engineer for each of the work elements specified in paragraphs 9 and 10. The evaluation factors such as fuel cost and operating regimen are furnished by CEB as an attachment to this request for tender. Cost estimates and benefits derived for each task shall be submitted by the Engineer for review and approval by CEB prior to implementing the work. 12. The Engineer shall submit a procedure to show how the additional work items are to be tendered to CEB and how the work is to be performed and managed by the Engineer. 13. The Engineer shall provide as soon as possible after the award of contract, comprehensive procedures for placing the plant in cold standby. These procedures shall include the shutting down of the plant in its entirety and the recommissioning of the plant after a period of six months to possibly several years time. A nitrogen blanketing system shall be included. 14. The Engineer shall include in his tender, the number of head office and field engineering technical staff he proposes to employ for this work. The time required at the plant site to perform inspections and to research records shall be given, including rates and expenses. 15. The Engineer shall include the services of a construction manager who shall manage the work and liaise with CEB. The field labor is to be provided by CEB except for foreign specialized service personnel or special skilled craftmen. The latter shall be furnished by the Engineer as required and shall be for CEB's account. Field supervision, administration and accounting personnel shall be furnished by CEB for the Engineer's ¶anagement. 16. The Engineer shall plan and schedule the work. The schedule shall reflect the normal operating regimeln of the CEB system whereby the thermal unit repair work is performed during the wet season. The tender document shall give a first estimate of the overall time required to complete the specified work. 17. Resumes of the key personnel being offered to perform this work shall be included in the tender document. ! G13NEf ING PLANT TECEaCAL SMPPLEE TO PROJECT IDENTIFICATION REPORT - ii - TABLE OF CONTENTS Page No. SECTION I - INTRODUCTION ................... ....................... 1 SECTION II - GENERATING SYSTEM , ..................................... 1 Generating Capacity and System Demand .................. 1 Generation Expansion Program ...... ..................... 4 Thermal Plant Operations ............................... 5 SECTION III - REHABILITATION PROJECT * ................................ 6 Background ............................ 6 Kelanitissa Steam Plant ................................ 6 Proposed Rehabilitation Project ......................... 6 Terms of Reference . .................................... 7 Summary of Rehabilitation Cost and Benefits ......0...... 7 Cold Standby of KPS Units . 8 The Kelanitissa Gas Turbine Plant ...................... . 9 Pettah Diesel Station ................................ 9 Other Diesel Units ................................ 10 ukuwela Hydroelectric Station .......................... 10 Other Hydroelectric Stations............................ 10 LIST OF TABLES IN TEXT 2.1 Generation and Demand Forecasts ....................................... 2 2.2 List of Standby Generators With Public Sector Corporations and Private Customers ......................... 3 2.3 CEB Generating Capacity, 1983 ........... ..................... 4 2.4 New Generation Expansion . .................................... 4 2.5 Sequence of Plant Operation .................................. 5 3.1 Estimated Cost and Benefit ................................... 8 ANNEXES 1. CEB Generating Capacity and System Demand..................... 11 2. Known Problems at the Kelanitissa Steam Plant ................. 13 - 1 - SRI LANKA CEYLON ELECTRICITY BOARD UNITED NATIONS DEVELOPME1NT PROGRAMME POWER SYSTEM EFFICIENCY STUDY GENERATING PLANT TECHNICAL SUPPLEMENT TO PROJECT IDENTIFICATION REPORT SECTION I INTRODUCTION 1.01 This Technical Supplement to the Project Identification Report reviews the operating efficiency of generating plants in the Ceylon Electricity Board System. It also summarizes the Generating Expansion Plan and comments on certain aspects. A thermal plant loss reduction project is described and a cost estimate and terms of reference for a consultant are included. SECTION II GENERATING SYSTEM Generating Capacity and System Demand 2.01 In reviewing tile generating capacity requirement of the Ceylon Electricity Board (CEB) from 1981 to the present it should be noted that CEB, as a result of an unusual period of drought did not have sufficient generating capacity to supply system demand. toad shedding had to be applied by cutting service and by requesting captive plants to supply their own needs and wher- possible in 1982 sell energy to CEB. Throughout the period system voltage has been abnormally and unacceptably low and will continue to be so until new plant and transmission lines can be placed in service about 1985. 2.02 CEB had a 1982 total maximum demand of 431 MW and a total energy generation of 2,065 GWh. Table 2.1 lists the generation and peak demand for 1981 and 1982 art. the forecasts from 1983 through 1991, based on information supplied by CEB.7 I/ Commercial Division, Forecast By Sales, 19 May 1982. Does not include railway electrification scheme originally scheduled to start 1985. A new forecast is being prepared which will be lower. -2- Table 2.1: Generation and Demand Forecasts Year 1981 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 A E Oamatimn GWh 1,871 1,957 2,065 2,427 2,743 3,007 3,288 3,554 3,840 4,151 4,486 4,848 Peiak Dzd NW 413 432 431 513 579 650 712 751 811 877 948 1,024 2.03 It will be seen from Table 2.1 that two figures are given for 1981, the column marked "A" is the actual demand and energy generated, while the column marked "E" is CEB's estimate of what the demand and energy generated would have been if the system had been able to supply the demand without restriction. 2.04 Annex 1 gives details of projected system generating capacity and system demand; forecast generation by type of plant and the forecast generation by thermal generating plant. The conclusions are summarized below: (a) Available generating capacity plus new capacity coming into service during 1983 and 1984 will not be sufficient to supply forecast peak demand and provide an adequate margin of spare capacity to meet maintenance and forced outage requirements. Consequently system low voltage problems will persist until about 1985 after which sufficient generating capacity should be available to meet forecast system demand and energy requirements until about 1989. New generating capacity for the 1990's is now being studied by CB. (b) The shortage of generating capacity on the CEB system during past years and the present situation described in (a) has led a number of industries and large hotels to install diesel generating plants on their premises as stand-by units to supply their own power needs in the event of a CUB system failure. At the present time there is an aggregate of about 22 MW installed, equivalent to about 5% of total system demand in 1983 (Table 2.2). In addition to this, the Electrical Department of Kandy Municipal Corporation haa two diesel units totaling 1,400 kW which are in regular use and energy is obtained on a net basis from the Steel Corporation. CEB purchased power from five of the organizations listed in Table 2.2. These power sources have higher generating costs than power purchased from CEB, hence they are usually only used when CEB service is not available. During the past two years, CB has encouraged these captive plants to operate so as to alleviate the shortage of generating capacity on its own system and has paid the owners the difference between their generating cost and what it would have cost them to use CEB power if it had been available. This situation is obviously not to CEB's advantage and will cease as soon as new generating capacity on the CEB system is available. -3- Table 2.2: List of Standby Generators With Public Sector Corporations and Private Customers Name of Customr Size of Pint Pim Flo Mtll Ctplex, Trncmaulee 6 x 1,000 kVA MfflnaIa Smnls Co r oration, Pulngds 2 x 750 kVA FertiLier Corporatin, SOPUPSkAwa O MN brbo Gemrators Tyre Corporation, Kelaniya 2 x 250 kVA Ce0lon xob Co., Ltd., alcbo 1 x 1,000 kVA tewr Bros., Ltd., Colcabo 1 x 1,000 kVA Hotel I:er T coti l lTd., Colambo 1 x 840 kVA Eqsly's Ltd., Colcado 2 x 300 kVA 1elos Bai Headqurters, Coloibo I x 750 kVA + I x 250 kVA Battoa NatmdPa1 Erk, Cbldbo 1 x 480 kVA O.T.S., Coloibo 2 x 500 kVA lInba beroi 6tel, Cblaubo 2 x 1,100 kVA (beirg instaUed) Associated Mbterways Group, Kalutara 2 x 550 kVA aiddm Fashion Knitwmar (lanka) Ltd., Negqto 2 x 500 kWA laika Waltlles ltd., Ebanda I x 1,065 kVA + 1 x 295 kVA Neatinal Mllk Board, Anbewla 2 x 250 kVA National Milk Board, Pallakalle 2 x 200 kVA 27,540 WV Aggregate about 22 MW Source: CEB Commercial Division, March 24, 1983. (c) Sri Lanka has a rainy season from June to November followed by a dry season from December to May. The CEB system is predominantly hydro, having 67% hydro capacity in 1983 and is expected to rise to 70% in 1988. The problem of power shortage arises when a dry season is prolonged and hydro energy falls below estimated firm energy generation. In these cases the thermal plant has to supply the deficit. However, if for any reason the thermal generating plant is not fully available, power shortages result, such as is now occurring. -4- Generation Exp!SLdion Program 2.05 CEB's existing generation plant is listed in Table 2.3. Table 2.3: CEB Generating Capacity, 1983 Plot ard Plant Plant T of Gme~ratlm Natiz - NW Capability - Eem*s nyon Sts 1 30 30 25 Nvt season op 'tion Iz~iniyagala 11 0 DIpens on water Wa Wlawe 6 0 reuireimts. New tmppna Hydro 100 100 CXLd Iwmpana IWRy 50 50 1 y Hydro 50 50 Poltiya Hydro 75 75 UUSW,n EIdo 40 38 BonaemHydro 40 40 FaIlnitissa Stm 50 40 7his plant is urgently in rsd of rehabilitation. Pettah Desel 6 4.5 oianikm Diesel 14 8 Kelanitissa Gas Turbi 120 120 2.06 Table 2.4 lists the new generation facilities under construction or planned for service from 1984 to 1990. Table 2.4: New Generation Expansion Plat Ne si and To tal Year Type of Ganertion Caaity - MW MW Pazuks Early 1984 SsRWcmxk Diese 4 x 20 80 late 1984 Vltotia Hydro 3 x 70 210 164 MWddry semn Tate 1984 CayzonHydro, Stge II 1 x 30 30 M?d 1985 Thtmle ghyo, Stage I 2 x 67 134 125 Mf dky sewn Ed 1986 RaiIenie a Hy&dro 2 x 61 122 177 MW dry smeon End 1987 Rantete Hydro 2 x 24.5 50 Ean 1988 Diesel Genration Plant 40 40 Ezd 1989 froadland Hydro 20/30 30 En 1991 Hydro 3 x 80 240 200MWdry season End 1991 Trncamle 1 x 120 120 Tota 1,056 Sourm: Future Ievelopimi Plm, GM, undated, by S. Bomiface, AQA (Generation). 2.07 In addition to the plant shown in Table 2.4, CEB has under considera- tion the construction of a coal plant and several hydro plants for service in the 1990's. The problems associated with the future generation plant program is that the plants scheduled for service during the 1980's may encounter some delays and thermal plants may have to be relied upon to meet system demand. Therefore the maintenance and rehabilitation of these thermal plants is a matter of prime importance (Section III). There is also serious doubt that the coal plant and hydro plants being studied for the 1990's can be completed in time to meet forecast system demand (Section IV). Thermal Plant Operations 2.08 The operation of the thermal plants is governed principally by the wet and dry seasons caused by the monsoon rains. Generally the thermal plants operate during the dry season. Table 2.5 illustrates the pattern of operation for the thermal and hydro plants. This combination of requirements makes it necessary to have a well organized maintenance program to keep thermal plant availability and efficiency high; it also is important to take appropriate measures when thermal plants are in cold standby to prevent deterioration due to the climatic conditions in Sri Lanka. Table 2.5: Sequence of Plant Operation Loading Period Dry Season Wet Season Peak Load Hydro Peak Plants Gas Turbines Large Diesels Intermediate Load Hydro Base Pln7ts Hydro Gas Turbinesal' Base Load Small Hydro Hydro Large Diesels y/ Kelanitissa Steam Planta / 1/ The heat rates of the various types of plant are: gas turbines 14,150 Btu/kWh; steam units after rehabilitation 11,000 Btu/kWh and large diesels 9,500 Btu/kWh. . -6- SECT!ON III REHABILITATION PROJECT Background 3.01 The two matters of principal concern in the generating system are: (i) the urgent need to rehabilitate the Kelanitissa Steam Plant (KSP) which has deteriorated badly, and (ii) the need to establish a well-organized maintenance program at the hydro, steam, diesel and gas turbine plants. These matters are discussed below. Kelanitissa Steam Plant Present Status: 3.02 Kelanitissa is an oil-fired steam plant consisting of two 25 MW units which are about 20 years old. In 1978 the plant was derated to 40 MW because the units could not give their full output. In addition, plant availability has been poor. In 1979 the situation became so bad that a consulting engineering firm was called in by CEB along with the turbine and boiler manufacturers to assess the condition of the units and to propose repairs to restore them to normal operating efficiency. The consultant's report was submitted in December 1979, in which, recommendations were made to replace the turbine blades in the last two stages and generally overhaul the turbine generator equipment. In addition, major repairs to the boiler were recommended. It was estimated that the boilers would have a four to five year life following this rehabilitation. However, the consultant's recommended repair and overhaul program has only been partially carried out to date. A program should be initiated by CEB immediately to complete the work recommended by the consultant and in addition carry out the work now required to properly rehabilitate these units, which are now in worse condition than they were in 1979. Proposed Rehabilitation Project 3.03 The rehabilitation of the KSP units involves a number of tasks which are highly specialized and will have to be done by the manufacturers of equipment or under the supervision of consultants or contractors familiar with the activity. CEB's workforce has been seriously depleted by skilled personnel leaving to go abroad so that it is now essential that it be rebuilt by a process of recruitment and training. The rehabilitation of the KSP would be an excellent opportunity for initiating this badly needed training process by having CEB's workforce work alongside and under the direction of overseas contractors. It is therefore recommended that offers be invited by CEB from qualified consulting engineers and contractors to undertake the rehabilitation of KSP and at the same time train CEB personnel to operate and maintain the plant after it has been rehabilitated. -7 - 3.04 The consultant would be required to make a visual onsite examination of KSP to include the following in his offer: (a) To review the status of the work already performed under the 1979 consultant's report, assess the items not completed and recommend whether these should now be done or whether changes in the original recommendation are warranted. (b) Recommend additional work required to restore RSP to its original capacity and bring the heat rate of the units -o acceptable levels. The present heat rate is about 14,250 Btu/kWh compared to 11,000 Btu/kWh, which might be expected from these uaits. (c) Review the work now being done by CEB and the items recommended in this report listed in Annex 2, to include these in the rehabilitation offer. (d) Present a detailed plan for the inclusion of CEB personnel in the rehabilitation work and the preoaration of a detailed training program for CEB personnel so that they will be fully competent to continue the efficient operation and maintenance of the plant after it has been rehabilitated. (e) Since the KSP units are expected to be in cold standby during the wet season each year (See Cold-Standby of ISP Units), the consultant will supply detailed procedures for placing these units in cold standby to prevent deterioration due to the effects of the Sri Lanka climate. (f) The consultant w.Lll include in his offer a work plan, schedule, cost estimate and justification of the work elements proposed. Terms of Reference 3.05 Suggested terms of refereuce for the consultant are given in the Project Identification Report. Summary of Rehabilitation Cost and Benefits 3.06 The rehabilitation of KSP should be viewed as being in two parts which, when taken together form the Rehabilitation Project. The first part consists of completing the work recommended in the 1979 consultant's report as modified by the new consultant. This part consists principally in repairs to the plant. The cost is assumed to have been budgeted by CEB so that neither cost nor benefits have been included in the proposed rehabilitation project. The second part would improve plant efficiency above that expected by doing part one and would result in additional benefits. Table 3.1 shows the principal items in part two with the associated cost and benefits. -8- 3.07 Table 3.1 shows that the major benefit is derived fiom the savings in fuel resulting from restoring the units to approach design efficiency levels and rated capacity. The overall benefit-to-cost ratio is about 2:1 in one year's operation. It is estimated that these units would have a useful life of about 10 years if the rehabilitation is properly done. The benefit from extending the useful life of the units has not been quantified because tnis depends upon the future utilization of the plant. However, the unit would be capable of base load service (Table 2.5) during the dry season each year and would be one of the more economical thermal units on the system. 