Review of the vesting contract parameters for 2013 and 2014 – Draft report LONG RUN MARGINAL COST (LRMC) PARAMETERS Ver. 6 (Draft) 31 May 2012
Review of the vesting contract parameters for 2013 and 2014 – Draft report
LONG RUN MARGINAL COST (LRMC) PARAMETERS
Ver. 6 (Draft) 31 May 2012
The SKM logo trade mark is a registered trade mark of Sinclair Knight Merz Pty Ltd.
Review of vesting contract parameters for 2013 and 2014 – Draft report
LONG RUN MARGINAL COST (LRMC) PARAMETERS Ver. 6 (Draft) 31 May 2012
SKM (Singapore) Pte. Ltd. UEN 198 905 658K A subsidiary of Sinclair Knight Merz 80 Marine Parade Road #18-01/04 Parkway Parade, Singapore 449269 Tel: +65 6345 3055 Fax: +65 6344 8441 Web: www.globalskm.com
COPYRIGHT: The concepts and information contained in this document are the property of SKM (Singapore) Pte. Ltd., a subsidiary of Sinclair Knight Merz. Use or copying of this document in whole or in part without the written permission of SKM (Singapore) Pte. Ltd., a subsidiary of Sinclair Knight Merz constitutes an infringement of copyright.
LIMITATION: This report has been prepared on behalf of and for the exclusive use of SKM (Singapore) Pte. Ltd., a subsidiary of Sinclair Knight Merz’s Client, and is subject to and issued in connection with the provisions of the agreement between SKM (Singapore) Pte. Ltd., a subsidiary of Sinclair Knight Merz and its Client. SKM (Singapore) Pte. Ltd., a subsidiary of Sinclair Knight Merz accepts no liability or responsibility whatsoever for or in respect of any use of or reliance upon this report by any third party.
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Contents 1. Executive summary 1
1.1. Introduction 1 1.2. LRMC technical parameters 1 1.3. Review of scale factor indices 1
2. Introduction 3 2.1. General 3 2.2. Financial parameters 3 2.3. Disclaimer 6
3. Performance parameters 7 3.1. Introduction 7 3.2. Generating technology 8 3.3. Capacity per generating unit 11 3.4. Impact of Gas Compression and Resulting Net Capacity 14 3.5. Heat Rate 17
4. Capital cost 21 4.1. Introduction 21 4.2. Method 21 4.3. Initial capital cost 24 4.4. Through-life capital costs 26 4.5. Land and Site Preparation Cost 26 4.6. Connection Cost 27 4.7. Owner's costs after financial closure 28 4.8. Owner's costs prior to Financial Closure 30
5. Operating costs 31 5.1. Fixed annual running cost 31 5.2. Variable non-fuel cost 35
6. Other parameters 37 6.1. Build duration 37 6.2. Economic life 37 6.3. Average expected utilisation factor 37
7. Results – vesting contract parameters 38 7.1. Introduction 38 7.2. Summary of technical parameters 38 7.3. Calculated LRMC 39
SKM (Singapore) Pte. Ltd.
PAGE i
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
SKM (Singapore) Pte. Ltd. PAGE ii
8. Review of Scale Factor Indices 40 8.1. Introduction 40 8.2. SKM Consideration 41 8.3. Other alternatives 43
Appendix A Prescribed procedures 46 Appendix B Market modelling 51 Appendix C Technical performance data 53
Thermodynamic analysis 53
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
1. Executive summary 1.1. Introduction
Sinclair Knight Merz (SKM) has been engaged by the Energy Market Authority (EMA) to provide recommended values for the technical parameters of the Vesting Contracts for electricity generation in Singapore for the period 2013 and 2014, and to review the Scale Indices method within the Vesting Contract Procedures.
1.2. LRMC technical parameters
The following values are recommended by SKM for use in the Vesting Contract parameters for 2013-14.
