Insert Table of Contents Here

FollowUpConsultancytothePowerSystemLossStudiesfortheFSMUtilities – Draft Report version 2

Kosrae Utility Authority

Submitted to : Pacific Power Association, Suva, Fiji

Submitted by : KEMA Inc.

Date : 23 September 2013

This document, and the information contained herein, is the exclusive, confidential and proprietary property of the PPA and is protected under the trade secret and copyright laws of the U.S. and other international laws, treaties and conventions. No part of this work may be disclosed to any third party or used, reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or by any information storage or retrieval system, without first receiving the express written permission of the PPA. Except as otherwise noted, all trademarks appearing herein are proprietary to the PPA.

Contents

1 Executive Summary 4

2 Introduction 8

3 Major conclusions and recommendations of the 2010 study 9

3.1 Generation efficiency 9

3.2 Technical and Non-Technical Losses 10

4 Current situation 11

4.1 Generation efficiency 11

4.2 Technical and Non-Technical Losses 11

5 Feasible options for improvement 13

5.1 Generation efficiency 13

5.2 Technical and Non-Technical Losses 13

6 Proposals 15

6.1 Proposal for purchasing a new diesel fired generating unit for Kosrae Utility Authority with a capacity of 1.2 MW 15

6.2 Proposal for optimizing fuel injection at existing, operable generating units 17

6.3 Proposal for purchasing Easy Power software for performing grid calculations 18

1 Executive Summary

Background

In 2010 KEMA Inc. hadconducted the project “Quantification of Energy Efficiency in the utilities of the US Affiliate States” for the Pacific Power Association, according to a contract between the PPA and KEMA Inc, signed on January 22, 2010. For each of these utilities a Report hadbeen provided containing (among others) study results and recommendations including a prioritized list of power system equipment needing to be replaced to reduce energy losses, while for each of the utilities a Data Handbook hadbeen provided, a grid model in Easy Power (a grid modeling tool) and Loss Calculation Sheets.

InJuly2012,the PPA received arequestforassistancefromthefourFSMUtilitiesto conductafollowupstudytothe Power System LossStudiescompletedin2010todiscusstherecommendationsofeachutility’sreportanddeveloplossreductionstrategiesaswellasformulatingproposalswithcostbenefitanalysisfortherecommendations.
Theseproposalscanthenbe usedbytheutilitiestoapproachdevelopmentpartnerstofundtheselossreductioninitiatives.DNV KEMA was contractedto undertake these services as requested.

Major findings

For the follow-up study as described DNV KEMA has visited KUA in Kosrae on August 26, 2013.

The maximum load was still – after 3 years - around 1.1 MW, but the energy consumption has slightly gone down since the rate increase implemented in April 2013.

Generation efficiency went up from 13.75 kWh/gallon in 2010 to 14.85 kWh/gallonat the time of this study. The only difference between the situation in 2010 and 2013 is that production in 2010 was done with a different mix of generators, with G-8 as the dominant generator, while now in 2013 the generation mix of G-4 and G-6 is being used which apparently brings a better efficiency. G-8 is currently inoperable and waits repair.

Part of the recommendation in 2010 was to install a new generator with a capacity that makes it possible to run at a more efficient level given the current load, which will not change much in years to come. With a new generator of 1.2 MW fuel efficiency can go up to 15.8 kWh/gallon. With an annual production of 6,000,000 kWh and a fuel price of $ 4.50 per gallon, this will bring savings of $ 109,000 per year when comparing with the current fuel efficiency of 14.85 kWh/gallon. Since the fuel efficiency in 2010 was only 13.75 kWh/gallon, the savings would have been $ 254,000 per year. The cost of a new generating unit 1.2 MW including site construction costs is estimated at $ 1,400,000, which means that payback time will be 17years based on the fuel efficiency in the current situation and < 7 years based on the fuel efficiency in 2010, taking an interest rate of 6% into account and assuming a fuel price of $ 4.50 per gallon per year. In case it is assumed that the fuel price will increase with 3% per year the pay-back times will be 15.5 years (based on current situation) and 6 years (based on situation in 2010).