3.08 The consultant's Terms of Reference require him to include two items which do not enhance the efficiency of the plant directly and the cost of these Items have not been included in Table 3.1. They are (i) the preparation and implementation of a maintenance and training program (3.04(d)), and (ii) the preparation of procedures for the cold storage of KSP (3.09). Table 3.1 Item Cost $ Benefit $/Yr 1. Consulting Services 250,000 - 2. Contractors 150,000 - 3. Burner Replacement 270,000 200,000 4. Control Rehabilitation 210,000 110,000 5. Chlorination System 60,000 250,000 6. Fuel Oil Analyzer .',000 - 7. Air Preheater/ I.D. Fan Repair 26,000 66,000 8. Variable Speed Fan Drives 175,000 60,000 9. Valve and Line Replacement 60,000 - 10. Various Improvements 30,000 - 11. Restoration of Unit to Rated Capacity and Efficiency - 2,125,000 12. Contingencies 144,000 - TOTAL 1,400,000 2,811,000 Cold Standby of KSP Units 3.09 Since the KSP units will be used essentially for dry season base load service after 1985 (Annex 1), it will be necessary for the consultant to prepare procedures for cold standby during the vet season to prevent deterioration of the equipment and ensure that it is maintained in good working order. This is especially the case with the boiler and condenser which previously suffered considerable dAmage. 3.10 Areas of special concern are the internal parts of the equipment, systems and materials which use steam, water, air, gas and other liquids. The proper procedures would call for nitrogen blanketing and/or dry storing with -9- heaters or desiccants for the internal pressure parts of the boiler, conden- ser, turbine, generator, switchgear, motors, the feedwater heater cycle (including the bleed steam lines), feedwater and condensate pipe, pumps, controls, etc. The lubricating and fuel oil systems and the water treating plant system also require a lay-up and storage procedure to be sure that corrosion or oxidation of the systems is minimized. Detailed discussions and procedural reviews should be held with the various suppliers on the best means of protecting their equipment and systems. The Kelanitissa Gas Turbine Plant 3.11 Six 20-MW gas turbine units are installed at the Kelanitissa Plant. All are fueled with diesel fuel. Three units were installed in 1980 by John Brown (UK) and three more were installed in 1982 by Alsthom Atlantique (France). The latter are still under guarantee. These units are expected to operate at high-load factors during 1983 and 1984 because of the shortage of thermal generating capacity and the prolonged drought discussed earlier. After 1985, the units are expected to be used mainly for standby and emergency service (Annex 1, Table 3). 3.12 All these units are performing well. However, CEB has no oil analyzer with which to check the level of contaminants in the fuel oil (Annex 2, paragraph 9). This is a matter of importance because gas turbines and diesel units are particularly susceptible to damage by fuel contaminants. A fuel analyzer should be purchased as soon as possible and used regularly to check all fuel used by all thermal plants. The manufacturers of thermal plants stipulate the acceptable level of contaminants and specify corrective measures when these limits are exceeded. 3.13 There is an urgent need to establish a regular maintenance schedule to keep the gas turbine units in good conditicn. The manufacturer's operating and maintenance manuals should be obtained and followed closely. It is especially recommended that CEB review the convenience of contracting with the manufacturer to undertake routine inspection and maintenance of the gas turbines. These units require the attention of specialized personnel who are trained to do this type of work. Many power entities have found it beneficial and economical to do the turbine maintenance under contract with the manufacturer. Pettah Diesel Station 3.14 The Pettah Station is located in the heart of Colombo and has three 30-year old, diesel units which were originally rated at 2 MW each. After being overhauled last year, the units were capable of 1.8 MW output and presently can operate at 1.5 MW each. These units have been scheduled for retirement in 1983; however, in view of the potential shortage of capacity in 1984 and 1985 (Annex 1, Table 1), these units should stay on standby service. Further, the Chunnakem units (not inspected) should be operated during system peak to hetp maintain voltage. The mission can see no justification for trying to improve these units due to the limited future operation of these plants. - 10 - Other Diesel Units 3.15 CEB is installing four 20 NW diesels for service in 1984 and plans 40 MW more in 1988 (Table 2.4). It would be important for CEB to arrange with the manufacturer to set up a maintenance program and train CEB personnel to do this work. It would be beneficial to have an inspector from the factory visit the plant once a year for two or three years to review performance and check the effectiveness of the maintenance work. Ukuwela Hydroelectric Station 3.16 This station has two 20 MW Francis-type units which are in good condition. The principal problem is experienced at the tunnel intake structure at the dam site. The intake is comprised of six sets of steel bars, with a five to six inch bar spacing. A trash rack is installed to remove debris from the face of the bars. During the heavy rains, the debris builds up so fast that it causes a three to four foot drop in water level across the bars. It is then necessary to take the plant out of service for five to ten minutes to clean the bars. This occurs two or three times a day. 3.17 This problem could be eliminated by installing a new trash rake drive unit which would operate the rake at a one-quarter to one-third faster speed thus eliminating the need to remove the plant from service to permit the rake to "catch-up" with the trash removal work. This should be reviewed with the manufacturer and it would probably cost less than $5,000 to install a new drive and appropriate gear box. Other Hydroelectric Stations 3.18 Technological advances made in hydroturbine design in recent years have made it possible to substantially improve efficiency of machines which are 15 years or older by changing runners and other parts. The following plants on the CEB system might be candidates: Old Laxapana, Inginiyagala, Uda Walave, Wimalasurenda and Politiya. However, this type of improvement would require participation by the manufacturer of the equipment who should be consulted. CEB should engage consultants to investigate the possibility of such improvements as part of the Phase I program. Such identification could alleviate the power shortages now contemplated prior to completion of the next major power plant (Trincomalle Coal). - A1 - TABLE 1 CEB GENERATING CAPACITY AND SYSTEM DEMAND Available Capacity MW Firm Capacity i A-/ System Surplus Demand (Deficit) Year Season Hydroa- Thermal Total Hydro./ Thermal Total MW MW 1983 Dry(6)! Wet 383 18 i/ 563 333 120 453 513 (60f/ 1984 Dry 383 220 i/ 603 333 160 493 545 (52) Wet 453 240 693 383 180 523 579 (16) 1985 Dry 547 240 787 465 180 645 613 32 Wet 727 250!/ 927 618 165 783 650 133 1986 Dry 672 250 922 571 185 756 680 76 Wet 727 250 977 618 165 783 712 71 1987 Dry 789 250 1,039 671 .185 855 731 125 Wet 849 250 1,099 722 165 887 751 136 1988 Dry *838 250 1,088 712 185 897 780 117 Wet 898 250 1,148 763 165 928 817 111 1989 Dry 838 290i/ 1,128 712 205 91' 843 74 Wet 898 290 1,188 763 200 963 877 86 1990 Dry 838 290 1,128 712 205 917 911 6 Wet 989 290 1,279 763 200 963 948 15 1991 Dry 1,098 290 1,380 933 205 1,138 984 154 Wet 1,138 290 1,428 967 200 1,167 1,024 143 a/ Future Development Plans, GGII, undated, by S. Boniface, AGM (Generation). b/ Dry season firm capacity is based on 25 MW steam unit (20 MW, 1984 and 1985), one 20 MW gas turbine out of service, and wet season firm capacity based on one large hydro unit, one 25 MW steam unit, and three 20 MW diesel or gas turbines out of service. c/ Decrease by 13 KW if Pettah and Chunnakan Diesel Plants are retained in service. d/ One unit fall, 1983, three units of new diesel plant, 1984. eI Kelanitissa Steam Plant upgrading. fl 40 MW additional diesel capacity. - 12 - TABLE 2 FORECAST GENERATION BY TYPE OF GENERATION Item Season 1983 1984 1985 1986 1987 1988 1989 1990 1991 System Energy Dry 1,189 1,344 1,471 1,610 1,741 1,881 2,033 2,198 2,376 Demand, GWh Wet 1,238 1,399 1,536 1,678 1,813 1,959 2,118 2,288 2,472 Total 2,427 2,743 3,007 3,288 3,554 3,840 4,151 4,486 4,848 Hydro Generation Dry 697 697 1,017 1,150 1,320 1,398 1,400 1,400 1,660 GWh Wet 927 1,372 1,538 1,556 1,790 1,900 1,900 1,900 2,130 Total 1,624 2,069 2,555 2,706 3,110 3,293 3,300 3,300 3,790 Thermal Generation, Dry 492 647 452 460 421 479 633 798 716 GWh Wet 311 27 - 122 23 63 218 388 342 Total 803 674 452 582 444 542 851 1,186 1,058 TABLE 3 FORECAST OF GENERATION BY CATEGORY OF THERMAL GENERATING PLANTS Item Season 1983 1984 1985 1986 1987 1988 1989 1990 1991 Large Diesels, Dry - 219 350 350 350 350 350 525 525 GWh Wet - 27 - 129 23 63 225 388 342 Total - 246 350 479 373 413 575 913 867 Kelanitissa Steam Dry 175 175 102 103 71 75 175 175 175 Plant, GWh Wet 70 - - - - - - - - Total 245 175 102 113 71 75 175 175 175 Kelanitissa Gas Dry 317 253 - - - 54 101 98 16 Turbines, GWh Wet 241 - - Total 558 253 - - - 54 101 98 16 _ ~ ~ ~ ~ _ _* Grand Total Thermal 803 674 452 582 444 542 851 1,186 1,058 Generation, GWh nou ==C m.. mum mum minmmm - 13 - ANNEX 2 KNOWN PROBLEMS AT THE KELANITISSA STEAM PLANT 1. The boiler feed pump drives limit unit output. Both pumps operate at full motor amperage when unit outputs are only at 60% to 80% full load. These drives should be replaced. 2. The boiler feed pumps of both units are believed to be leaking a large flow through the leak-off valve system. Air Preheater - ID Fan Repairs 3. The ID fan drives on both units operate at full rating between 60 and 80% of full load on the units. The ID fan inlet gas temperatures are also well below design values at all loads. Both these readings indicate that there is an excess of air entering the ID fans because of either air heater tube leaks, flue or duct leaks and/or fan casing leaks. Auxiliary Power 4. The KSP auxiliary plant power usage has been in the 10% range for many years. An acceptable auxiliary power rating for an oil-fired steam plant of this type and age would be in the 3 to 4% range. Boiler feed pump and ID fans operation may be a major cause. Oil Burners 5. The existing boiler oil burners are a standard design typical of that used in the latter part of the 1960s. Burners of present day design are high efficiency units built to operate under low excess air ratings with low nitrogen oxide and carbon monoxide emissions and very low carbon losses. These new burners can improve overall plant efficiency and reduce air pollution emissions. The installation of these burners with new oil atomizers, light oil ignitors and flame safeguard systems, can also reduce fuel and auxiliary power losses. Additionally, these new burners should be designed to burn residual oil as well as furnace oil which would give even more flexibility and savings. Control Systems - Instruments 6. The plant inspection revealed that essentially the entire boiler plant operation is on hand control. The only device on automatic control is the boiler pressure control system. Essential controls on hand operation include drum level, feedwater flow, steam temperature, excess air including the FD and ID fan controls and several other less critical systems. The plant people advise that these controls 'never really worked correctly." - 14 - 7. In addition to the efficiency of operation being compromised by thls situation, it is also a highly dangerous mode of operation since it was also found that two of the six boilers' safety valves on each unit have been installed with isolation valves which have been closed for the past 10 to 15 years. This is even more critical as there is no burner flame safety system installed on these units. 8. Generally, the functioning instrumentation is in fair condition. The plant staff advise that most of these devices had been calibrated about a year ago. New water treatment monitoring devices have also been installed per the consultant's strong recommendations in 1979. However, many recorder charts are inoperative and some of these new recorders have no chart paper because they either did not order them with the new devices or they simply are not in stores. Chlorination System 9. As a result of the examination of the condenser and cooling water system in 1979, it was recommended that a chlorination system be installed. It was reported that should the RSP units operate on a regular basis, the chlorination system would prevent the growth of algae. In addition, this system is needed to chemically flush the corrosion pits which formed during the long shut-down periods during the wet season. Fuel Oil Analyzer 10. CEB had advised that the most recent analysis of oil purchased from Ceylon Petroleum Corp. (CPC) is more than a year old. This refers to both diesel and furnace oil. 11. Neither CEB nor CPC have a fuel oil analyzer by which to identify the fuel product being burned in the CEB thermal units. If CEB't only thermal units were the Kelanitissa steam units, this would not be too serious. However, with the forecast operation of the relatively new gas turbines burning diesel oil and the new diesel units operating on furnace oil, it is imperative that an analyzer be available to test these fuels for contaminating metals and compounds such as sodium, vanadium, lead, potassium and sulphur. Valve and Line Rehabilitation Work 12. The inspection of the RSP plant revealed many areas where maintenance work is required on the piping systems. Many of the larger pipe systems have valves which are leaking badly, resulting in excessive energy losses. The repair or replacement of the other valves can greatly reduce the energy losses being experienced at the KSP plant. Plant Safety 13. In addition, it was observed that two of the six safety valves on each boiler had isolating valves installed between the safety valve and the boiler in violation of boiler safety codes. It is Imperative that these isolating valves be removed as soon as possible. - 15 - 14. The treated water make up line between the deuineralizer and the deaerator is now undersized because of the large percentage of loss being experienced due to the leaking valve problem. In order to assure the proper quantity of make up water reaching the units, this line should be increased. . TRANSMISSION AND DISTRIBUTION SYSTEM TECIICAL SUPPLDENT TO PROJECT IDENTIPICATION REPORT . - i - TABLE OF CONTENTS Page No. SECTION I - INTRODUCTION ................. ......................... 1 SECTION II - EXISTING AND FORECAST SYSTEM LOSSES ................... 