Table 1 Summary of recommended technical parameters
Item Parameter 2013-14 Value 6 Economic capacity of the most economic technology in
operation in Singapore (MW) 383.47 MW net at 32oC
7 Capital cost of the plant identified in item 6 ($US/kW) 1004.21 USD/kW
8 Land, infrastructure and development cost of the plant identified in item 6 ($Sing million)
SGD 147.81M
11 HHV Heat Rate of the plant identified in item 6 (Btu/kWh) 6886 btu/kWh net HHV
12 Build duration of the plant identified in item 6 (years) 2.5 years
13 Economic lifetime of the plant identified in item 6 (years) 20 years
14 Average expected utilisation factor of the plant identified in item 6, i.e. average generation level as a percentage of capacity (%)
72.8%
15 Fixed annual running cost of the plant identified in item 6 ($Sing)
18.297M SGD
16 Variable non-fuel cost of the plant identified in item 6 ($Sing/Mwh)
5.21 SGD/MWh
1.3. Review of scale factor indices
The formulation of the index forecast for quarter D+1 using the trend from quarter D-2 to D does potentially introduce volatility to the calculation of the cost indices. Where the index growth of two consecutive quarters exhibit either high positive or high negative numbers, the forecast for D+1 quarter will exacerbate the trend. Also the use of quarterly data that constantly moves while the base quarter is fixed means that the trend since the base quarter is not necessarily reflected. Using a longer period between quarters however is likely to reduce volatility. Fixing the index at the base quarter plus using the trend going forward from the base quarter is likely to both reduce volatility as well as reflect the trend in cost movements since the base quarter.
SKM (Singapore) Pte. Ltd. PAGE 1
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Alternatives to reduce volatility that the EMA may be considered include:
• removing indexation or reducing the frequency of the adjustments (eg. Annual adjustments rather than quarterly adjustments)
• using a less volatile index (eg the MAS core inflation index) instead of the CPI/DSPI
• undertaking an annual review of capital cost
Given the dissatisfaction expressed by various parties regarding the volatility of vesting prices due to the indexation process, SKM recommends that changes be made to the method of determining the LRMC Scale Factor Indices within the two year price period. We recommend that for the overhead cost component, quarterly indexation be removed and replaced with an annual adjustment to reflect forecast inflation over the year.
For capital costs, if the EMA is of the opinion that the cost of undertaking an annual review is justified by the likely savings due to the use of an unsuitable index, SKM recommends that an annual review be made of the capital cost components.
SKM (Singapore) Pte. Ltd. PAGE 2
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
2. Introduction 2.1. General
The Energy Market Authority (EMA) has implemented Vesting Contracts to control market power of generation companies in the National Electricity Market of Singapore. The parameters for setting the Vesting Price associated with these contracts are to be reviewed every two years. The current review relates to the setting of these parameters for 1 January 2013 through to 31 December 2014.
EMA has engaged Sinclair Knight Merz (SKM) to:
• Conduct a comprehensive review and recommend the value of each vesting contract parameter (items 6 through 8 and 11 through 16 in section 2.3 of the Vesting Contract Procedures) for the setting of the vesting price for the period 1 January 2013 to 31 December 2014; and
• Review and recommend improvements to the method for calculating the LRMC Scale Factor Indices (set out in Section 3.8 of the Vesting Contract Procedures), taking into consideration the objective to reduce the quarterly volatility of the vesting price.
This review of the vesting contract parameters follows the method adopted by SKM (as sub-consultant to PA Consulting) in the review of parameters for the period 1 January 2011 to 31 December 2012 (the “2011-12” review)1.
The parameters of the Vesting Contract determine the Vesting Price associated with these contracts and are reviewed every two years, covering the subsequent two-year period. The fifth of these two yearly reviews is the subject of this project, covering the period 1 January 2013 to 31 December 2014.
2.2. Financial parameters
Financial parameters are to be provided by EMA and will be updated prior to the final report.
For the purposes of the Draft report, financial parameters have been provided by EMA and recent market observations by SKM.
1 PA Consulting Group “Review of the Parameters for Setting the Vesting Contract Price for 2011 and 2012 – Final Report”, 27 September 2010.
SKM (Singapore) Pte. Ltd. PAGE 3
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Table 2 Finance parameters applied (pending EMA recalculation)
Parameter Value Notes WACC 5.26% post-tax, nominal
3.20% pre-tax, real Nominal rate advised by EMA
CPI 3.03% Average year-on-year core inflation, Dec 2011, Jan 2012, Feb 2012. Refer Figure 1.
Gas price $22.100 SGD/GJ (PNG) Advised by EMA. For the PNG price, EMA have used the average quarterly forward HSFO prices and USD/SGD exchange rates for the period Dec 11 to Feb 12
Exchange rates 1.2764 SGD/USD 1.6716 SGD/EUR
Average bid and ask, daily, 1/12/2011 to 29/2/2012. Refer Figure 2 and Figure 3.