An important aspect is - as further indicated in section 4 of this report – that the new generator is not only needed for the reason of increased efficiency but also for getting KUA back to at least an n-1 situation (a situation where the peak load can still be served if the largest generator is out of service), with which also the amount of non-served energy will be reduced due to a better availability of production units.

Finally it will be recommended to have all generators that will keep on playing a role in KUA’s production being treated towards optimal fuel injection (and this way increased efficiency) by a specialized firm like for example Baileys Diesel Group.

Power station losses did not change (4.98 now 5.00) but system losses (including street lighting) went up from 11.76% to 13.51%. Since the technical infrastructure and the load did not change, the technical losses should still be at the level of 2010 (5.91%) so the surplus as measured in 2013 is attributed tonon-technical losses. Non-technical losses went up this way from 3.27% to 5.02%.

A possible reason for this increase could be the fact that 300 of the 1500 cash power meters (20%) have been malfunctioning and needed to be replaced. By the end of August 2013 300 new meters have arrived at KUA and will be installed to replace the malfunctioning meters. It is advised to check out the loss figures again after replacement. If non-technical losses go back to around 3% again, this can be considered a reasonably low figure, also when taking into account that energy meters are not revenue class and have an impact on calculated losses due to inaccuracy.

Some recommendations of 2010 have been followed up, like purchasing an infrared camera for identification of hot spots, rotating distribution transformers for reducing situations with oversized distribution transformers.

The power factor at KUA is around 0.96 which is quite well and does not call for the need of installing capacitor banks.

When looking at transformer losses it can be seen that the power transformers at the KUA power station are oversized (both transformers have a capacity of 2.5 MVA while the maximum load is only 1.1 MW at a rather high power factor of 0.96). When looking at savings on core losses when replacing the two transformers by two new ones at 1.6 MVA each forReplacement Costs New of around $ 80,000the savings on losses would be around $ 5,000 per year and as such the payback time would run over 20 years which unfortunately is not a feasible option.

Proposals

As already indicated there is a number of feasible investments in efficiency improvement justified by a reasonable pay-back time. Proposals for these investments are attached to this Report and are related to:

- New generator 1.2 MW

- Fuel injection service

Finally it should be mentioned that possibly the peak load will go up with some 0.35 to 0.5 MW in the coming years (new hospital, Chinese fishing company), that additional solar will be introduced power (now 50 kW), and that KUA is waiting for wave power funding. It will be important for KUA to be able to do load flow and short circuit calculations as well as protection studies in grid modeling software Easy Power for calculating the impact of these changes, also in order to see the impact on losses and to identify options for keeping losses at an optimal level. Last but not least the grid modeling software will be needed for performing dynamic simulations in order to see the impact of the planned renewables on the grid stability and to identify options for keeping up grid stability.

In the light of these new developments it is recommended that KUA will also have grid calculation software available to anticipate on future load flow and stability issues ant to at the same time have a tool for better calculating losses situations. Buying Easy Power, inclusive of training, as recommended for the other FSM utilities for the reason of getting better hold on efficiency issues, is at the same time recommended for KUA, particularly while a discount of 10% may be possible if all 4 FSM utilities buy Easy Power.

Savings and pay-back times of the proposals for KUA are summarized below.

PROPOSAL / COST (US$) / SAVINGS/YEAR (US$) / PAY-BACK TIME
New generator 1.2 MW / 1,400,000 / $ 109,000 to $ 254,000 dependent on available generators / Between 7 to 17 years
Fuel injection services / 75,000 / Estimated $ 25,000 per year / Around 5 years
Easy Power software and training / 37,500 / Save on consulting costs for calculating impact on losses, grid situations and stability in case of extensions and integration of renewables / Most probably fully paid back per the first study needed

2 Introduction

In 2010 KEMA Inc.hadconducted the project “Quantification of Energy Efficiency in the utilities of the US Affiliate States” for the Pacific Power Association, according to a contract between the PPA and KEMA Inc, signed on January 22, 2010. For each of these utilities a Report hadbeen provided containing (among others) study results and recommendations including a prioritized list of power system equipment needing to be replaced to reduce energy losses, while for each of the utilities also a Data Handbook hadbeen provided, a grid model in Easy Power (a grid modeling tool) and Loss Calculation Sheets.