1 SECTION III - LOSS ANALYSIS .............5..., .............. 5 Areas of Study ......... ................................ 5 Methodology .......................... 5 Methods of Loss Reduction ............................. 6 ............. ........ ....... ....... 6 Cost of Modifications.................. ................ 7 Capacity and Energy Costs ............................00 0 7 Economic Analysis...................................... 8 SECTION IV - DISTRIBUTION STUDY RESULTS ............................. 9 alle ..................................................... 9 Horana Fedr..................................13 SECTION V - TRANSMISSION SYST........ ................................ 17 Trau~smission.... .................. .......... ...... ; 17 Conductor Sie.................17 Operation .............. ............................ 17 Existing 66-kV System .................. .... 17 SECTION VI - TECHNICAL LOSSES ............... .................. 18 SECTION VII - OBSERVATIONS ............. 22 Distribution System, .......... 22 Voltage....... .................... .... ................. 24 Distribution Construction Standards .................... 25 System Protection ............. ......................... 26 Distribution System Engineering. ....................... 26 Staff ................................................. 27 Computers ...........o.................................. 27 Local Authoritieso .... o .................................... 28 SECTION VIII - CONCLUSIONS ............................................... 28 Construction. ......................... ................ . 29 Planning ............................................... 29 Implementation ............. .. 29 Cos ................................ .............................29 - li - TABLE OF CONTENTS Page No. SECTION IX - RECOMMENDATIONS ........................... 32 Phase I ..... . 33 Phase I I ..... 33 LIST O?F TABLES IN TEXT 2.1 Existing and Forecast Energy Requirements, Demands and 3............................................... * 3 2.2 Distribution of CEB System Losses - 1982 ............... 4 3.1 Long Run Marginal Costs ....... ..a. .................... 8 4.1 Galle Substation - Semidale Feeder..................... 10 4.2 Galle Substation - Udugaia Feeder....................... 11 4.3 Padduka Substation - Horana Feeder..................... 12 4.4 Loss Analysis - Colombo Cable.......................... 14 4.5 Loss Analysis - Colombo Substation A, Transformer 12 ... 15 4.6 Loss Analysis - Colombo Substation B, Transfromer 22 ... 16 6.1 Estimated Distribution System Loss Reduction........... 19 6.2 Estimated Distribution System Loss Reduction and Generation P.equirement........... ............... 21 8.1 Projected Costs - Loss Reduction and Rehabilitation, Ceylon Electricity Board............................. 31 ANNEXES 1 Statistical Data, Ceylon Electricity Board.............. 34 2 Summary of Circuits Analyzed........................... 35 3 Adopted Unit Costs .................................... 36 4 Typical Computer Printouts. ....................... 38 5 Long Run Marginal Cost for Valuing Technical Losses in the Electric Power System......................... 45 6 Street Lighting .... ............................. 53 7 Metering ............................ 55 - 1 - SRI LANKA UNITED NATIONS DEVELOPMENT PROGRAME POWER SYSTEM EFFICIENCY STUDY TRANSMISSION AND DISTRIBUTION TECHNICAL SUPPLEMENT TO PROJECT IDENTIFICATON REPORT SECTION I INTRODUCTION 1.01 This report is the Transmission and Distribution Supplement to the Project Identification Report. This Technical Supplement concerns itself primarily with the power and energy losses in the electrical system of the Ceylon Electricity Board. Included, however, are comments and recommendations on other parts of the system based on observations made during the loss investigation. The results of the study indicate a need for system rehabili- tation in addition to reduction in losses. Cost estimates and Terms of Reference for consulting assistance are included in the Identification Report. 1.02 The power sector in Sri Lanka is composed of three major groups, the Ceylon Electricity Board (CEB), the Mahaweli Development Authority and the local authorities. The CEB essentially controls all thermal generation, existing hydroelectric generation, the transmission system and part of the distribution system (Annex 1). The Mahaweli Development Authority is building several multi-purpose dams. When completed, the associated generation facilities and transmission lines will be transferred to the CEB. The local authorities consist of 218 distribution systems owned and operated by various municipal corporations, and urban, township and village councils. SECTION II EXISTING AND FORECAST SYSTEM LOSSES 2.01 An analysis of past and forecast energy that is lost and unaccounted for is shown in Table 2.1. Although not shown in Table 2.1, existing losses for years earlier than 1978 were 14 to 15% each year. For reasons not known, in 1981 and 1982 energy lost and unaccounted for increased to 18% of generation. The increase may be due to underestimating energy use for customers without watthour meters. At one time, some 20,000 customers were without meters. After 1982, the forecast losses are taken from the system -2 load forecast prepared by the Commercial Division, where, apparently based on historical data, the Commercial Department estimated losses as 15% of gross generation. No attempt was made in the forecast to consider the source of the losses and the changes that might occur with the start-up of new hydroelectric generation, the current transmission line construction program, and whether or not losses will be reduced. Actual energy use and demand have been less than forecast. The Commercial Division Manager advised that a new forecast is to be made which will show a lower energy use. The impact of losses and generation is discussed in Section V. 2.02 Table 2.2 presents the distribution within CEB's system of peak demand and energy losses for the year 1982. This is an order of magnitude estimate based on data from a load flow program, distribution loss analysis made as part of this study and data on average transformer losses. Two things of note from this study are the high peak demand loss, which is equal to capacity of five gas turbines and the unaccounted for energy, i.e., losses that can be considered non-technical due to billing and metering errors, no meters, or theft. In 1982, total technical losses on CEB's system were 14.6% of generation. Distribution accounted for 45% of the loss. Data on the local authorities is limited, but losses in their system is conservatively estimated at 46 GWh, which would make overall energy losses for the country 16.7%. 2.03 The cost of the losses is high. Based on the current gas turbine fuel cost, the estimated value of CEB's 1982 energy losses is Rs 778 million (US$ 34 million). The value of the non-technical losses--that is lost revenue -is Rs 120 million.l.! (US$ 5.2 million). 2.04 An alternative cost evaluation using long run marginal costs (LRMC) for capacity and energy is shown in Table 2.1. 1/ Based on Rs 1.66/kWh, 1982 average revenue. Compare these costs with the net projected surplus in 1982 (taken from loan documents for the Sapugeskanda Diesel Plant Project) of Rs 661.3 million and a system fuel cost of Rs 1,820 million. -~~~ ~~~~~~~~~~~~~~~~~~~~~~~~ . t SRI LANKA Table 2,1: Existing and Forecast Energy Requirements, Demands and Losses 1978 1979 1980 1981 1982 1983 1984 1985 1987 1989 1990 Sales, (Gh 1161 1299.6 1393.8 1504.3 1680 Station Servie 9.7 17.5 15.5 17 Total, GWh 1161 1309.3 1411.3 1519.8 1697 2332 2556 3021 3528 3813 Gros (neration 1385 1527 1668.3 1871.6 207f-/ 2427 2743 3007 3554 4151 4486 Inrease, X 14 10 9 12 11 17 13 10 8 8 8 %3ergy Inat ardl Thiacamte1 For, Qi 224 215.7 257.0 351.8 375 364 411 451 533 623 672 As % of Gierat1m 16.2 14.1 15.4 18.8 18.1 1i 15 15 15 15 is Loaa Factor, % 54.8 52.9 51.7 51.2 54.9 54 54 54 54 54 54 MaxInunDmd,NW 291 329 369 413 431 513 580 650 751 877 948 CalLilte Ped Da1 Loss. NW 62 65 73 110 106 102 114 124 14$ 170 188 As % of MNadum Dsaml 21 22 29.1 26 25 20 20 20 20 20 20 Fstinated value b. es2 based on IWX ' Rs mtl]ion 838 809 911 996 1181 1372 1494 a/ Actual to 1981, estimated 1982. b/ Load shed. c/ From 1983 onwards, data taken from load forecast dated Hay 1982 by Commercial Division, except proposed railway electrication load is deleted. d/ Taken as 15% of gross generation in load forecast. e/ Includes 6.9 GWh of purchased power. f/ Capacity charge @ Rs 2833/kW, energy i Rs 1.43/kWh. -4- SRI LANKA Table 2.2: Distribution of CEB System Losses - 1982 Mw GHh System Peak Demand 430.8 Generation (Gross) 2,072 Sales 1,680 Station Service (Estimated) 17 Energy Lost and Unaccounted for 375 … - - - - - - - - - - - - - - - - - _ - - - - - - - - - - - - - - - - - - - _ Transmission System 25.0 70.1 Grid Substation and Power Plant Transformers 23.0 95.9 Primary Distribution Lines Colombo Area 5.3 16.3 Rural Area 21.7 47.5 Distribution Transformers Colombo Area 3.0 12.0 Rural Area 14.0 23.3 Secondary Distribution Lines and Services Colombo Area 8.0 24.5 Rural Area 6.0 13.1 Technical Losses 106.0 302.7 Unaccounted, Non-technical Losses 72.3 TOTAL 375.0 …-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Cost of Losses USS 36.6 million Rs 838.2 million Note: Load Factor: System - 55; Colombo Area - 58; Rural Area - 50. Loss Factor: System - 32; Colombo Area - 34; Rural Area - 25. -5- SECTION III LOSS ANALYSIS 3.01 The purpose of this study is to identify areas where loss reduction might be achieved. The analysis of the complete power system is not contem- plated. Instead, sample distribution circuits that appeared representative, with certain exceptions, were selected as test cases. The anaylsis is limited to technical losses in the primary (MV) or secondary (LV) circuit itself and does not include losses in distribution system transformers, or the losses attributable to theft or billing errors. The corrective action considered to reduce losses in each test case may not be the optimum solution, due to assumptions regarding load characteristics made in each individual circuit and various combinations of capacitor banks and conductor sizes. however, the intent is to demonstrate, through a series of examples, that system losses can be reduced economically. Areas of Study 3.02 Circuits selected for a comprehensive loss analysis were two 11-kV underground fteders within the City of Colombo and four rural 33-kV circuits in the area south of Colombo. The Colombo circuits represent two extremes-- one an urban residential supply ad the other a semi-industrial/commercial load. The rural circuits supply a predominantly residential load with scattered small industrial load. These 33-kV circuits also include 33-11 kV bulk supply points to scattered 11-kV systems; however the 11-kV circuits are not included in the analysis. Two other 33-kV circuits and three 400 volt circuits were reviewed but an extended analysis was not made (Annex 2). Methodology 3.03 The methodology used for making a technical analysis consists of preparing a computer model of the circuits to be analyzed. The computer model requires data on circuit length, type and size of conductors, connected distribution transformer capacity and for each feeder, maximum demand, voltage, load factor and power factor at the substation. The computer allocates total feeder load along the circuit proportional to the connected transformer capacity. The computer can then calculate the existing losses and future losses using forecast load growth. The computer model can be modified to simulate the addition of capacitors, changing conductor sizes, shortening the circuit or changing voltage levels. The computer also calculates the percentage conductor loading and voltage at different points on the feeder since these indicate the limits of circuit capacity. Typical computer printouts may be found in Annex 4. 3.04 Using suitable costs for capacity and energy, the present worth of losses can be calculated for the time period under consideration. When construction costs for each modification are computed, suitable benefit-cost ratios can be determined. -6- Methods of Loss Reduction 3.05 Power and energy losses can be reduced by one or more of the following procedures: (a) change to a larger size conductor (b) install shunt capacitors (c) transfer load to another less-loaded circuit (d) change to a higher circuit voltage (e) balance loads, particularly in 400-volt circuits 3.06 In this study the two principal methods considered were changing conductor size and installing capacitors. Shortening a circuit or transferring a le,ad depends on the physical arrangement of the distribution system and requires a more detailed study than is feasible under the present study. Installation of capacitors on the high voltage distribution circuits usually provides the greatest benefit in most power systems. Alternatives are large banks of capacitors at a substation bus at the low voltage side of the distribution transformers or, for motor loads, small individual capacitors for each motor. For this study 1,170 to 2,750 kVAR capacitor banks were used as the minimum basic size because of the low circuit loading. Other capacitor bank sizes are available with a lower cost per unit as the bank size increases. The local standard conductor sizes of 100 sq.mm. (0.186 in) ACSR (aluminum conductor, steel reinforced) have been used where reconductoring has been considered. In addition, a 170 sq.mm. ACSR conductor was considered. Assumptions 3.07 Circuit lengths were scaled from distribution system maps which were used for this study. The short branch circuits along with the effect of conductor sag were ignored. Circuit electrical character4s. ,cs were obtained from CEB. 3.08 Connected transformer load on the various ci" .tp Jnalyzed was based on data prepared by CEB staff. Much of this data was leld checked. 3.09 Existing circuit load was derived in part from ammeter readings from the substation operation log and in part by special readings made by CEB's staff. 3.10 Load factor and power factor data for individual feeders was limited. Principal metering available in most substations were ammeters and voltmeters, except the Colombo substations which had transformer watthour and power factor meters. Load factor data used is based on that for the overall system and engineering judgment. Power factors computed from data taken from system control center's load flow sheets gave power factors that appear too low. Power factor used in the rural circuits ranged from 86 to 90% at peak -7- due to a high incandescent lighting load. Off-peak circuit loads are basically industrial/commercial with a very poor power factor. A large number of rice mills and saw mills were noted which are normally poor power factor loads. In the future power factors will probably drop unless some corrective action is taken. In Colombo, for example, air conditioning load will become a major load component, and in the rural areas, fluorescent lighting load will replace incandescent. Cost of Yodifications 3.11 Data for estimating the cost for distribution circuit modifications was obtained from CEB, except the cost of conductors and capacitors and accessories which are based on current prices from U.S. manufacturers. These costs include an allowance for shipping and customs. The unit cost data is summarized in Annex 3. 3.12 The method used for neutral grounding of the 33-kV system at the grid substations requires equipment with a 200-kV basic insulation level (BIL). As a result, an expensive capacitor bank and a two-pole structure is required. As an alternative, since the system supplies a number of 33-11 kV bulk supply substations, lower cost capacitors (about 25X less) might be installed on the ll-kV system and may prr,vide greater benefits. These 11-kV systems, however, were not included in this study. Reconductoring costs include an allowance for removing old conductors and some pole replacements or additions. Salvage was not included. Capacity and Energy Costs 3.13 For the purposes of valuing technical losses in the electrical network, the long run marginal cost (LRMC) has been calculated as a two-part cost, an incremental capacity related cost and an incremental energy cost. 3.14 The long run marginal cost may be defined as the economic cost of providing an incremental unit of demand or energy on the power system. 