Figure 1 Singapore CPI data2
‐
1.0
2.0
3.0
4.0
5.0
6.0
Year on year %
CPI (Y‐o‐Y)
MAS core CPI (Y‐o‐Y)
2 Monthly data Department of Statistics, Singapore, http://www.singstat.gov.sg/news/news/cpimar2012.pdf and earlier editions
SKM (Singapore) Pte. Ltd. PAGE 4
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Figure 2 Fx data Dec 2011 to Feb 2012
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
1/12/2011 1/01/2012 1/02/2012 1/03/2012
fx reate
SGD/USD
SGD/EUR
SKM (Singapore) Pte. Ltd. PAGE 5
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Figure 3 Foreign exchange rate trends3
0.00
0.50
1.00
1.50
2.00
2.50
Jan 2007 Jul 2007 Jan 2008 Jul 2008 Jan 2009 Jul 2009 Jan 2010 Jul 2010 Jan 2011 Jul 2011 Jan 2012
fx rates, daily
USD/SGD
EUR/SGD
2.3. Disclaimer
This report has been prepared for the benefit of EMA for the purposes of setting the vesting contract price for the 2013 to 2014 period. This report may not be relied upon by any other entity and may not be relied upon for any other purpose.
3 Data based on Reserve Bank of Australia implied cross rates
SKM (Singapore) Pte. Ltd. PAGE 6
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
3. Performance parameters 3.1. Introduction
Parameters for the existing generation fleet in Singapore4 are shown in Table 3.
Table 3 Registered capacity, large CCGT units
Large CCGT units Reg Cap, MW
Date Licence
Tuaspring Pte Ltd HFLXCC‐1 411 2H 2014 EMA/GE/015
SNK CCP 1 (Senoko) 425 1996 EMA/GE/012
SNK CCP 2 (Senoko) 425 1996 EMA/GE/012
SNK CCP 3 (Senoko) 365 2002 EMA/GE/012
SNK CCP 4 (Senoko) 365 2004 EMA/GE/012
SNK CCP 5 (Senoko) 365 2004 EMA/GE/012
SNK CCP 6 (Senoko) 431 2H 2012 EMA/GE/012
SNK CCP 7 (Senoko) 431 2H 2012 EMA/GE/012
SembCorp Cogen SKACCP1 392.5 2001 EMA/GE/004
SembCorp Cogen SKACCP2 392.5 2001 EMA/GE/004
SembCorp Cogen SKACCP3 400 1Q 2014 EMA/GE/004
SembCorp Cogen SKACCP4 500 TBA EMA/GE/004
TUAS Stage 2 CCP1 367.5 2001 EMA/GE/009
TUAS Stage 2 CCP2 367.5 2002 EMA/GE/009
TUAS Stage 2 CCP3 367.5 2005 EMA/GE/009
TUACCP4 367.5 2005 EMA/GE/009
Power Seraya CCP1 368 2002 EMA/GE/016
Power Seraya CCP2 364 2002 EMA/GE/016
Power Seraya CCP3 370 2010 EMA/GE/016
Power Seraya CCP4 370 2010 EMA/GE/016
Keppel Merlimau Cogen GRF 3 420 1Q 2013 EMA/GE/006
Keppel Merlimau Cogen GRF 4 420 3Q 2013 EMA/GE/006
GMR Unit 1 400 Nov‐13 EMA/GE/005
GMR Unit 2 400 Jan‐14 EMA/GE/005
4 http://www.ema.gov.sg/page/115/id:129/
SKM (Singapore) Pte. Ltd. PAGE 7
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
The technical performance parameters for the notional new entrant plant are estimated in this Section.
3.2. Generating technology
The parameters for the existing relevant power stations in Singapore, extracted from the 2010 report for 2011-12, and updated with the selected gas turbines for plants recently committed, are given in Table 4:
Table 4 Existing Singapore station parameters (large CCGT units)5
Power station Train capacity
MWe
Number of trains
Total station Frame F
capacity MWe
CCGT technology
GT type Original Equipment
Manufacturer (OEM)
Senoko Converted CCGT
365 3 1095 Type F GT26 Alstom
Senoko repower (CCP6&7)
431 2 862 Type F M701F Mitsubishi
TUAS CCGT 367.5 4 1470 Type F M701F Mitsubishi
Seraya CCGT 368 364 370 370
4 1472 Type F V94.3A (SGT5-4000F)
Siemens
Sembcorp Cogen6 392.5 2 785 Type F 9FA General Electric
Sembcorp cogen 400 1 (committed)
400 Type F GT26 Alstom
Keppel Merlimau 420 2 840 Type F GT26 Alstom
GMR 400 2 800 Type F SGT5-4000F
Siemens
5. KEMA 2009 op cit. Adjustments based on licensed capacity (EMA) as per Table 3 and as updated by SKM 6 Evaluations have been made based on CCGT performance only
SKM (Singapore) Pte. Ltd. PAGE 8
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
The Vesting Contract procedures published by EMA7 indicate that:
The EMA implemented Vesting contracts on 1 January 2004 as a regulatory instrument to mitigate the exercise of market power by the generation companies (“Gencos”). Vesting Contracts commit the Gencos to sell a specified amount of electricity (viz the Vesting Contract level) at a specified price (viz the Vesting Contract price). This removed the incentive for Gencos to exercise their market power by withholding their generation capacity to push up spot prices in the wholesale electricity market. Vesting Contracts are only allocated to the Gencos that had made their planting decisions before the decision was made in 2001 to implement Vesting Contracts.