The four electric utilities in the Federated States of Micronesia that were studied under this assignment are:

1. Chuuk State Public Utilities Corporation (CPUC)
2. Kosrae Utility Authority (KUA)
3. Pohnpei Utility Corporation (PUC)
4. Yap State Public Service Corporation (YSPSC)

After receipt of the Reports and other deliverables these utilities felt the need for having further clarifications on the Report and particularly on the conclusions and recommendations, including recommendations on suggested measures to be taken and equipment to be purchased or replaced, all with the purpose of increasing efficiency.

Theseproposalscanthenbe usedbytheutilitiestoapproachdevelopmentpartnerstofundtheselossreductioninitiatives.

DNV KEMA was contractedto undertake these services as requested.

3 Major conclusions and recommendations of the 2010 study

The major conclusions and recommendations of the 2010 Study for KUA are summarized in headlines in this chapter.

3.1 Generation efficiency

On Generation Efficiency we found in 2010 an average fuel consumption of 13.75 kWh/gallon, while generation was only done by unit G-8 (1050 kW) with units G-4 (580 kW) and G-6 (1500 kW) being standby. G-2 (580 kW), manufactured in 1980, was for repair (and still is) and should be retired like G-5 (1984) which is not operational for a long time already. Generator G-7 (1650 kW) was (and is still) for repair and needs a new stator, which would cost around $ 100,000 while the generator would not run very efficient after repair, given its capacity of 1650 kW.

In our Report of 2010 we already mentioned that the fuel efficiency of G-8 is 13 kWh/gallon only, while G-6 has a fuel efficiency of 15 kWh/gallon. This explains why fuel efficiency in 2013 (with G-6 as the dominant generator) is much better than it wasin 2010. G-8 is currently for repair and needs a new stator.

The most optimal fuel efficiency could be reached when purchasing a new generating unit of 1.2 MW, but the payback time would be long (> 10 years), based on a fuel price in those days of $ 3.50 per gallon.

Furthermore we mentioned power station losses in our 2010 report of 4.98% and recommended some further investigations on reduction of power station losses, if possible down to 4%.

3.2 Technical and Non-Technical Losses

When looking at the KUA statistics we found in 2010 that power delivery by KUA for Street Lighting was included in the figure of 11.76% for System Losses. This is because individual streetlights are not metered. We found that 2.58% of all power entering into the grid is used for street lighting, which is partly paid, which means that KUA has a financial loss on power delivery for street lighting.

The real System Losses were for that reason 11.76% - 2.58% = 9.18%. Our losses calculations showed that 5.91% were technical losses and the remainder of 3.27% was for non-technical losses. These are quite good figures for system losses, but our report gave a number of recommendations for further improvement of these losses figures, such as better sizing of distribution transformers, avoiding unbalance in the feeder phases, identifying hot spots with infrared equipment, monitoring the power factor while no direct measures were recommended for the power factor of 0.96 as identified in 2010.

For non-technical losses it was recommended in 2010 to resolve problems with Cash Power meters as soon as possible (customers could for example re-enter tokens) and to audit the process from meter reading to billing in order to identify any hidden irregularities and furthermore some recommendations were given for combating theft, meter tampering, etc.

4 Current situation

4.1 Generation efficiency

As already mentioned in section 3.1 it was found in 2010 that G-8 (1050 kW) had a lower fuel efficiency than G-6 (1500 kW, currently running as the baseloadgenerator for a maximum load of 1.1 MW).

After 2010 G-8 had to be put out of service because of a stator problem. KUA found funding with help of the Micronesian Government to replace the stator of G-8. The replacement will take place by the end of this year 2013.

KUA will then more or less be back in an n-1 situation, which means that if the largest generator trips still another generator will be available to cover the full load, if needed together with the still running, smaller unit, G-4.

It has anyway shown that with G-6 as the baseloadgenerator the fuel efficiency has increased from 13.75 to 14.85 kWh per gallon. This saves KUA $ 145,000 per year, based on a yearly production of 6,000,000 kWh and a current price for diesel # 2 of $ 4.50 per gallon.

Power station own usage remained almost the same and it is still recommended to perform an energy audit in the power plant with all its auxiliaries.