3.15 The peak kilowatt charge is based on the investment and fixed operation and maintenance cost of the incremental capacity related facilities adopted as the least cost means of providing the incremental peak kilowatt demand. For this study, the costs used are those for the powerhouse and associated electro-mechanical equipment of the Rantembe/Randenigala hydro scheme and the incremental additions to the network in the investment program from 1983 to 1990. 3.16 The marginal energy cost is derived on the basis of incremental fuel and variable operation and maintenance costs of the generating facilities that would be used to provide the incremental kilowatt-hour. The marginal energy cost is calculated as a weighted average incramental cost of peak and off-peak energy. The generating system simulation indicates that the existing gas turbines would be needed to provide the energy at peak times in the dry season and heavy oil-fired diesels in the wet season. Off-peak energy in the wet and dry seasons would respectively be provided by hydro and steam plants. -8- 3.17 Table 3.1 shows the estimated LRMC's at 11% discount rate, as the opportunity cost of capital to Sri Lanka, and at the international border price of fuel at Rs 4.76/litre for diesel 2 and Rs 3.88/litre for furnace oil (Annex 5). Table 3.1: Long Run Marginal Costs Capacity Cost erbgy Cost System location Rs/KW/gear(US$/Mt/year) Rs/kNh (US$/kW) mrmratin 1,312 (57) 1.260 (.0550) H48hVoltage 2,193 (93) 1.38 (.0594) Medium Voltae (37-11kV) 2,833 (124) 2.43 (.0624) Low Volta (400 V) 3,112 (136) 1.45 (.0633) Economic Analysis 3.18 The basis of the economic analysis is the trade-off between increased distribution system investment to reduce the level of technical losses and the economic value of the savings in technical losses. 3.19 On the grounds of efficient allocation of scarce economic resources, an incremental demand on the power system must .be valued at the long run marginal cost of supply. The long run marginal cost of supply reflects the economic cost of the incremental facilities required to provide for the incremental demand. 3.20 Since incremental technical losses constitute a component of the incremental demand on the system, incremental technical losses must also be valued at the long run marginal cost of supply. Therefore, peak kilowatt and kilowatt-hour losses are valued at the long run marginal capacity cost and the long run marginal energy cost, respectively. 3.21 To determine the optimum design of the distribution system, the system's improvement program should be continued up to the point where the marginal cost of improvement is equal to the marginal value of a unit of technical loss. The analysis is performed in constant prices of a base year and over the long term, on the assumption of a static distribution system, in order to capture all long run effects. 3.22 For the purposes of identifying the benefits of reducing the level of technical losses in the Sri Lanka distribution system, a benefit to cost ratio test is performed on the basis of six representative circuits that were analyzed. The cost represents the estimated incremental cost of system improvement comprising the installation of capacitors and/or reconductoring certain parts of the distribution system. The benefits are measured in terms of the present value of the reduced technical losses. -9- 3.23 The analysis is carried out in constant 1983 prices and limited to a five year period and a static distribution system. The discount rate used 11%. A longer time period would be preferred, but in the case of Sri Lanka, either circuit changes will occur due to the addition of substations or, in several cases, the circuit requires some type of corrective action within the next several years due to excessive voltage drop. 3.24 The results show substantial benefits by reducing technical losses with an overall benefit to cost ratio of 3 to 1. These results are very conservative. If a longer test period is used that approaches the economic life of the factlities (20 years), even larger conductor sizes or more extensive reconductoring could be justified. SECTION IV DISTRIBUTION STUDY RESULTS Galle Region 4.01 The results of the analysis are summarized in the following tables and discussion. The base case referred to in the tables represents the losses that would occur if no corrective action is taken to reduce losses and is used to compare the benefits obtained by reducing losses. 4.02 Galle (Bataduwa) Substation supplies the southern tip of Sri Lanka with three 33-kV feeders and, through step down transformers, Galle Munici- pality at 11 kV. The three 33-kV circuits, Semidale, Ambalangoda and Udugama, were considered for analysis. 4.03 The Ambalangoda circuit has the heaviest load. A preliminary circuit analysis was made (Annex 2) which indicated a circuit voltage of 24 kV at the far end, which was confirmed by prior measurtients of 24 kV made by the CEB staff. Present voltage levels are unacceptable. This circuit will be shortened when the new 132-33 kV Mataguma Grid Substation, now under construc- tion, is completed, which should alleviate the low voltage problem. A compre- hensive analysis was not made due to lack of data on how the circuit would be changed after the new station is completed. 4.04 The analysis of the Udugama feeder, a lightly loaded circuit, is summarized in Table 4.2. The analysis indicates a potential saving resulting from the application of capacitors. - 10 - Table 4.1: Galle Substation - Semidale Feeder Present Worth Case 1982 1985 1989 "000" Rupees_/ 1. Base Case - Existinga/ Demand, kVA 8,957 10,784 13,181 Demand, kW 7,723 9,274 11,336 Energy, Million kWh 33.2 39.8 48.6 Losses, kW 1,088 1,595 2,438 22,789 Losses, Million kWh 2.3 3.4 5.2 24,688 Total Present Worth, Losses 47,477 Minimum Voltage Level, kV 24.7 23.0 20.5 2. Installing Two 2,290-kVAR Capacitor Banks Demand, kVA 8,611 10,323 Demand, kW 8,602 10,228 Energy, Million kWh 36.9 43.9 Losses, kW 922 1,330 12,770 Losses, Million kWh 2.0 2.9 13,833 Present Worth, Losses 26,603 Present Worth, Savings 20,874 Investment in Capacitors 847 Benefit-To-Cost-Ratio 25:1 Minimum Voltage Level, kV 28.2 26.6 3. Reconductor 12.9 Miles of Line Demand, kVA 10,544 12,552 Demand, kW 9,068 10,794 Energy, Million kWh 38.9 46.3 Losses, kW 968 1,410 13,506 Losses, Million kWh 2.1 3.0 14,632 Present Worth, Losses 28,138 Present Worth, Savings 19,339 Investment in Conductor 4,672 Benefit-to-Cost-Ratio 4:1 Minimum Voltage Level 26.4 24.7 4. Reconductor as Above, Plus Two 2,290 kVAR Capacitor Banks Demand, kVA 8,714 10,340 Demand, kW 8,707 10,240 Energy, Million kWh 37.4 43.9 Losses, kW 607 885 8,382 Losses, Million kWh 1.3 1.9 9,081 Present Worth, Losses 17,463 Present Worth, Savings 30,014 Investment in Conductor and Capacitors 5,519 Benefit-to-Cost-Ratio 5.4:1 Minimum Voltage Level 29.3 28.1 a/ Assumptions: (1) load Factor - 49%; (2) Loss Factor - 24.5%; (3) Base Case Power Factor - 86%; (4) Load Growth - 5% p.a. (initial). b/ For the period 1985-1989. Table 4.2: Galle Substation - Udugama Feeder Present Wort, Case 1982 1985 1989 "000" Rupees_/ 1. Base Case - Existlnga/b/ Demand, kVA 3,442 4,005 4669 Demand, kW 2,965 3,448 4,017 Energy, Mllion kWh 12.7 14.8 17.2 Losses, kW 79 107 146 1,446 Losses, Million kWh 0.17 0.23 0.31 1,566 Total Present Worth, Losses 3,012 Minimum Voltage Level, kV 30.5 30.2 29.8 2. Installing one 1,380 kVAR Capacitor Bank Demand, kVA 3,473 4,088 Demand, kN 3,416 3,974 Energy, Million kWh 14.7 17.1 Losses, kW 75 102 1,011 Losses, Million kWh 0.16 0.22 1,095 Present Worth, Losses 2,106 Present Worth, Savings 906 Investment in Capacitors 316 Benefit to Cost Ratio 2.9:1 Minimum Voltage Level, kV 32.6 32.2 a/ (1) Load Factor - 49%; (2) Loss Factor - 24.5%; (3) Base Case Power Factor - 86%; (4) Load Growth - 5% per annum (initial). b/ Over the period of 1985 to 1989 - 12 - Table 4.3: Padduka Substation - Horana Feeder Present Worth, Case 1982 1985 1989 "000 Rupees-/ 1. Base Case Existinga/ Demand, kVA 4,651 5,571 6,758 Demand, kW 4,100 4,909 5,952 Energy, Million kWh 18.3 22 26 Losses, kW 106 153 226 2,147 Losses, Million kWh 4.7 0.36 0.53 2,563 Total Present Worth, Losses Minimum Voltage Level, kV 31.6 31.4 31.0 2. Installing 2,290 kVAR Capacitor Banks Demand, kVA 4,888 5,969 Demand, kW 4,878 5,904 Energy, Million kWh 21.8 26.4 Losses, kW 121 177 1,694 Losses, Million kWh 0.29 0.42 2,023 Present Worth, Losses 3,717 Present Worth, Savings 993 Investment in Capacitors 424 Benefit-to-Cost Ratio 2.3:1 Minimum Voltage Level, kV 31.8 31.5 a/ Assumptions: (1) Load Factor - 51%; (2) Loss Factor - 27%; (3) Base Case Power Factor - 88%; (4) Load Growth - 6% p.a. (initial). b/ Over the period of 1985 to 1989. - 13 - 4.05 A more detailed analysis of the Semidale circuit was made in which several alternates were considered. The circuit is long and it is understood to supply the Hambantota area on the eastern end except when the Uda Walawe Hydroelectric Plant is in operation. 4.06. The base case analysis for the Semidale feeder indicates that without some corrective action, voltage levels are completely unacceptable after 1984. The base case analysis also indicates that 64% of the circuit losses occur in the 13 Ipile (21 km) section between the Galle Substation bus and the Semidale Gantry1r. The loss analysis by reconductoring was based on the use of 170 sq.mm. ASCR. Although this may seem to be a large conductor size by current CEB standards, this size and larger has been used in other power systems. Note that the benefit to cost ratio is quite large and a substantial amount of money can be spent strengthening the existing tower line. Calculat- ing the optimum conductor size and the cost of reconductoring an existing circuit requires more study than can be given here. Strength of the pole or tower line must be checked and the impact of outages on customars considered; however, the saving in losses is large enough to warrant a detailed investi- gation. 4.07 In the analysis, the busbar voltage at Galle was set at 34 kV in order to improve the end of the line voltage. However, with all combinations of conductor size and/or capacitors, the calculated voltage level remained low. Some form of load relief must be obtained for this circuit or a feeder voltage regulator installed. A suggested location for a voltage regulator is at the Tangelle Gantry, since at this point, the regulator could correct voltage on Hambanto end irrespective of whether the supply is from Galle or Uda Walawe. Horana Feeder 4.08 This circuit is a 33-kV line running south from 66-33-kV Padukka Substation. The circuit is relatively short and circuit voltage is not at present a problem. Losses are also relatively low, however, as can be seen in Table 4.4, they can be reduced economically with the addition of one 2,290 kVAR capacitor bank. Colombo 4.09 The two circuits analyzed in Colombo are underground with a primary voltage of 11 kV. Both circuits have two things in common: the first segment from Substation A or B to the first distribution transformer No. 12 or 22 in the ring main is a major contributor to circuit losses and the other, segments of 0.04 sq.in. cable contributed to the remaining losses (Table 4.4). Compared to the 33-kV system, the losses are relatively low. 1/ A steel switching structure. - 14 - 4.10 Methods of reducing circuit losses in an underground system are the same as in an overhead system; bowever, the cost is higher. For the appli- cation of capacitors, either metal-enclosed 11-kV capacitors can be installed, tapped from the underground system (the appearance would be similar to a small transformer) or 400 volt units installed at each transformer. The 400 volt units are quite expensive per kVAR. Their use may be justified when reduction in transformer losses are considered. For this study, 11-kV units are used. Table 4.4: Loss Analysis - Colombo Cable Transformer Number 22 12 Supplied from Grid Substation A B Total Losses - kW 37.2 67.5 First Segment, 0.1 sq.in. - kW 13.4 36.7 Length of First Segment - ft. 4,400 5,400 Small Cable, 0.04 sq.1n. - kW 22.0 26.2 Length - ft. 4,700 9,600 Total Circuit Length - ft. 25,090 23,900 4.11 The results for these circuits are shown in Tables 4.5 and 4.6. Although the losses are far less than those in the 33 kV system, they can be reduced. 4.12 A loss analysis on one 400-volt circuit supplied from a distribution transformer in Colombo (Substation 248) was made. Even on a 400 volt circuit, losses can be reduced. Using 1982 loads, 10 kVAR of capacitors reduces losses from 10.1 kW to 9.4 kW and improves voltage. This indicates what might happen if low voltage capacitors are used to reduce losses. A comprehensive analysis was not made. - 15 - Table 4.5: Loss Analysis - Colombo Substation A, Transformer 12 Present WortF Case 1982 1985 1989 "000" Rupees-/ 1. Base Case - Existinga/ Demand, kVA 3,074 3,672 4,523 Demand, kW 2,615 3,122 3,845 Energy, Million kWh 13.0 15.6 18.3 Losses, kW 33 47 71 664 Losses, Million kWh 0.10 0.14 0.21 999 Total Present Worth, Losses 1,663 Minimum Voltage Level, kV 10.7 10.7 10.6 2. Installing 1,170 kVAR Capacitor Banks Demand, kVA 3,204 4,107 Demand, kW 3,114 3,833 Energy, Million kWh 15.5 19.1 Losses, kW 38 58 543 Losses, Million kWh 0.11 0.17 817 Present Worth, Losses 1,T60 Present Worth, Savings 303 Investment in Capacitors 237 Benefit-to-Cost Ratio 1.3:1 Minimum Voltage Level, kV 10.8 10.7 a/ Assumptions: (1) Load Factor - 57%; (2) Loss Factor - 34%; (3) Base Case Power Factor a 85%; (4) Load Growth - 6* p.a. (initial). b/ Over the period of 1985 to 1989. - 16 - Table 4.6: Loss Analysis - Colombo Substation B, Transformer 22 Present Wort t Case 1982 1985 1989 '000" Rupees-_ 1. Base Case - Existinga/ Demand, kVA 4,484 5,182 6,209 Demand, kV 3,570 4,146 4,967 Energy, Million kWh 18 21 25 Losses, kW. 68 91 132 1,272 Losses, Million kWh 0.21 0.28 0.40 1,968 Total Present Worth, Losses 3,240 Mlnimum Voltage Level, kV 10.5 10.4 10.3 2. Installing 1,570 kVAR Capacitor Banks Demand, kVA 4,411 5,400 Demand, kW 4,123 4,935 Energy, Million kWh 21 25 Losses, kW 68 100 960 Losses, Million kWh 0.2i 0.31 1,485 Present Worth, Losses 2,445 Present Worth, Savings 795 Investment in Capacitors 260 Benefit-to-Cost Ratio 3:1 Minimum Voltage Level, kV 10.6 10.5 a/ Assumptions: (1) Load Factor - 58%; (2) Loss Factor - 35%; (3) Base Case Power Factor = 80%; (4) Load Growth 7% p.a. (initial). bf For the period 1985-1989. - 17 - SECTION V TRANSMISSION SYSTEM Transmission 5.01 Although this report is concerned more with the CEB distribution system, a brief review has been made of the transmission system. Based on the examination of system losses in Table 2.2, the transmission lines alone have a current peak demand loss of 25 MW. This will increase as system demand increases and as line extensions are made, although the parallel 132 kV circuit to Chunnakam, now under construction, will make a small redtction. Conductor Size 5.02 As far as can be determined, los1 evaluation has not been a factor in selecting transmission conductor size. A prime example is in the study for the transmission system for Mahaweli 220-kV and 132-kV lines where conductor size was selected on a thermal loading only. Operation 5.03 The method of system operation also will affect system losses. Most circuits are operated on a radial basis. For example, two double circuit 132- kV lines, from Polpitiya to Anuradhapur and from Laxapana to Rurunegale operate with one circuit only. The other circuit is open. Operating the two Anuradhapura cIrcuits in parallel would have reduced circuit losses from an estimated 4.4j/ megawatts to 2.2 megawatts and with an energy saving of 7.1 GWh (1982). The savings in fuel cost alone is Rs 10 million (US$ 440,000). The reason for this method of operation is not clear, whether from inadequate protective relays or switchgear with a low interrupting rating; however, the savings in losses warrant changing system operation. Operating the Anuradhapura line with two circuits in parallel would also improve Chunnakam bus voltage, currently as low as 104 kV. All transmission lines should be operated with circuits parallel to minimize circuit losses. A comprehensive load flow analysis and fault study can provide detailed data. Existing 66 kV System 5.04 One area of relatively high transmission losses is the 66-kV system which accounts for about 20% of the transmission system losses. Conductor size is small and the system has low voltage. In addition, the system is old and many t!nsformers appear to be the original ones installed when the system was built./!. The existing substation transformers will have high losses 1/ Source: Load flow study by CEB for year 1983. System demand 438.9 MW at 93% power factor. Power factor appears high. Load flow study does not include transformer losses. 