And:
The Allocated Vesting Price approximates the Long Run Marginal Cost (LRMC) of a theoretical new entrant that uses the most economic generation technology in operation in Singapore and contributes to more than 25% of the total demand.
The underlying concept of LRMC is to find the average price at which the most efficiently configured generation facility with the most economic generation technology in operation in Singapore will cover its variable and fixed costs and provide reasonable return to investors. The plant to be used for this purpose is to be based on a theoretical generation station with the most economic plant portfolio (for existing CCGT technology, this consists of 2 to 4 units of 370MW plants). The profile of the most economic power plants is as follows:
• Utilises the most economic technology available and operational within Singapore at the time. This most economic technology would have contributed to more than 25% of demand at that time.
• The generation company is assumed to operate as many of the units of the technology necessary to achieve the normal economies of scale for that technology.
• The plants are assumed to be built adjacent to one another to gain infrastructure economies of scale.
7 Energy Market Authority, "EMA's procedures for calculating the components of the vesting contracts", March 2011, Version 1.7
SKM (Singapore) Pte. Ltd. PAGE 9
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
• The plants are assumed to share common facilities such as land, buildings, fuel supply connections and transmission access. The cost of any common facilities should be prorated evenly to each of the plants.
• The plants are assumed to have a common corporate overhead structure to minimise costs. Any common overhead costs should be prorated evenly to each of the plants.
SKM believes that the technology that should be selected according to these criteria would be CCGT units based on "F" class gas turbines. The existing large CCGT/Cogen plants in Singapore are based on "F" class gas turbine technology (refer Table 4).
SKM expects that any new plant in Singapore would be optimised for performance at the site Reference Conditions. For this review it is taken that the site Reference Conditions are the all-hours average conditions of:
• 29.5ºC dry bulb air temperature,
• 85% Relative Humidity (RH);
• Sea-level;
• 28ºC cooling water inlet temperature.
Operation at other ambient or sea water conditions represents off-design operation. This includes operation at the ambient conditions specified in the Singapore Market Manuals for the Maximum Generation Capacity, which includes an ambient temperature of 32ºC. Consistent with the treatment in 2010 for the 2011-12 review, a correction factor for the plant's capacity to 32ºC has been applied.
As shown in Table 4, the Singapore market includes "F" class units from each of the following OEMs8:
• Alstom;
• Siemens;
• General Electric (GE); and
• Mitsubishi.
8 Original Equipment Manufacturers
SKM (Singapore) Pte. Ltd. PAGE 10
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
The market for supply of such plants is competitive and it generally cannot be determined, without competitive bidding for a specific local project, which design is the most economic generation technology on an LRMC basis for new built plant. It is often the case for example that the configuration offered with the lowest heat rate is the bid with a higher capital cost. In order to model the performance of the most economic generator it is therefore considered appropriate to consider the performance of all these OEM's appropriate "F" class CCGT configurations and to use an arithmetic average of the performance parameters of each of these OEMs' plants in CCGT configuration9.
In order to estimate these performance parameters, the GTPro/GTMaster10 (Version 22 Release dated 16 March 2012) thermodynamic analysis software suite was applied. Representative schematics of the resulting configurations are shown in Appendix C.
3.3. Capacity per generating unit
The generation capacities of new entrant CCGT configurations, on a clean-as-new condition, and at the Reference Conditions of 29.5ºC are given in Table 5. Note that upgrades of gas turbine technologies occur frequently and judgement must be applied as to whether a new entrant developer would choose the very latest announced version for a project in Singapore or not. In this review SKM has decided not to apply the very latest announced models of the Mitsubishi gas turbine (the 701F5) and the Alstom GT26 2011 upgrade but to instead select the variants that have been available in the market for longer (considering commercial operating experience).
Table 5 Generation capacity of new entrant CCGT units (clean-as-new at Reference Conditions, excluding gas compression impacts)
Configuration Gross MW Net MW
Frame 9FB 406.0 397.5 M701F 432.4 423.7 GT26 416.1 407.6 SGT5-4000F 389.5 381.7 Average 411.0 402.6 This thermodynamic modelling includes all corrections necessary for:
• Ambient conditions of 29.5ºC;
• Boiler blow-down; and
• Step-up transformer losses.
9 It is noted that the Sembcorp plant is a cogeneration plant. In evaluating the parameters for this review the performance of plants in CCGT configuration only has been applied.
10 TM, Thermoflow, inc
SKM (Singapore) Pte. Ltd. PAGE 11
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
No further allowances need to be made for these factors except as discussed below regarding ambient temperature.
The impact of gas compression requirements is discussed separately below (Section 3.4).
The capacities and heat rates of operating gas turbine and CCGT power plants degrade from the time the plant is clean-as-new11. The primary drivers for performance degradation are fouling, erosion and roughening of the gas turbine compressor and material losses in the turbine section. A CCGT plant has a slightly reduced degradation profile than a simple cycle gas turbine installation due to partial recovery of this effect in the steam cycle, and that the gas turbine only comprises approximately 2/3 of the plant output. This degradation effect is typically described as having two components:
• "Recoverable" degradation is degradation of performance that occurs to the plant that can be recovered within the overhaul cycle. Recoverable degradation can be substantially remediated by cleaning of air inlet filters, water washing of the gas turbine, ball-cleaning of condensers and the like. These cleaning activities are typically undertaken several or many times within a year depending on the site characteristics and the economic value of performance changes; and
• "Non-recoverable" degradation is caused by the impacts of temperature, erosion and corrosion of parts within the plant. This type of degradation is typically substantially remediated over the overhaul cycle of the plant as damaged parts are replaced with new parts. Because the typical industry repair philosophy uses an economic mix of new and refurbished parts within overhauls, it is typically the case that not all of the original clean-as-new performance is recovered at the overhauls.
The average capacity reduction due to recoverable degradation is estimated at 1%. That is, the degradation amount varies from approximately zero to approximately 2% over the cleaning cycle.
Additional to this, an allowance for the non-recoverable degradation of capacity should be made. These typically have the form similar to that shown in Figure 4.
11 Refer GE publication “Degradation curves for Heavy Duty Product Line Gas Turbines” for example
SKM (Singapore) Pte. Ltd. PAGE 12
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Figure 4 Form of CCGT recoverable and non-recoverable degradation
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
4.5%
0 5 10 15 20 25
Degrada
tion
from
clean
‐as‐‐new
Years
Power degr
HR degr
Based on plants operating up to 93.2% of hours in the year12, the degradation allowance amounts for average capacity degradation rate over the plant's life of 3.05% is suggested (calculated as a weighted average using the pre-tax real discount rate to weight each year in the plant’s life). Note that the average capacity degradation is not materially affected by the OEM's nomenclature for describing major outages for the gas turbine units. Degradation rates are dominated by compressor fouling rates and the timing of major compressor refurbishments and scouring, similar between OEM's.
Variations in ambient temperature affect the capacity of the generating units. The modelled impacts of variations in ambient temperature on the new entrant configurations and the average impact across the four modelled configurations are shown in Table 6 and Figure 5.
12 Which is the estimated Available Capacity Factor for the plant, from the review for the 2012-13 period
SKM (Singapore) Pte. Ltd. PAGE 13
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Table 6 Variation in net power output with ambient temperature (relative to Reference Conditions)
Config. Ambient temperature (dry bulb), ºC
0 5 10 15 20 25 30 35 40
GT26 108% 107% 106% 104% 103% 102% 99% 97% 94% Frame 9FB 110% 110% 109% 108% 105% 103% 100% 95% 89% 701F 112% 110% 108% 106% 104% 102% 100% 98% 95% SGT5-4000F
110% 110% 109% 108% 105% 103% 100% 97% 94%
Average 110% 109% 108% 107% 104% 102% 100% 97% 93% Figure 5 Effect of ambient temperature on power output
80%
85%
90%
95%
100%
105%
110%
115%
120%
0 5 10 15 20 25 30 35 40
Power, %
of P
ower at R
eferen
ce Con
ditio
ns
Ambient dry bulb temperature
GT26
9FB
701F
4000F
Average
The correction factor for operation at 32ºC relative to the Reference Conditions of 29.5ºC is a reduction in capacity of 1.48% (averaged over the four models), or 5.98MW. Note that for variations of ambient relative humidity between 75% and 95% there is negligible difference in the performance of CCGT plants with once-through cooling.
3.4. Impact of Gas Compression and Resulting Net Capacity
Gas compression is now required for new entrant “F” class CCGT plants in Singapore.