2/ The 66-, 33- and 11-kV bulk oil switchgear is essentially obsolete. - 18 - compared to current designs. Demand in the 66-kV system is approaching the 66-33-kV transformer capacity, although this will be reinforced as part of the Mahaweli Project. In view of limited capacity and high losses, a study of the advantages that might be obtained by either replacing the 66-kV system with 132-kV, cutting back the system, or possibly abandoning parts of the system is warranted. 5.05 Installation of reactive power compensation at various substation buses for voltage Improvement is recommended in the Revised Project Report for Mahaweli 220-kV and 132-kV Transmission Project. Part of this compensation can and should be in the form of capacitor banks In the distribution system rather than at the substation. 5.06 Delay in construction of transmission system and related substation has contributed to system losses. As an example, timely construction of Mataguma Substation would have reduced losses in the 33-kV Amblandgoda Feeder. SECTION VI TECHNICAL LOSSES 6.01 Technical losses can and should be reduced. Two different approaches are shown here on the impact of loss reduction. In both of these, the base case is the present Commercial Department forecast which assumes that energy lost and unaccounted for will be a constant 15% of generation. Table 6.1 shows a conservative estimate of loss reduction through the application of capacitors and some reconductoring. The estimate is based on an extrapolation from data in Chapter 4 where approximately 6% of the 33 kV overhead and 11 kV underground systems were given a complete analysis. The data in Table 6.1 does not include the reduction in losses that will occur in grid substation transformers and the transmission system that result from the reduction in the distribution system. LO the system expande and heavier conductors are used in the future as the result of suitable economic analysis, a further reduction will occur. 6.02 The other approach to loss reduction is based on establishing a target to be met to reduce amount of energy lost and unaccounted for. Loss levels as low as 5% have been achieved economically in some systems. For CEB's system, a target of 10% energy lost and unaccounted for is proposed by 1990. This is a relatively modest target. The reduction would be the result of reconductoring high loss sections of the 33- and 11-kV systems, possible replacement of the II-kV systems with 33-kV, the application of larger conductors for new construction in the transmission, the 33-kV, 11-kV and 400 volt distribution systems, application of shunt capacitors, eliminating high loss power and distribution transformers, controlling excess transformer capacity and Improving system operation by operating transmission lines in parallel, and controlling reactive power flow in the system. The new control center should be set up for economic dispatch of generation. Maintaining a very narrow band of voltage generation will also contribute. Table 6.1: Estimated Distribution System Loss Reduction 1983 1984 1985 1986 1987 1988 1989 1990 Without Loss Reduction Demand, MW 513 579 650 712 751 811 877 948 Peak Loss, MW 102 114 124 133 148 156 170 188 Generation, GWh 2,427 2,743 3,007 3,288 3,554 3,840 4,151 4,486 Energy Lost and Unaccounted for, GWh 364 411 451 493 533 576 623 672 As a % of Generation 15 15 15 15 15 15 15 15 After Loss Reduction Loss Reduction, MW per year 3.7 9.0 14.4 19.7 22.6 24.0 Loss Reduction, GWh per year 10.1 24.1 38.8 53.6 61.3 64.7 New Demand, MW 646 703 737 791 861 924 New Generation Requirement, GWh 2,997 3,256 3,515 3,786 4,090 4,422 Energy Lost and Unaccounted for, GWh 441 469 494 522 562 608 As a X of New Generation Requirement 14.7 14.4 14.1 13.8 13.7 13.5 Capacitors, MVAR per year 60 84 89 89 63 54 Reconductor, miles per year 63 63 63 58 20 - Investment, Rs Million per year 26.7 32.6 34.2 34.5 15.3 9.3 Benefits, Rs Million per year 24.9 60.1 81.3 132.9 151.8 160.5 Present Worth Investment, 1985-1990, Rs Million 124.0 Present Worth Benefits, 1985-1990, Rs Million 437.0 Benefit to Cost Ratio 3.5 - 20 - 6.03 Table 6.2 shows the result of this approach. Three significant items are shown in the table--the savings in losses range from Rs 21 million in 1985 to Rs 431 million in 1990; the reduction in required generation which will reduce the need for Kelanitissa Steam Plant operation and reduces the gas turbine operation to emergency and most important, the reduction in demand could theoretically eliminate the need for an additional diesel generating plant In 1988. In practice, what will happen is a significant improvement in service quality as a result of voltage increases. Peak system demand could actually increase unless steps are taken to reduce voltage to former levels, which would be undesirable, However, once voltage is brought to acceptable levels, further reduction in losses will result in lower system demand. Consequently, the reduction in losses may permit rearranging the generation exptnsion program. 6.04 Note that an aggressive program in loss reduction in the distribution systems owned by the local authorities could also reduce the need for new generation facilities. 6.05 Table 6.2 is also based on the assumption that energy lost and unaccounted for will not be reduced significantly in 1983 and 1984 from the 1981 and 1982 18% levels. Some reduction may occur in 1984 due to new transmission lines. Not until 1985, assuming the recommendations made in this study, will losses start to drop. The result will be to increase system demand in 1983 atd 1984. 6.06 No attempt has been made to segregate technical and nontechnical losses in the tables. Reduction in nontechnical losses should be made; how- ever this reduction would show as an increase in revenues rather than as a reduction in true economic losses. 6.07 Maintaining losses at a minimum acceptable level is a continuing operation. An economic evaluation of losses should be made a part of all new projects whether line extensions or the purchase of transformers. It also includes periodic re-evaluation of the existing system at all voltage levels-- transmission and distribution, overhead and underground--and taking corrective action where economically justified to reduce losses. Transformer application is particularly important. Under-utilized transformers contributed excess no- load losses and investment, but overloaded transformers are even less desirable. Transformer load management programs are available to monitor transformer loading and to optimize loading. Table 6.2: Estimated Distribution System Loss Reduction and Generation Requirement 1983 1984 1985 1986 1987 1988 1989 1990 Forecast Sales, GWh 2,323 2,556 2,795 3,021 3,264 3,528 3,813 Existing Forecast, Generation, Losses and Demand Required Generation, GWh 2,427 2,743 3,007 3,288 3,554 3,840 4,151 4,486 Energy Lost and Unaccounted for, GWh 420 451 493 *533 576 623 673 As a % of Generation 15 15 15 15 15 15 15 15 Demand, MW 513 579 650 712 751 811 877 948 Peak Loss Demand, MW 102 114 124 136 148 159 170 180 Revised Forecast - Assuming Loss Reduction After 1985 Required Generation, GWh 2,263 2,323 3,000 3,216 3,472 3,709 3,964 4,237 Energy Lost and Unaccounted for, GWh 407 477 440 421 451 445 436 424 As a % of Generation 18.0 17.0 14.2 13.9 13 12 11 10 Demand, MW 523 597 648 692 728 775 828 886 Peak Loss Demand, MW 112 132 122 116 125 123 121 118 Loss Reduction, MW/Year (10) (18) 2 20 23 36 49 62 Loss Reduction, GWh/Year 11.0 72 82 131 187 249 Value of Loss Reduction, Re Million - - 21.4 159.2 170.8 289.3 405.4 431.0 Existing Forecast Generationb/ All Hydro, Gqh 1,624 2,069 2,555 2,706 3,110 3,294 3,300 3,300 Diesel, GWhS/ - 246 350 479 373 413 570 913 Kelanitissa Steam, GWh 245 175 102 113 71 75 175 175 Kelanitissa Gas Turbine, GWh 559 253 - - - 62 101 98 Revised Generation Requirements - Assuming Loss Reduction After 1985 All Hydro, GWh 1,624 2,069 2,555 2,706 3,110 3,294 3,300 3,300 Sapugaskanda Diesel Only, GWh - 246 350 479 373 413 570 763 Kelanitissa Steam, GWh 245 102 113 71 75 175 - 175 Kelanitissa Gas Turbine, GWh 559 - - - - - - - Available Generation Without 1988 Diesel Plant Addition 453 563 783 783 887 928 928 928 Surplus (Deficit) Assuming Loss Reduction (70) (34) 135 91 195 153 100 50 a/ Based on Long Run Marginal Cost. b/ See also Technical Supplement, Generation, Table 3 for more detail and wet/dry season requirements and Annex 1 for capacity balance. - _ __s_] .. -1 r" - -4:*-- 102'4 n%om bn't - 22 - SECTION VII OBSERVATIONS Distribution System 7.01 As part of this study it became evident that CEB has technical, staff and organizational problems in addition to high losses, 7.02 CEB's distribution system is lightly built. Maximum size overhead conductor is 75 sq. mm ASCR or AAC. Poles are a mixture of wood, reinforced concrete, lattice steel towers and poles and tubular steel poles. Cross arms are usually steel. Underground construction uses principally three conductor paper-insulated cables. Largest conductor size is 0.3 sq. in. copper (about 185 sq. mm). Transformers are all three phase, usually on a concrete plinth or a two pole structure. Distribution circuit protection is limited to overcurrent relays and circuit breakers at the substations with some sectionalizing fuses in the lines. 7.03 Various problems and defects were noted, apparently caused by lack of maintenance and poor construction which, if indicative of the whole system, shows a need for general rehabilitation and review of design standards, in addition to a need for a loss reduction program. 7.04 Specific items noted: - Low voltage fuses at transformers were jumpered. High voltage fuses were not in suitable fuse holders. - (round mounted transformers had inadequate enclosures; fences are in poor shape, are too low, or have gates open. At one transformer on a busy street, the door to the secondary fuse compartment was open, exposing switches and fuses. - Clearances between energized conductors and buildings and ground appeared marginal at several locations. - Innumerable aluminum to aluminum and copper to aluminum connections made by wrapping wires together were noted. Oxide inhibiting compound connections are not always used. These connections were primarily in the low voltage distribution system. These contribute to high loss connections and ultimately failure. Suitable connectors are available. - Armour rods or line guards to prevent conductor chaffing at insulators are not used. - On one new transformer installation, a bare aluminum strip was connected to the transformer case and ran underground to a grounding electrode. This is apparently current design. The underground aluminum is subject to corrosion. The electrode was too far away from the transformer to provide good surge suppression. - 23 - Transformers currently in use are designed for veatlt irstalla- tion. Present installation utilizes fenced concrete plinths or two-pole structures with platforms--all expensive. Alternate designs are available which permit installation on a single pole. Surge arresters are not installed at isolating switches or major changes in line angle. Installacion of surge arresters at these locations improves the lightning performance of a line. Few sectionalizing fuses were noted. For long lines, in particular, the use of reclosers and sectionalizers will either restore service or quickly isolate faulted sections of the line and permit rapid restoration of service to customers on unfaulted parts of the circuit. Substation transformers leak oil, usually at radiator connections or tank tops. Several transformers were seen where silica gu.l breathers that needed drying. Improved maintenance is required. Bushings on oil circuit breakers were leaking oil. Concrete poles in coastal areas are spalling and exposing reinforcing bar. Replacement is required. Steel core of ACSR conductor has been corroding with resultant cable failure. Several alternate reinforcing materials are available that reduce corrosion. Street lights in many cases use inefficient incandescent lamps (Annex 8). Voltage levels are low. Metering at substations is not always adequate. Operators do not believe meters are accurate and very often disregard meter readings. Logged readings are often simply rated values. System Control prepares monthly load flow reports but part of the data is estimated-" . Storage of materials is haphazard. 1/ Some improvement will occur when new load control center, which includes telemetering, is completed and magnetic tape recording equipment for substation load analysis is installed. The latter equipment will require upgrading substation wiring. - 24 - - Material control is poor, resulting in material shortages. - Method of system grounding in the 33-kV system requires high insulation levels and for distribution, expensive equipment. - Underground system uses paper insulated cable, which is relatively expensive and requires more skill by the electrician in making joints. Cross linked polyethlene insulated wire is available, usually at lower cost and is easier to install. Joints are very easy to make. - Space availability is limited in the meter test and repair shop&/. Nonstandard watthour meters complicate meter testing. - Transformer shop is crowded and a fire hazard*2/ - If the long lines and crowding at Customer Service facilities at CEB's head office is an indication, expansion of customer service facilities is required. - Equipment available to line crews appears limited. One crew observed were using metal ladders tied together with scrap wire. Further use of metal ladders in or around electrical facilities is considered a major hazard. Voltage 7.05 Low voltages were noticed in different locations in the system. The declared voltage is 230 volts with a minimum of 216 volts and a maximum of 244 volts (*6%). A mid-day spot check on one low-voltage system yielded voltages of 210 to 215 volts. To what extent this drops during the evening peak load is not known, but 200 volts is a conservative estimate. A more graphic example is along the Galle-Colombo road at night where the incandescent lamps glow and the kerosene lights are bright. 7.06 Mid-day voltages of 10.4 to 10.6 kV were noted at Fullerton Sul.itation 11-kV bus and at the 11-kV service point to the Kandy Municipal Corporation; at Padukka Substation 33-kV bus, 31 kV was noted. Independent field checks by CEB staff found voltages as low as 27 kV in the 33-kV system. The voltage calculations made as part of the computer analysis also indicated voltages as low as 24 kV. 7.07 Taps on distribution transformers can correct to some extent for low voltage conditions. To what extent tap changes are made is not known. 1/ A recommendation has already been made to split the shop with one shop for each of the two operating regions. 2/ A proposal has been made to turn over all transformer repairs to Sri Lanka Transformer Company. - 25 - 7.08 As far as can be determined, no attempt is made by substation operators to change substation bus voltage to match load conditions (where on load tap changing is installed). System Control does shift reactive power (var) flow in the system to maintain some control over voltage at the substation bus!!. Data from the System Control Center indicates voltages as low as 104 kV occur at Chunanakan. This has been confirmed by load flow studies which also indicate 55 to 59 kV at substations in the 66-kV system. 7.09 Low voltage exacerbates very poor quality service. For example, motor power factor increases and thereby increases system losses. Light bulb efficiency drops and television operation is poor. 7.10 In addition, low voltage causes a loss in revenue. This, at system peaks, must be offset by incremental high fuel costs. The estimated loss in 1982 of revenue, however, is Rs 56.4 million (US$ 2.5 million). 