SKM (Singapore) Pte. Ltd. PAGE 14
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Three of the CCGT configurations noted use natural gas at approximately 35 Barg and one configuration (the GT26) uses natural gas at approximately 50 Barg at the site boundary. The gas compressor power requirements calculated for the relevant gas turbines at varying site boundary gas pressures are shown in Figure 6. Allowances are made for pressure losses between the site boundary and the gas turbine unit.
Figure 6 Gas compressor power requirements for relevant gas turbines
‐
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
20 21 22 23 24 25 26 27 28 29 30
Gas com
pressor po
wr, per unit, kW
Gas pressure at site boundary, Barg
GE, Mitsubishi, Siemens
Alstom
Data for gas pressures in the TUAS area of Singapore is shown in Figure 7, for the months of January 2011 to May 2012. The Network 1 pressure may be downstream of a regulator in which case the upstream pressure will be higher.
SKM (Singapore) Pte. Ltd. PAGE 15
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Figure 7 Gas pressures in TUAS area, 2011 to May 2012
0
5
10
15
20
25
30
35
40
45
Jan
2011
Feb
2011
Mar
2011
Apr
2011
May
2011
Jun
2011 Jul
2011
Aug
2011
Sep
2011
Oct
2011
Nov
2011
Dec
2011
Jan
2012
Feb
2012
Mar
2012
Apr
2012
May
2012
System
pressures, Barg
Network 1 Tuas Power Inlet Pressure (Barg)
Network 2 Tuas Gatepost Pressure (Barg)
The data indicates that gas compression is sometimes required under current conditions. Should the system pressures reduce further (e.g. because of load growth) then gas compression would be required more often13.
For the purposes of this review it is assumed:
• Gas compressors would be incorporated in a new plant in the TUAS vicinity;
• The specification of the compressors would allow for further reductions in local gas pressures from those presently seen. It is assumed they would be capable of operating from a site boundary gas pressure of 20 Barg; and
• The average pressure at the site boundary during operation is 31.7 Barg in the relevant period, being the average pressure in the Network 2 from Jan 2010 to date.
13 The introduction of LNG should support local gas pressures. LNG re-gasification plants necessarily incorporate gas compression.
SKM (Singapore) Pte. Ltd. PAGE 16
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
On this basis the calculated average gas compressor auxiliary/parasitic load impact is 0.909 MW per unit based on the averaged pressure requirements of the four gas turbine models under consideration.
The resulting net capacity calculation after considering the above is shown in Table 7.
Table 7 Generation capacity of new entrant CCGT units
Parameter/factor MW
Gross capacity (clean-as-new, reference conditions) 411.0
Less parasitics = net capacity at Reference Conditions (clean-as-new) -8.4 = 402.6
Less allowance for gas compression -0.909
-5.978 Adjust for 32ºC maximum registered capacity (-1.48%)
Adjust for average degradation (-3.05%) -12.278
Net capacity 383.47
3.5. Heat Rate The heat rates of new entrant CCGT configurations, on a clean-as-new condition, and at the Reference Conditions of 29.5ºC are given in Table 8.
Table 8 Heat rate of new entrant CCGT units (clean-as-new at Reference Conditions excluding gas compression)
Configuration Net HR, LHV, GJ/MWh
Net HR, HHV, GJ/MWh
Net HR, LHV, Btu/kWh
Net HR, HHV, Btu/kWh
Frame 9FB 6.295 6.981 5.967 6.617 M701F 6.344 7.035 6.013 6.669 GT26 6.263 6.946 5.936 6.584 SGT5-4000F 6.274 6.958 5.947 6.595 Average 6.294 6.980 5.966 6.616 This thermodynamic modelling includes all corrections necessary for:
• Ambient conditions of 29.5ºC;
• Boiler blow-down; and
• Step-up transformer losses.
No further allowances need to be made for these factors except as discussed below regarding ambient temperature and gas compression impacts.
As noted in Section 3.3 above, heat rates for CCGT plants are also subject to degradation. A weighted average heat rate degradation over the plant's life of 1.89% is estimated (weighted by the pre-tax real discount factor for each year).
SKM (Singapore) Pte. Ltd. PAGE 17
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Variations in ambient temperature affect the heat rates of the generating units. The modelled impacts of variations in ambient temperature on the new entrant configurations and the average impact across the four modelled configurations are shown in Table 9 and Figure 8.
Table 9 Variation in net heat rate with ambient temperature (relative to Reference Conditions)
Ambient temperature (dry bulb), ºC
Config. 0 5 10 15 20 25 30 35 40
GT26 100.6% 100.4% 100.2% 100.1% 100.0% 100.0% 100.0% 100.0% 100.3%
Frame 9FB 101.1% 100.7% 100.3% 100.0% 99.9% 99.9% 100.0% 100.4% 101.4%
701F 100.5% 100.4% 100.3% 100.3% 100.2% 100.1% 100.0% 100.1% 100.2%
SGT5-4000F 101.8% 101.3% 100.8% 100.3% 100.2% 100.1% 100.0% 100.0% 100.2%
Average 101.0% 100.7% 100.4% 100.2% 100.1% 100.0% 100.0% 100.1% 100.5%
Figure 8 Impact of ambient temperature on heat rate
95%
100%
105%
0 10 20 30
HR, % of H
R at Referen
ce Con
dition
s
Ambient dry bulb temperature
40
GT26
9FB
701F
4000F
Average
Note that for variations of ambient relative humidity between 75% and 95% there is negligible difference in the performance of CCGT plants with once-through cooling.
SKM (Singapore) Pte. Ltd. PAGE 18
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
The use of fuel by the plant will reflect average operating conditions and hence the heat rate at the Reference Conditions has been applied. It is not appropriate to consider the Standing Capability Data criterion for capacity (i.e. at 32ºC) to also apply for the plant's heat rate except in as much as it impacts on the average part load factor as discussed below.
Whenever the power plant is operated at less than the Maximum Continuous Rating (MCR) of the plant at the relevant site conditions, the heat rate is affected. The modelled variation in heat rate with the part load factor of the plant is shown in Table 10 and Figure 9
Table 10 Variation of heat rate with part load (%)
Power 55 % 60% 65% 70% 75% 80% 85% 90% 95% 100%
Average HR relative to full load
110.1% 108.3% 106.7% 105.2% 104.0% 102.9% 101.9% 101.2% 100.6% 100%
Figure 9 Variation of heat rate at part load
100%
102%
104%
106%
108%
110%
112%
60% 70% 80% 90% 100%
Heat rate, % of full load HR
Part load
9FB
701F
GT26
4000F
Average
An average load when operating at 86.9% of registered capacity has been applied. This reflects the shared obligations for providing frequency control ancillary services and is consistent with the market modelling in Appendix B.
SKM (Singapore) Pte. Ltd. PAGE 19
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
The apparent part load factor for the plant's performance is slightly reduced since the registered capacity would only be 98.5% of the nominal capacity. The resulting overall part load factor is 85.6% for which the part-load factor for heat rate would be 1.85%.
An additional adjustment is made to reflect the natural gas used in starts through the year14. The gas usage for starts is estimated at 10 hours of full-load operating equivalent, or 0.1%.
In reviews prior to 2010, an additional allowance on account of regulation service is added (+0.5%). It is not considered that the AGC requirement in Singapore is materially different from other jurisdictions where minor perturbations of output on account of AGC (for those units in the system providing AGC service) or on droop-control are part of normal operations for which no specific extra allowance is considered appropriate. Note that the impact of operating the plant at part-load on account of the need for regulation and contingency reserve ancillary services is already accounted for within the load factor correction.
An adjustment is applied for to account for the gas compressor auxiliary load. As noted in Section 3.4, the auxiliary load of the gas compression has an impact on net output and also on net heat rate.
The resulting overall heat rate calculated is shown in Table 11.
Table 11 Heat rate of new entrant CCGT units
Parameter/factor Heat rate Net HR (clean-as-new, reference conditions) - after recognition of parasitic loads
6.980 GJ/MWh HHV
Adjust for overall part load factor (+1.85%) +0.129 Adjust for average degradation (+1.89%) +0.132 Adjust for starts gas usage (+0.1%) +0.007 Adjust for gas compressor impact +0.017 Adjusted heat rate 7.265 GJ/MWh HHV Net HR 6,886 Btu/kWh HHV
14 Based on 16 hot starts, 3 warm starts and 0.5 cold starts in an average year. These exclude starts due to economic shutdowns, the cost of which should be factored into the operator's decision to shut-down.
SKM (Singapore) Pte. Ltd. PAGE 20
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
4. Capital cost 4.1. Introduction
Capital cost includes:
(i) facility costs (ancillary buildings, demineralisation plant, sea water intake/outfall structures, constructing the jetty for emergency fuel unloading facility and gas receiving facilities) classified under land and site preparation cost in previous reviews,
(ii) emergency fuel facilities classified under land and site preparation cost in previous reviews,
(iii) civil works for the plans, erection and assembly, detailed engineering and start-up costs, and contractor soft costs classified under connection cost in previous reviews and
(iv) discounted through life capital cost classified under miscellaneous cost in previous reviews.