7.11 At present, with limited generating capacity and high incremental fuel cost during peak, the preferred method of system operation would be to maintain system voltage as high as possible without damaging equipment during off peak hours and letting voltage drop to just acceptable levels during peak. High daytime voltage will increase revenue from industrial/commercial customers. The low voltage during peak will suppress load during peaks, which is desirable with present limited generating capacity. Substation transformer automatic on-load tap changing, switched capacitor banks and much larger con- ductors are required. Distribution Construction Standards 7.12 Present 33-kV distribution system is based on the use of 200 kV basic impulse level (BIL) equipment. This is made a necessity by the use, in general, of grid substation transformers with delta connected secondary and obtaining a neutral through a grounding transformer and a Petersen coil. Although 200 kV BIL is in agreement with British and International Electro- technical Commission standards, electrical equipment is available for 33 kV systems at 150 and 125-kV BIL and at substantially lower cost for all major distribution system components-transformers, switches, surge arresters, capacitors and regulators. Utilization of low BIL equipment requires an effectively grounded substation transformer. For existing substations, a very low impedance grounding transformer would be used; for new stations, a wye connected secondary. 7.13 In conjunction with this, the installation of a neutral conductor will permit the installation of very low cost rural electrification, using single phase construction. In the CEB system, the grounding conductor now installed under the 33-kV crossarms can be converted to a neutral without major cost. Alternatively, a common primary and secondary neutral conductor in new construction will reduce cost. 1/ A prior report has recommended installation of shunt capacitors for voltage control. - 26 - 7.14 Hajor disadvantages are the need for a low impedance grounding transformer at existing substations and the possiblility that existing switchgear that may have inadequate interrupting capacity. 7.15 Typical cost for a 200 kV BIL 2,290 kVAR capacitor bank used here is Rs 423,700; the same bank for a 150-kV BIL system is Rs 350,000. The saving comes in part from capacitor costs and in part because the capacitors can be installed on a single pole Instead of two poles. A 75-kVA single phase trans- former installed is Rs 83,000 compared to Rs 115,000 for a 63 kVA 3-phase transformer on a two pole structure. The 75-kVA transformer can be installed on a single pole. Where single-phase construction is acceptable, line cost goes from Rs 223,000 per kilometer for 3-phase construction to Rs 148,000 per kilometer for single phase construction. System Protection 7.16 Feeder system protection primarily in the 33-kV system is relatively rudimentary, with overcurrent relays only on most substation feeder circuit breakers. Use of reclosing relays, reclosers,±' sectionalizers and multiple fused cutouts can reduce outage time and the number of customers affected by the outage. This is particularly important in the rural areas where repair personnel must travel long distances to reach the point where the fault occurred. Distribution System Engineering 7.17 Overall distribution system engineering and distribution planning in CEB's system appears fragmented and, in fact, completely overlooked within the new organizational plan which splits CEB's system into two operating regions that appear to be semi-autonomous. Although the operating districts and regions have a planning section, there is no Central Distribution Engineering Section. In addition, a separate section handles rural electrification. Planning is presently on a three-year cycle which is acceptable for prepara- tion of annual budgets and detail construction programs, however, no long- range plan apparently exists. The Transmission Division plans 33-kV lines, but responsibility for materials and construction standards rests with dif- ferent sections. A proposal now exists for transferring design standards and specifications to the Commercial Department. 7.18 Distribution expenditures should run 20% to 25% of total invest- ment. Distribution system engineering has become a specialized technical area. The techniques of distribution system engineering are different than that of transmission engineering. Different materials and equipment are used --capacitors, feeder voltage regulators, reclosers, fused cutouts, for example. A need exists for a Department of Distribution Engineering at the same organizational level as Power and Transmission Planning or Generation. This department would prepare and maintain overall system design and 1/ An experimental unit is installed on an 11-kV circuit at Galle Substation. - 27 - application standards, material standards and specifications and standard procedures, analyze distribution system performance, distribution system protection, perform computer analysis /I make economic studies and do long range planning. A more detailed outline of organization and duties is contained in Annex 6. Staff would consist of engineers who are specialists in distribution system engineering. Technical aid to local authorities could be included in this department. 7.19 The establishment of this department, it is true, would downgrade the position of system planning at the regional level but system planning at the district level should not be affected, except for coordination with the proposed Distribution Engineering Department. Staff 7.20 Professional staff is in short supply. Engineers with any reasonable experience have left in large numbers for more attractive employment else- where, mainly in the Middle East. Remaining professional personnel are either senior engineers in management or newly hired engineers. Further, the profes- sional staff has limited exposure to new concepts and trends in the electric utility field. The result is that work is often done on a crisis basis and time is not available for detailed forward planning or investigation of alternative methods. The praparation of good standard construction designs and application guides will ease the training of new staff and also reduce the need for a large number of engineers. Computers 7.21 CEB has made some load flow studies utilizing an existing small mini- computer. The load flow program has apparently been developed internally. Whether or not fault level studies can be made is not known. CEB recently requested proposals for computer software for distribution system analysis and is in the process of purchasing a mini computer similar to the existing computer but with a larger capacity. Whether or not the computer under consideration is the best for the proposed use should be reexamined. Many of the software programs of the type requested by CEB have been structured around a specific operating system. Attempts to transfer programs from one computer to anothei are not always satisfactory. 7.22 In addition to the programs specifically requested by CEB, other engineering programs are available that could profitably be used by CEB, that might reduce the need for engineering staff, an important consideration in view of the present shortage of personnel. I/ Currently, CUB is purchasing the computer hardware and software for load forecasting, distribution system analysis and transformer load management and special metering equipment for obtaining substation data. - 28 - local Authorities 7.23 Local authorities--distribution systems operated by various municipalities, township councils, village councils and urban councils-- provide energy to about the same number of domestic customers as CEB. Local authorities constitute 218 CEB customars and take delivery at 615 different points. Of the 218 customers, 5 take service at 11-kV, the balance at 400 volts. Some are small--15 to 20 kV demand. The large ones, such as the Kandy Municipal Council, have a demand in excess of 2,000 kW and, in the case of Kandy, operate two diesel driven generators on a regular basis. 7.24 No attempt was made to analyze the distribution systems owned by the local authorities. The local authorities do not have much in the way of technical assistance and they appear to copy CEB's practices. 7.25 Loss levels are considered high. From CEB's viewpoint, reduction in losses could reduce requirements for new generation and from the viewpoint of the Sri Lanka Government, a reduction of fuel imports. Local authorities that would have very small systems should be discouraged. 7.26 For existing local authority systems, CEB should establish standards, uniform regulations for service and provide techncial assistance and purchases of materials for system improvement and loss reduction. Alternatively, a separate and independent organization could be established to provide these services. A comprehensive review of the local authority systems, their problems, standards and methods of operation appears warranted. This review should include recommendations for loss reduction. SECTION VIII CONCLUSIONS 8.01 Energy lost and unaccounted for in the CEB system is quite high. As a benchmark, energy lost and unaccounted for in the United States and Western Europe is currently 7 to 7.5X of generation. 8.02 Technical losses can be reduced economically by either inscalling capacitors or reconductoring with larger size conductors as can be seen in all of the preceeding examples (Table 6.2). Highest benefit-to-cost ratios are obtained with the application of shunt capacitors. 8.03 The continued use of very small conductor sizes and "tack-on' line extensions has been a major contributor to high losses. Present practice is to use conductor sizes established many years ago without considering current economic conditions. The loss situation is further aggravated by delays in circuit reinforcement, transmission and grid substation construction. 8.04 Non-technical losses can also be reduced, but it requires a management effort to improve meter reading and billing efficiency and to ensure that all customers are metered. This area is beyond the scope of the study (see however Annex 8). - 29 - Construction 8.05 The distribution system has deteriorated from lack of maintenance. Present methods of construction are expensive. Materials are now available that permit lower cost construction, improved voltage regulation, and improved circuit protection. System rehabilitation and design changes should be undertaken. 8.06 Distribution engineering is fragmented. A central distribution engineering department is needed. Planning 8.07 Distribution system planning at present is limited to a three year period. A longer term plan is needed to provide guidance for future expansion and to avoid 'tack on" extensions and use of undersized conductors that currently cause problems with voltage drop and losses. Computer programs and improved data gathering techniques now available at reasonable cost permit rapid calculation of alternate system arrangements for a least-cost solution for system expansion. Implementation 8.08 The magnitude of work outlined herein is such that even a well organized utility company would have difficulty in accomplishing this work in- house. In addition, CEB has a shortage of trained and experienced profes- sional staff. The most practical method of acquiring the necessary expertise and manpower is through the use of engineering consultants and contractors. The consultants should establish a group of engineers and support staff in Colombo either in a separate office, or if space and commmmications are avail- able, in CEB's office. The consultants should be able to draw on their home office for specialists if necessary and for back up services. The Project Identification Report includes suggested Terms of Reference for this work. 8.09 Any work-modifications, rehabilitation or extensions--must be coordinated with system operations. As much work as possible should be organized by the consultants into a series of contracts that can be handled by local contractors thereby expediting rehabilitation and freeing CEB's staff for emergencies or work that is very difficult for an outside contractor. The consultants' work should include the services of a field engineer to supervise the activities of contractors and coordinate work with CEB's various departments. Cost 8.10 Loss reduction and rehabilitation require investment in additional materials in addition to normal investment for system expansion. Table 8.1 presents an order of magnitude estimate of the total funds required for loss reduction and rehabilitation, although in fact, loss reduction and rehabili- tation is an integral part of distribution system expansion. - 30 - 8.11 The estimte for Phase I includes engineering, approximately 233 MVAR of capacitors, reconductoring or rehabilitating 189 km of 33 and 11-kV line and miscellaneous materials required for rehabilitation. This estimate does not include materials and funds required for current expansion requirements. Phase II includes additional engineering, 152 MVAR of capacitors, 78 km of line, 280 MVA of distribution transformers and 190 MVA of substation capacity and related materials. Table 8.1: Projected Costs - loss Reduction and Rehabilitation, Ceylon Electricity oand, TocAl fve4p local Foreig morel Rs Million RB Mllion Us$ MWiui 1s$ MLion U.S. Million Phase I Engineering 10.9 80.6 .5 3.5 4.0 Tnitial Rehabilitatio and loss Reduction 170.0 230.0 7.4 10.1 17.5 (botingpncy 27.1 46.6 1.2 2.0 3.2 208.0 357.2 9.1 15.6 24.7 II ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I- Thase l E&ginserivg 4.1 30.0 .2 1.3 1.5 Complete RehabilitatIon and Loss Reduction 456.7 653.3 20.0 28.5 48.5 Cbntiincies 69.1 102.5 3.0 4.5 7.5 l¢AL 529.9 785.8 23.2 33.3 56.5 - 32 - SECTION IX RECOMMENDATIONS 9.01 On the basis of the results of the studies made on the system and observations during this iavestigation, the following steps for loss reducti4,. and system improvements are proposed. Initially: (a) Select engineering consultants for use with Phases I and II See Terms of Reference included in the Project Identification Report. (b) Accelerate the construction of the Mataguma Substation and related 132-kV line. The reduction in losses and improved service in the 33-kV system in the area that will be supplied by the substation will pay for an early completion of the station. This is equally true for the 132-kV lines to Chunnakam and Valaichchenai. (c) Until new standards are developed, as a matter of policy all new primary and secondary line construction, except for very short line taps, should be made with the heaviest conductor in stock. (d) Eliminate the use of 0.04 sq.in. 11-kV cable in future underground construction. (e) Purchase approximately 130 MVAR of 33-kV and 15 MVAR 11-kV of capacitors in banks of 1,200 to 2,400 kVAR (nominal) for voltage improvement. Specific locations can be determined by CEB and consultants during the time the capacitors are manufactured and shipped. (f) Monitor all grid substation bus voltages. Direct station operators to change transformer taps as necessary and as directed by System Control to maintain acceptable levels commensurate with available system generating capability. (g) Convert all substation transformers with manually-operatee. on- load tap changers to automatic control where feasible. (h) Initiate a program of monitoring voltages throughout the system to identify areas with the poorest voltage level to start identifying areas that are in need of system improvement first. (i) Establish an improved reporting system to improve the data base for system studies, pending installation of the new magnetic tape metering equipment now on order. New metering equipment should be provided for all substations. In addition, this type of metering should be used for revenue metering of large customers (Annex 8). - 33 - 9.02 Phase I - Loss Reduction and Rehabilitation (a) Establish a Department of Distribution Engineering (Annex 6). (b) Prepare a feasibility study and prepare cost estimates in adequate detail that would permit C1B to obtain financial assistance from outside sources. Cc) As a parallel operation, establish a lose reduction program and a system rehabilitation program, initially utilizing resources available to CEB. (d) Run studies regarding voltage levels, neutral grounding, replacement of the 11-kV system, fault levels, system operations and 66-kV system replacement as part of the feasibility study. (e) Review design standards and choice of materials. (f) Provide training to staff of new Distribution Engineering Department and CEB distribution engineers. (g) Run studies on voltage levels, economic conductor levels, and selection of surge arresters. (h) Run studies on improvements in transmission line operation. (i) Review local authority distribution systems with respect to possible loss reduction, improvement in efficiency and operations. Prepare a feasibility study and cost estimates for loss reduction and system rehabilitation in adequate detail that would permit the Government to obtain financial assistance from outside sources. 9.04 Phase II (a) After initiating the first phase of a loss reduction and rehabilitation plan, prepare a rolling 10-year plan and budget for further rehabilitation and expansion of the distribution and transmission system. The plan for the first 2 years would be in adequate detail to obtain additional funding, if necessary, for outside assistance. (b) Establish a material ident'fication system and computer program for material purchase, iuventory control and preparation of bills of materials. (c) Establish new mapping standards that are suitable for computer- ized system analysis. (d) Establish space requirements for operation and maintenance and equipment requirements for the next 10 year period. ANNEX 1 Page 1 of 1 - 34 - ANNEX 1 STATISTICAL DATA CEYLON ELECTRICITY BOARD Characteristics Frequency 50 Hz Transmission, kV 132, 66 Distribution, Medium Voltage, kV 33, 11, 6.6, 3.3 Distribution, Low Voltage, V 400 Declared Voltage 230 + 6 percent Transmission (End of 1982) 132-kV Double Circuit Tower Line 9 km 132-kV Single Circuit 289 km 66-kV Single Circuit 548 km Number of Grid Substations 22 Installed Substation Capacity (33 and 11 kV Supply) 474 kVA Distribution (End of 1982) 33-kV Double Circuit 161.9 km 33-kV Single Circuit 7,090 km 33-kV Underground 80.6 km 11-kV Overhead 1,588 km 11-kV Underground 549 km Other Voltages 45.5 km 400 Volt 1,194 km Distribution Transformers 4,739 Generation (End of 1982) Plants Nameplate Capability mw MW Hydra 8!.I 402 383 Steam 1 50 40 Piesel 2ˇ/ 20 12.2 3as Turbine 1 120 120 Total 59 a/ In addition, three plants are now under construction, totalling 404 MW. b/ In addQtion, one plant is under construction with 80 MW of capacity. ANNEX 2 -35 - Page 1 of 1 ANNEX 2 SUMMARY OF CIRCUITS ANALYZED Substation Circuit 1982 1982 Ratio & Circuit Length Voltage Connected Demand. Losses Capacity Name Miles kV kVA kVA kW ±1 kW to Load Galle (Bataduma) Rating: 132-33 kV, 3 x 10 MVA Transformers Semidale 206 33 24,917 8,917 7,723 1,088 2.79 Ambalangoda 74 33 16,918 11,442 9,855 1,680 1.48 Udugama 99 33 10,523 3,448 2,965 79 3.05 379 52,358 20,543 2,847 Padukka Rating: 6-J3kV, 4 x 3 MVA Transformers Pagoda 23 33 7,318 4,211 3,579 70 1.74 RHorana 97 33 12,245 4,659 4,100 106 2.63 120 19,563 7,679 T7-6 Colombo A Rating: 33-11 kV, 3 x 12.5 MVA Transformers Transformer No. 12 Ring Main 4.2 11 6,250 3,067 2,615 33 2.04 Colombo B Rating: 33-11 kV, 3 x 12.5 MVA Transformers Transformer No. 12 Ring Main 4.5 11 11,500 4,410 3,570 68 2.61 8.7 17,750 6,185 101 Padukka, Pagoda Feeder Transformer 11-.433 kV, 150 kVA Boys Town 2 .400 - 25 22 0.9 6.0 Colombo A, Feeder to Substation No. 22 Ring 'fain Transformer No. 248, 11-.433 kV, 500 kVA DeFonseka Place 1 .400 182 164 10.1 2.74 * Galle, Semidale Feeder Transformer 11-.433 kV, 50 kVA Samadago 1 .400 - 7 7 .1 7.1 4 186 11.1 a/ Non coincident demands. b/ Ratio of the sum of connected distribution transformers kVA to circuit demand kVA. ANNEX 3 Page 1 of 2 - 36 - ANNEX 3 ADOPTED UNIT COSTS 1. Capacitors - 33-kV installed in 2-pole structure, unswitched Bank Rating, kVARa/ 1,380 2,290 2,750 Capacitors, US$ 3,840 7,665 7,680 Support frame, fuses, surge arrestors 2,157 2,157 2,487 Subtotal 5,997 9,822 10,167 Export pack, shipping 900 1,473 1,524 Total, USS 6,897 11,295 11,691 Rs 157,941 258,656 267,727 Say, Rs 158,000 258,700 268,000 Customs 79,000 13,000 13,400 Support structure 83,000 83,000 86,000 Installation 10,000 12,000 14,000 Engineering and overhead 57,000 57,000 57,000 Total, Rs 315,900 423,700 438,400 11-kV, installed in an enclosure for underground distribution, unswitched Bank rating, kVAR!a/ 780 1,170 1,560 Capaeitors, US$ 2,274 3,708 4,548 Enclosure, fuse, surge arrestors 2,225 2,425 2 425 Subtotal 4,499 6,133 6,973 Export pack, shipping 675 920 1,046 Total, US$ 5,174 7,053 8,019 Rs 118,481 161,514 183,634 Say, Rs 118,500 161,600 183,700 Customs 6,000 8,100 9,200 Installation 30,000 35,000 35,000 Engineering and overhead 31,900 31,900 31,900 Total, Rs 186,400 236,600 259,800 a/ Based on standard 19-kV 60 Hertz capacitors, derated to 50 Hertz and 33-kV or 11-kV, as applicable. ANNEX 3 37 - Page2of2 2. Line Reconductoring - Conductor Size, sq. mm 100 125 170 Cost per 1,000 ft., US$ 316 448 575 Export pack, shipping 16 22 29 Total, US$ 332 n5 604 Rs 7,603 10,763 13,832 Customs at 5% 380 538 692 Cost per 1,000 ft., Rs 7,983 11,301 14,524 Three phase, including: Sag allowance 25,150 35,600 45,760 Accessories 1,490 1,520 1,950 Remove old conductor 3,000 3,000 3,000 Install new conductor 3,700 3,900 4,200 Transport 800 800 800 Subtotal, Rs 34,140 44,820 55,700 Engineering and Overhead 5,120 6,720 8,360 Total, per 1,000 ft., Rs 39,260 51,540 64,060 Per mile, Rs 207,300 272,130 338,240 ANNEX 4 Page 1 of 7 - 38 - ANNEX 4 TYPICAL CCMPUTER PRINTOUTS 1.0 The following are typical computer printout sheets of the results of the loss analysis program. The basic computation is loss and voltag2 drop (Program BALVOL or Program CAPLOC). The results can be obtained either in a detailed computation of losses for each part of the circuit, or a summary by years showing losses, demand voltage level and line capacity. The other printouts (Program CSTLOS) are a financial calculation, which provides the present worth of the cost of losses. 2.0 Most column headings are self explanatory, except the following: Sect - Section number assigned to identify subdivisions of the circuit Lgth K - Section length in 1,000 feet Cond % or - Percentage loading of the conductor as based on the X Cap rated conductor capacity in the data base Tran - Cumulative number of distribution transformers through the sect ton Specific examples are: Colombo Substation A Base Case (Program BALVOL) Coloubo Substation A -Add 1,170 kVAR capacitors (Program CAPLOC) Colombo Substation A Summary Colombo Substation A Financial Analysis (Program CSTLOS) ANNEX 4 Page 2 of 7 - 39 - Program BALVOL Program CAPLOC Columbo Substation A, Transformer 22 Ring Main Locations of Ring Main Transformers Section Transformer Number 2 22 4 248 9 183 12 65 15 165 18 77 22 277 25 179 29 214 32 309 36 238 Proposed capacitors are located adjacent to Transformer 77. ANNE% 4 - 40 - Page 3 of 7 PROGRAM BALVOL DATE 6/1/83 CASE 3 BASE 1982 FEEDER 32 COLUMBO'A"-S/S*22, VOLTAGE - lbOO KV LINE TO LINE SECT LGTH PHAS COND ---LOAD THRU SECTION--- VOLTAGES (PERCENT) LOSSES K- CONF SIZE COND KW KVAR AMPS TRAN SECT ACCU LEVEL KW UNITS ' DROP DROP SUBSTATION TOTALS 2615. 1616. 100.0 1 4.4 3 11/.3/Cu 42.5 2615. 1616. 161.3 11. .7 .7 99.3 13.40 2 .0 3 11/.3/Cu 9.1 553. 344. 34.4 1. .0 .7 99.3 .00 3 .7 3 11/04/Cu 39.9 808. 490. 49.9 4. .2 .9 99.1 1.48 4 .2 3 ll/.l/Cu 22.7 806. 488. 49.9 4. .0 .9 99.1 .15 5 .0 3 11/.1/Cu 4.4 160. 87. 9.7 1. .0 .9 99.1 .00 6 .2 3 11/.1,'Cu 18.3 646. 401. 40.3 3. .0 .9 99.1 .10 7 1.5 3 11/04/Cu 32.2 646. 401. 40.3 3. .4 1.3 98.7 2.06 e .9 3 ll/.l/Cu 18.3 643. 399. 40.3 3. .1 1.4 98.6 .48 9 .0 3 l1/.1/Cu 3.7 132. 81. 8.2 1. .0 1.4 98.6 .00 10 .9 3 1I/.1/Cu 14.5 511. 317. 32.0 2. .1 1.4 98.6 .30 11 .8 3 11/04/Cu 25.6 S11. 317. 32.0 2. .1 1.6 98.4 .65 12 .0 3 11/04/Cu 21.7 432. 267. 27.1 1. .0 1.6 98.1 .00 13 .3 3 11/04/Cu 3.9 78. 49. 4.9 1. .0 1.6 98.4 .01 14 .6 3 l/.1/Cu 2.2 78. 49. 4.9 1. .0 1.6 98.4 .00 15 .0 3 11/.1/Cu 2.2 78. 49. 4.9 1. .0 1.6 98.4 .00 16 1.3 3 11/04/Cu 61.6 1241. 766. 77.1 6. .6 1.3 98.7 6.54 17 .3 3 11/.1/Cu 35.0 1235. 759. 77.1 6. .1 1.3 98.7 .56 18 .0 3 1I/.1/Cu 10.4 367. 228. 23.0 1. .0 1.3 98.7 .00 19 .2 3 11/.1/Cu 24.6 867. 531. 54.1 5. .0 1.4 98.6 .18 20 1.2 3 11/04/Cu 43.3 867. 530. 54.1 5. .4 1.8 98.2 3.10 21 1.1 3 11/.1/Cu 24.6 864. 527. 54.1 5. .1 1.9 98.1 1.04 22 .0 3 11/.1/Cu 2.9 102. 63. 6.4 1. .0 1.9 96.1 .00 23 1.1 3 11/.1/Cu 21.7 761. 463. 47.7 4. .1 2.0 98.0 .79 24 .4 3 11/04/Cu 38.1 760. 462. 47.7 4. .1 2.2 97.8 .87 25 .3 3 11/.1/Cu 21.7 759. 461. 47.7 4. .0 2.2 97.8 .21 26 .0 3 ll/.1/Cu 9.5 331. 206. 20.9 1. .0 2.2 97.8 .00 27 .3 3 11/.1/Cu 12.2 428. 255. 26.7 3. .0 2.2 97.8 .07 28 .7 3 11/04/Cu 21.4 428. 255. 26.7 3. .1 2.3 97.7 .42 29 .0 3 11/04/Cu 10.0 197. 123. 12.5 1. .0 2.3 97.7 .00 30 .9 3 11/04/Cu 11.4 230. 132. 14.3 2. .1 2.4 97.6 .16 31 .3 3 11/.1/Cu 6.5 230. 132. 14.3 2. .0 2.4 97.6 .02 32 .0 3 11/.1/Cu 4.5 160. 87. 9.8 1. .0 2.4 97.6 .00 33 .3 3 11/.l/Cu 2.0 70. 45. 4.5 1. .0 2.4 97.6 .00 34 .4 3 11/04/Cu 3.6 70. 45. 4.5 1. .0 2.4 97.6 .01 35 .4 3 11/.1/Cu 2.0 70. 45. 4.5 1. .0 2.4 97.6 .00 36 .0 3 11/.1/Cu 2.0 70. 45. 4.5 1. .0 2.4 97.6 .00 END OF FEEDER TOTAL LOSSES ON FEEDER 32 KW 32.61 KVAR w 36.35 NOTE: KW AND KVAR LOADING THRU SECTION INCLUDES DEMANDS WITHIN SECTION, LOAD ING FROM DOWNSTREAM SECTIONS, AND LOSSES. ANEK 4 -41 Page 4 of 7 IOGRAM CAPLOC DATE 6/1/83 .ASE 4 ADD 1170 KVAR CAP BANK 1982 FEEDER 32 -- COLUMBO"A"-S/S#22 VOLTAGE - 11.00 KV LINE TO LINE THIS RUN WITH ALL CAPACITORS TEMPORARILY PLACED SECT LGTH NO. COND --- LOAD THRU SECTION --- VOLTAGES (PERCENT) LOSSES K- PHAS SIZE COND KW KVAR AMPS TRAN SECT ACCU LEVEL KW UNITS D DROP DROP SUBSTATION TOTALS 2609. 439. 100.0 1 4.4 3 11/.3/Cu 36.5 2609. 439. 138.9 11. .4 .4 99.6 9.92 2 .0 3 11/.3/Cu 9.0 553. 344. 34.3 1. .0 .4 99.6 .00 3 .7 3 11/04/Cu 39.8 808. 490. 49.8 4. .2 .7 99.3 1.47 4 .2 3 111.1/Cu 22.6 806. 488. 49.8 4. .0 .7 99.3 .15 5 .0 3 11/.1/Cu 4.4 160. 87. 9.6 1. .0 .7 99.3 .00 6 .2 3 11/.1/Cu 18.3 646. 401. 40.2 3. .0 .7 99.3 .10 7 1.5 3 11/04/Cu 32.1 645. 401. 40.2 3. .4 1.1 98.9 2.05 8 .9 3 11/.1/Cu 18.3 643. 399. 40.2 3. .1 1.2 98.8 .48 9 .0 3 11/.1/Cu 3.7 132. 91. 8.2 1. .0 1.2 98.8 .00 10 .9 3 11/.1/Cu 14.5 511. 17. 31.9 2. .1 1.2 98.8 .30 11 .8 3 11/04/Cu 25.5 511. 317. 31.9 2. .1 1.4 98.6 .65 12 .0 3 11/04/Cu 21.6 432. 267. 27.0 1. .0 1.4 98.6 .00 13 .3 3 11/04/Cu 3.9 78. 49. 4.9 1. .0 1.4 98.6 .01 4 .6 3 11/.1/Cu 2.2 78. 49. 4.9 1. .0 1.4 98.6 .00 .5 .0 3 11/.1/Cu 2.2 78. 49. 4.9 1. .0 1.4 98.6 .00 16 1.3 3 11/04/Cu 55.0 1239. -407. 68.7 6. .3 .7 99.3 5.20 17 .3 3 11/.1/Cu 31.2 1234. -412. 68.7 6. .0 .7 99.3 .44 18 .0 3 11/.1/Cu 10.4 367. 228. 22.8 1. .0 .7 99.3 .00 19 .2 3 11/.1/Cu 25.9 866. -640. 56.9 5. .0 .7 99.3 .20 20 1.2 3 11/04/Cu 45.6 866. -640. 56.9 5. .2 1.0 99.0 2.23 CAPACITOR IN SECTION 20 1170 21 1.1 3 11/.1/Cu 24.4 864. 527. 53.6 5. .1 1.1 98.9 1.03 22 .0 3 11/.1/Cu 2.9 102. 63. 6.4 1. .0 1.1 98.9 .00 23 1.1 3 11/.1/Cu 21.5 761. 463. 47.3 4. .1 1.2 98.8 .77 24 .4 3 11/04/Cu 37.8 760. 462. 47.3 4. .1 1.4 98.6 .85 25 .3 3 11/.1/Cu 21.5 759. 461. 47.3 4. .0 1.4 98.6 .21 26 .0 3 11/.1/Cu 9.4 331. 206. 20.8 1. .0 1.4 98.6 .00 27 .3 3 11/.1/Cu 12.1 428. 255. 26.5 3. .0 1.4 98.6 .07 28 .7 3 11/04/Cu 21.2 428. 255. 26.5 3. .1 1.5 98.5 .42 29 .0 3 11/04/Cu 9.9 197. 123. 12.4 1. .0 1.5 98.5 .00 30 .9 3 11/04/Cu 11.3 230. 132. 14.2 2. .1 1.6 98.4 .15 31 .3 3 11/.1/Cu 6.4 230. 132. 14.2 2. .0 1.6 98.4 .02 32 .0 31 1/.1/Cu 4.4 160. 87. 9.7 1. .0 1.6 98.4 .00 33 .3 3 11/.1/Cu 2.0 70. 45. 4.4 1. .0 1.6 98.4 .00 34 .4 3 11/04/Cu 3.5 70. 45. 4.4 1. .0 1.6 98.4 .01 35 .4 3 11/.1/Cu 2.0 70. 45. 4.4 1. .0 1.6 98.4 .00 36 .0 3 11/.1/Cu 2.0 70. 45. 4.4 1. .0 1.6 98.4 .00 END OF FEEDER rAL LOSSES ON FEEDER 32 KW 2 26.76 KVAR - 29.59 NOTE: KW AND KVAR LOADING THRU SECTION INCLUDES DEMANDS WITHIN SECTION, LOADING FROM DOWNSTREAM SECTIONS. AND LOSSEa. AM 4 -42-a sge 5 of 7 CASE 3 BASE 6/1/83 COLUMBO " A"-S/S#22 SUMMARY FOR FEEDER 32 YEAR X --- -KW---- KVA KVAR --MAX LOAD--- MAX VOLT DROP GROW DEM LOSS DEM DEM SEC %CAP AMPS SECT LEVEL 82 0.00 2615.2 32.6 3074.0 1615.7 16 62. 77.1 36 97.6 83 6.90 2774.2 36.7 3261.6 1715.1 16 65. 81.8 36 97.4 84 6.00 2943.1 41.4 3460.8 1820.7 16 69. 86.8 36 97.3 85 6.00 3122.5 46.6 3672.4 1933.0 16 74. 92.1 36 97.1 86 6.00 3312.9 52.5 3897.2 2052.4 16 78. 97.w 36 96.9 87 5.00 3481.5 58.0 4096.2 2158.3 16 82. 102.8 36 96.7 88 5.00 3658.7 64.1 4305.6 2269.7 16 87. 108.1 36 96.6 89 5.00 3845.2 70.9 4526.0 2387.2 16 91. 113.7 36 96.4 90 4.00 4002.1 76.8 4711.5 2486.1 16 95. 118.4 36 96.3 CASE 4 ADD 1170 KVAR CAP BANK 6/l/83 COLUMBO"A" -S/S#22 SUMMARY FOR FEEDER 32 YEAR % - ---KW ----- KVA KVAR --MAX LOAD- MAX VOLT DROP GROW DEM LOSS DEM DEM SEC %CAP AMPS SECT LEVEL 82 0.00 2609.3 '26.8 2646.0 439.0 16 55. 68.7 36 98.4 e3 6.00 2767.5 30.0 2819.2 537.4 16 57. 71.9 36 98.2 84 6.00 2935.5 33.7 3004.9 642.0 16 60. 75.3 36 98.1 85 6.00 3113.8 38.0 3203.6 753.2 16 63. 79.2 36 97.9 86 6.00 3303.2 42.8 3416.3 871.4 16 67. 83.5 36 97.7 87 5.00 3470.8 47.4 3605.5 976.2 16 70. 87.4 36 97.6 88 5.00 3647.1 52.5 3805.6 1086.6 16 73. 91.6 36 97.4 89 5.00 3832.6 58.2 401S.9 1202.9 16 77. 96.2 36 97.2 90 4.00 3988.6 63.3 4195.4 1300.9 16 80. 100.1 36 97.1 AMX 4 "ae 6 of 7 -43- PROGRAM CSTLOS -- 6/14/83 CASE 3 BASE COLUMBO A SUB - TRANSFORMER 22 RING MAIN DEMANDS AND ENERGY YEAR/. DEMAND ENERGY ---- LOSSES ---- PERCENT LOSSES FACTORS(%.) SEASON KW MWH PEAK KW MWH PEAK ENERGY LOAD LOSS 85 1 3122.5 15591.3 46.6 138.8 1.49 .89 57.0 34.0 86 1 3312.9 16542.0 52.5 156.4 1.58 .95 57.0 34.0 87 1 3481.5 17383.8 58.0 172.7 1.67 .99 57.0 34.0 88 1 3658.7 18268.6 64.1 190.9 1.75 1.05 57.0 34.0 89 1 3845.2 19199.9 70.9 211.2 1.84 1.10 57.0 34.0 NUMBER OF STUDY YEARS 3 5 NUMBER OF SEASONS I 1 INTEREST RATE (X) 11.00 MONTHS IN 1ST SEASON 12 CURRENCY; SRI LANKA RUPEES PROGRAM CSTLOS -- 6/14/83 CASE 3 BASE COLUMBO A SUB - TRANSFORMER 22 RING MAIN CoSrs AND PRESENT WORTH OF COSTS YEAR/ ANNUAL COST OF LOSSES PRESENT WORTH OF COST OF LOSSES SEASON CAPACITY ENERGY TOTAL CAPACITY ENERGY TOTAL 85 1 132017.8 198474.6 330492.4 132017.8 198474.6 330492.4 86 1 148732.5 223603.4 372335.9 133993.2 201444.5 335437.7 87 1 164314.0 247028.5 411342.5 133360.9 200493.9 333854.8 88 1 181595.3 273009.1 454604.4 132780.9 199621.9 332402.8 89 1 200859.7 301971.1 502830.8 132312.5 198917.7 331230.2 TOTALS 827519.3 1244086.7 2071606.0 664465.4 998952.6 1663418.1 FUTURE 4017194.0 6039421.510056616.0 949232.3 1427069.2 2376301.5 SERIES (25 YRS) TOTALS 4844713.5 7263508.512128222.0 1613697.7 2426021.7 4039719.7 NUMBER OF STUDY YEARS 8 S NUMBER OF SEASONS - I INTEREST RATE (7) 11.00 MONTHS IN 1ST SEASON = 12 CURRENCY: SRI LANKA RUPEES ANX$ 4 44 -A -M Page 7 of 7 PROGRAM CSTLOS -- 6/14/83 CASE 4 ADD 1170 KVAR CAP BKt COLUMBO A SUB - TR.fNS 22 RING M DEMANDS AND ENERGY YEAR/ DEMAND ENERGY ---- LOSSES ---- PERCENT LOSSES FACTORS(%) SEASON KW MWH PEAK KW MWH PEAK ENERGY LOAD LOSS 85 1 3113.8 15547.8 38.o 113.2 1.22 .73 57.0 34.0 86 1 3303.2 16493.5 42.8 127.5 1.30 .77 57.0 34.0 87 1 3470.8 17330.4 47.4 141.2 1.37 .81 57.0 34.0 88 1 3647.1 18210.7 52.5 156.4 1.44 .86 57.0 34.0 89 1 3832.6 19136.9 58.2 173.3 1.52 .91 57.0 34.0 NUMBER OF STUDY YEARS S NUMBER OF SEASONS 1 INTEREST RATE (%) - 11.00 MONTHS IN 1ST SEASON - 12 CURRENCY: SRI LANKA RUPEES PROGRAM CSTLOS -- 6/14/83 CASE 4 ADD 1170 KVAR CAP SK, COLUMBO A SUB - TRANS 22 RING M COSTS AND PRESENT WORTH OF COSTS YEAR/ ANNUAL COST OF LOSSES PRESENT WORTH OF COST OF LOSSES SEASON CAPACITY ENERGY TOTAL CAPACITY ENERGY TOTAL 85 1 107654.0 161846.3 269500.3 107654.0 161846.3 269500.3 86 1 121252.4 182290.0 303542.4 109236.4 164225.2 273461.6 87 1 134284.2 201881.9 336166.2 1O8988.1 163851.9 272840.0 88 1 148732.5 223603.4 372335.9 108751.9 163496.9 272248.8 89 1 164880.6 247880.4 412761.0 108612.0 163286.5 271898.4 TOTALS 676803.7 1017502.0 1694305.7 543242.3 816706.7 1359949.2 FUTURE 3297612.2 4957607.0 8255219.5 779200.6 1171444.9 1950645.5 SERIES (25 YRS) TOTALS 3974416.0 5975109.0 9949525.0 1322442.9 1988151.6 3310594.7 NUMBER OF STUDY YEARS 5 NUMBER OF SEASONS I 1 INTEREST RATE (7.) = 11.00 MONTHS IN 1ST SEASON - 12 CURRENCY: SRI LANKA RUPEES ANNEX 5 Page 1 of 8 - 45 - ANNEX 5 LONG RUN MARGINAL COST FOR VALUING TECHNICAL LOSSES IN THE ELECTRIC POWER SYSTEM Introduction 1. This note sets out the computation of the economic long run marginal cost (LRMC) for valuing technical losses in the power system in Sri Lanka. 2. The economic LRMC may be defined as the economic resource cost of providing for an incremental demand on the power system. 3. For the purpose of valuing losses, the LRMC is estimated as a two- part cost, an incremental capacity related cost and an incremental energy cost. Existing Facilities 4. Presently, CEB's total installed capacity is 592 MW. A summary of generation and lines is presented in Annex 1. Currently, load dispatch and control is done from one center location at Kelanitissa. A new center, which will include economic dispatch facilities, is under construction at Kolawanna. Historical Consumption and Load Forecasts 5. From 1961 to 1980, electrical energy consumption grew at the rate of 9 5X per annum on the average, ranging from 17.8% in 1966 to 3.3% in 1974. Gross energy generation is expected to increase from 2,065 GWh in 1982 to 4,485 GWh in 1990, an average annual rate of increase of 10.5%. Peak demand is expected to increase from 430 MW in 1982 to 948 MW in 1990. The system's load factor is forecast to remain at about 54%. Load Characteristics 6. The country has two seasons, a wet season and a dry season, each of 6 months duration. The daily load curve shows the same pattern in both the wet and dry seasons, with the wet season load about 8% higher than the dry season load. Currently during the dry season, some self-producers generate their own power due to CEB's inability to meet their requirements. However, during the wet season, CEB is capable of meeting the entire demand including those of self-producers. 7. The peak demand on the system occurs between 1800 and 2200 hours, with a lower daytime peak from 0600 hours to 1200 hours. Large industries consume about 27% of the total energy sales. The contribution to the daytime peak is from domestic lighting in the early hours of the morning and the small ANNEX 5 - 46 - Page 2 of 8 and medium commercial and industrial loads. The system nighttime peak consists primarily of lighting and other domestic loads. The shape of the daily load curve is expected to remain the same in the foreseeable future. Expansion Program, 1983 - 1991 C. CEB's 1983 - 1991 generation expansion program shown in Table 9 attached, indicates that up to 1991 future capacity additions will be mostly in hydro plants. Additional hbdro capacity is estimated at 500 MW and new diesel at 40 MW. The first unit of 120 MW for coal-fired steam plant will not be in service until 1991 .1. 9, Transmission and distribution additions to the system under the World Bank Sixth and Seventh Power Projects are expected to be completed in 1984 and 1985. A rural electrification program being partly financed by ADB and OPEC is expected to be completed in 1984. CEB plans to carry out additions and improvements annually to the transmission and distribution system. In addition, it is assumed that the systems owned by the local authorities will expand. Marginal Capacity Cost 10. The long term investment program provides the basis for estimating the LRMC of capacity. The relevant portion of the investment program is that which appears to be most sensitive to changes in the incremental kilowatt demand. 11. Investments in Kotmale and Vittoria have already been committed and changes in kilowatt demand would not affect the timing of this investment; however, changes in demand would affect both timing and amount of investment in Rantembe/Randenigala. Rantembe/Randenigala therefore appears as the marginal investment or as a representative of the least cost means of prcviding the incremental kilowatt of capacity. Marginal capacity cost at generation is therefore to be based on the cost of Rantembe/Randenigala. 12. The relevant marginal capacity related costs are the investments and fixed operation and maintenance costs of the power house and associated electro-mchanical equipment of Rantembe/Randenigale and those of the incremental additions to the network. 13. All costs are expressed in constant 1983 price levels and in economic terms to reflect the economic cost of scarce resources. 1/ We have modified CEBts investment program which shows a 60 MW steam plant coming into service in 1989. We feel the timing is unrealistic and also since the maximum demand is increasing at the rate of 95 MW per year on the average, we have adopted 120 MW steam unit instead of 60 MW in order to allow for an appropriate reserve margin. At the moment the cost for the coal-fired plant is uncertain. ANNEX 5 Page 3 of 8 - 47 - 14. Using an 11% discount rate as the opportunity cost of capital to Sri Lanka, the incremental capacity related costs for major components of capacity are estimated as follows: TABLE 1 Cost Voltage Level Rs/kW/Year US $/kW/Year Generation 1,312 57 High (132-66 kV) 697 30 Medium (33-11 kV) 480 21 Low (400 V) 202 9 Losses 15. There are two main type of losses in an electrical network; technical and non-technical. Technical losses are heat losses which are determined by the electrical characteristics and the amount of load on the network. Non- technical losses consist of unmetered consumption and unaccounted for losses. A summary of kilowatt and kilowatt-hour technical losses is given below: TABLE 2 Voltage Kilowatt Loss Kilowatt-Hour Loss Level (% of max. demand) (X of gross generation) High (132-66 kV) 11.1 8.0 Medium (33-11 kV) 9.2 4.8 Low (400 V) 2.7 1.8 Marginal Capacity Cost Matrix 16. In order to obtain a value of the long run marginal capacity cost at various voltage levels, the incremental capacity related costs shown in Table 2 are adjusted for kilowatt technical losses up to the various voltage levels. The resulting long run marginal costs are shown in Table 3. Marginal Energy Cost 17. Marginal energy cost is given by the incremental fuel, operation and maintenance costs of the generating facilities that would be used to provide the incremental kilowatt-hour. ANNEX 5 -48- Page 4 of8 TABLE 3 LONG RUN MARGINAL COST MATRIX PM/klW tear (USSwwa7%r) Gen. H (132-66 W) Medim (33-11 kV) tI (400V) (Bn 1,312 1,457 1,592 1,635 4 Nigh (132-66 kV) - 697 761 782 Medium (33-11 kV) - - 480 493 Tm (400N) - - 202 Total 1,312 2,123 2,833 3,112 (57) (93) (124) (136) Iceses M%) 11.1 9.2 2.7 18. A simulation of the generating system aimed at providing the required energy at the least cost and which shows the stacking sequence of existing plants and incremental generating additions up to .1990 indicates that at all peak times in the conceivable future, with the exception of the period from 1985 to 1988, the existing gas turbines would be used to provide the energy at critical periods in the dry season. Heavy oil-fired diesel plants would be used to provide energy at peak periods in the wet season in all years from 1984 onwards. 19. Currently, the gas turbines are operated at an average of 12.4 hours per day. As from 1984, this is expected to reduce to about 6 hours per day and only in the dry season to meet peak load requirements. 20. As a result of the pronounced seasonal variation in the availability of water to meet power generation requirements, the stacking sequence of plants to meet the demand on the system varies from season to season as shown in Table 4. 21. In the wet season, the steam plant is placed in cold storage and the gas turbines are assumed to be in reserve in periods that they are not required. The sticking sequence is based only on per unit cost of generation. 22. The international border price of fuel is used to compute thermal generation cost. Table 5 summarizes the results of incremental fuel cost per kilowatt-hour. ANNEX 5 49 - Page 5 of 8 TABLE 4 SESONAL PLANT STACKING SEQUENCE Year Dry Season Wet Season 1984 Gas Turbine Diesel Steam Hydro Diesel Hydro Hydro Hydro 1985 - 1988 Steam Diesel Diesel Rydro Hydro Hydro 1989 Gas Turbine Diesel Steam Hydro Diesels Hydro Hydro lydro TABLE 5 ANYSIS OF FUEL COST Plant Gas Heat Rate Turbine Steam Diesel (Btu/kWh) 14,150 11,000 9,500 Type of Fuel Diesel 2 Furnace Oil Furnace Oil Price of Fuel (Rs/Litre) 5.76 3.88 3.88 Fuel Cost (Rs/kWh) 2.18 1.110 0.9375 23. Marginal energy cost of thermal generation is obtained by adjusting the incremental fuel cost for incremental variable operation and n:aintenance cost. 24. The cost of energy from base load hydro plants is based on the 16 average capital cost of reservoirs at Victoria and Kontmale. 25. The average off-peak energy cost is based on the energy cost of facilities required to provide the off-peak energy in the dry and wet seasons. The relative duration in hours of peak and off-peak are used as weights in determining the weighted average energy cost per year. Table 6 sumrizes the marginal energy cost per type of plant. ANNEX 5 -50 - Page 6 of 8 TABLE 6 MWRGINAL ENERGY COST Rs/IkWh Gas Steam Diesel Base T.rbine Plant Plants Hydro 2.70 1.41 1.20 0.251 26. Table 7 shows the peak, off-peak and average incremental energy cost per year. Table 7 Discount Off- Factor Year Peak Peak Average (r - 11%) 1983 2.70 2.70 2.70 1 1984 1.95 1.09 1.23 0.90 1985 1.31 0.725 0.82 0.81 1986 1.31 0.725 0.82 0.73 1987 1.31 0.725 0.82 0.66 1988 1.31 0.725 0.82 0.59 1989 1.95 1.09 1.23 0.53 1990 1.95 1.09 1.23 0.48 Infinity 1.95 1.09 1.23 27. The present value of incremental energy cost is given by: WiXi - 11.96 - 1.26 Wi 9.46 where, Wi - discount factor Xi = annual average incremental energy cost The long term average incremental cost of energy is therefore Rs 1.26/kWh at the generation level. ANNEX 5 - 51 - Page 7 of 8 28. To obtain the average incremental energy cost at various voltage levels, the cost at generation is adjusted for average technical energy losses up to the respective voltage levels. Table 8 summarizes the average incre- mental energy cost at the various voltage levels. TABLE 8 LONG-RUN AVERAGE INCREMENTAL ENERGY COST Rs/kWh (US $/kWh) Generation High Voltage Medium Voltage Low Voltage (132/66 kV) (33/11 kV) (400 V) 1.26 1.36 1.43 1.45 (0.0550) (0.0594) (0.0624) (0.0633) Losses M%) 8 4.8 1.8 TABLE 9 POWER SECTOR INVESTMENT PROGRAM 1981-1990 (Rs Million in 1980 Prices) Total 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 Generation CEB: Bowatenne/Canyon 631 140 211 280 Thermal phase 378 378 - - - - - - - - - Thermal phasea/ 1,240 120 870 250 - - - - - - Non-CEB_k: Kotmale 6,600 1,330 1,630 1,620 1,510 510 - - - - - Victoria 4,850 1,090 1,260 1,500 1,000 - - - - - - L Randeigala/ Rantembe 5,350 - 1,070 620 920 i,110 250 320 610 450 - Transmission/Distribution Sixth IDA 760 80 320 250 110 - - - - - Seventh IDA 1,152 - 319 716 69 48 - - - - - Other 1,097 266 227 172 253 179 - - - - - Rural Electrification ADB project 490 313 120 57 - - - - - - - Other workss/ 15,036 297 359 330 300 300 1,850 3,400 3,300 2,900 2,000 o 0 Total Power Sector 37,584 4,014 6,386 5,795 4,162 2,147 2,100 3,720 3,910 3,350 2,000 a/ Assuming diesel plant (80 MW). b/ Mahweli Development Authority. Includes complete headworks. c/ From 1986, includes all CEB capital investment. ANNEX 6 Page I uf 2 ANNE 6 STREET LIGHTING * 1.0 Of the 22,000 street lighting fixtures owned by CEBJl/ a large percentage appear to be either incandescent or self-ballasted mercury vapor. Typical lumens per watt and cost per lumen are tabulated in Table 1. These are tabulated for lamps manufactured in North America but are considered typical. TABLE 1 Rated Annual Life Lumens/ Cost of En?;gy Light Source Hours Watt Rs/Lumeni$ 100 watt incandescent .750 17.5 .4803 160 watt self-ballasted Mercury vapor_/ 24,000 10.6 .7905 100 watt Mercury vapor 24,000 22.1/ .3798 100 watt high pressure sodium vapor 24,000 63.3 .1288 50 watt high pressure sodium vapor 24,000 52.9 .1588 a/ Data on 100 watt self-ballasted not available. b/ Ten hours per day operation, Rs 3,112 per kilowatt capacity, Rs 1.45 per kilowatt hour. These are long run marginal costs. c/ Based on mean lumens, includes ballast losses. -~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 2.0 Assuming 11,000 of 22,000 lights in CEB's system are 100 watt incandescent, the annual cost of energy is Rs 9.24 million. Assuming that these are replaced with 50 watt high pressure sodium vapor fixtures, the cost of energy is reduced by one third and the level of lumination will increase two fold. In addition, lamp life is much longer. The savings in energy is approximately 1.3 GWh per year (not quite half because of ballast losses for high pressure sodium vapor). Changing mercury vapor to sodium vapor lamps also provides a savings, although not as great. 1/ It is assumed that the local authorities also own a substantial number of lights. ANNEX 6 54 - Page 2 of 2 TABLE 2 ANNUAL COST, INCANDESCENT VS SODIUM VAPOR LIGHTS 100 Watt 50 Watt Incandescent Sodium Vapor Annual cost, energy Rs 840 Rs 572 Annual cost of lamps 28a/ 140 Annual cost of a new sodium vapor fixture - 206 TOTAL Rs 866 Rs 918 a/ Probably less since lamp life is extended due to chronic low voltage in the system. ANNEX 7 Page i of 2 - 55 - ANNEX 7 METERING 1. Table 2.1 classifies the difference between generation and sales as "energy lost and unaccounted for." Table 2.2 is an attempt to segregate losses into technical and non-technical losses and into different parts of CEB's system. 2. The non-technical losses are classified as metering and billing errors, inaccurate or lackt of metering, or theft. A reduction in non- technical losses can be made. The foNllowing are recommended: (a) Meters that are defective should be replaced promptly. All customers should have meters. In addition, station service for all grid substations should have meters. Some substations, it has been noted, include warehousing and staff ho%sfing. Energy used for these facilities should be accounted for_- (b) Energy used for street lighting at present is apparently estimated. Estimates of street lighting evergy use can be improved through the installation of watthour meters in several typical street lighting circuits. Sample data can be used as a check against estimates. (c) Watthour calibration should be checked at least once a year on all large customers. In addition, the calibration of watthour meters in all power plants and substations should be checked. (It is understood the accuracy of watthour meters at the power plants was checked when the plant was commissioned but has not been checked since.) (d) Installation of watthour meters at the grid substations in the low voltage side of all transformers (preferably with demand attachment for kilowatt measurement) can be used to determine energy losses in the transmission system and to monitor station demand. Although telemetering may be added to the grid substations, a permanent wattmeter serves as a back-up device should the telemetering system fail. The proposed cassette (magnetic tape) recorders proposed for installation at the substation may alto require a watthour meter as an input device. 1/ Energy use by CEB's facilities should be in separate category in system reports. ANNEX 7 -56 - Page 2-of 2- (e) Telemetering that is now being installed is concerned primarily with generation and the substations associated with the Mahaweli project. Telemetering stould be extended to all grid substa- tions. (f) CEB released requests for proposals for the supply of magnetic tape (cassette) type equipment for installation on the feeders to measure substation and feeder load characteristics. This is an excellent approach. The same type of equipment can also be used for large customer revenue metering. This type of equip- ment is far more accurate than conventional electromechanical equipment, permits easy application of time-of-day rate schedules, and custom"r power factor monitoring. (g) CEB does not have a power factor penalty clause in the rate schedules, but depends instead on a high kVA demand charge to encourage customers to correct power factor. Although the kVA charge may be high enough to cover CEB's additional capacity and energy losses rasulting from a low power factor load, this type of rate is a very indirect signal to the customer that he has a low power factor load. Further, the customer may not realize that the power factor of his load is low. On the other hand, a power factor penalty charge is a direct signal to the customer that he has a problem. Meters that are installed to obtain the customers' kVA demand are, in fact, kVA-hour meters. If the kVA hour data is read by the meter reader in addition to kilowatt hours and kVA demand and the billing computer programs are modified, an average power factor can be calculated and included in the customer bills, either for information or altirnatively used to compute a penalty.