4.2. Method
The capital cost of a new entrant CCGT plant using current costs is assessed using the following method.
• SKM has made enquiries to the four OEMs requesting advice on the current specific capital costs (on a greenfields EPC basis) for a specific generic CCGT configuration that SKM use to compare costs between projects and times on a consistent basis. This is based on a “1+1” single shaft “F” class unit with mechanical draft evaporative cooling tower and gas-only fuel. This enquiry was specific for the Singapore region;
• SKM modelled this configuration within the latest version of the PEACE software included with the GTPro software suite noted in Section 3 above and using the current regional cost factors in-built into PEACE for Singapore and other relevant countries;
• SKM are also assisting with other large “F” class project developments in the region and are in discussions (including regarding costs) with OEMs for turnkey supply;
• SKM have considered the latest version of Gas Turbine World Handbook;
• Considering this information SKM assesses that the current EPC cost (excluding connections and on an “overnight basis”) of a "standard" single-unit "F" class CCGT unit for the Singapore location is USD760/kW (based on net ISO output);
• SKM then evaluates whether the regional cost indices within PEACE require adjusting to produce the assessed market EPC specific cost. In the case of the current review no modification was considered to be necessary;
SKM (Singapore) Pte. Ltd. PAGE 21
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
• Models are then established within PEACE for the configurations being evaluated. These include once through cooling, dual fuel burners, gas compression, savings in infrastructure when shared between multiple units and considering the site reference ambient conditions. This produces a capital cost estimate for the basic plant;
• Further calculations are made to estimate costs for the site specific costs not able to be modelled in PEACE by direct calculation or by escalating from the previous review.
This method is consistent with the 2011-12 review.
SKM assesses that the capital costs of large CCGT plants for current procurement have reduced further between the 2011-12 review and this review.
This is notwithstanding that the latest release of the Gas Turbine World Handbook (2012) indicated that it expected prices to rise 5% to 7% relative to 2011 due to the Handbook’s expected firming up of gas turbine orders. SKM considers that the Handbooks are not as directly useful as market soundings and information from other projects are as the Handbook information has a time-delay from the time it was written, it is not geographically specific and scope differences occur between editions of the Handbook.
A comparison of data presented in recent editions of the Gas Turbine World Handbook for relevant gas turbines is shown in Table 12. The various qualifications given in the Handbook should be considered when evaluating this data.15
Table 12 Gas Turbine World Handbook budget plant prices for CCGT units, USD/kWISO
Gas turbine unit for a single shaft CCGT
block
Volume 26 2007-08
Equipment only, FOB
Volume 27 2009
Equipment only, FOB
Volume 28 2010
Turnkey
Volume 29 2012
Turnkey
Frame 9FB 520 551 494 536
M701F 529 539 491 533
GT26 521 549 497 539
SGT5-4000F 521 550 497 Not listed
SKM has also considered the trends in local construction cost parameters for Singapore as shown in Table 13 and Figure 10.
15 These are “bare bones” standard plant designs and exclude design options such as dual fuel and project specific requirements, are for sites with minimal transportation costs, site preparation and with non-union labour, and there can be a wide-range of prices for combined cycle plants depending on geographic location, site conditions, labour costs, OEM marketing strategies, currency valuations, order backlog and competitive situation.
SKM (Singapore) Pte. Ltd. PAGE 22
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Table 13 Local construction cost parameters for Singapore16
2006 2007 2008 2009 2010 2011 2012CPI (SingStats) 2009=100 91.3 93.2 99.4 100 102.8 108.2 114.1
Tradesman SGD/h 10 10.5 11.5 12 12 12.5 12.5
Labourer SGD/h 7 7 7.5 8 8 8 8.5
Building Price Index (re previous year) 3% 15% 9% -8% -1% -1% -1%
Industrial factories/wharehouses, owner occ., SGD/m2
900 1025 1200 1950 1700 1750 1600
Concrete (foundations) SGD/m3 88 92 160 160 150 127 137
Structural steel, UB, UC etc erected SGD/t 2700 3100 4500 6000 5200 5280 5230
Figure 10 Trends in Singapore local construction cost parameters, 2010 = 100
The apparent local construction costs are slightly below those of 2010 for the 2011-12 review.
0%
20%
40%
60%
80%
100%
120%
140%
2006 2007 2008 2009 2010 2011 2012
Inde
x relative
to 20
10
CPI (SingStats) 2010=100
Tradesman SGD/h
Labourer SGD/h
Building Price Index
Industrial factories/wharehouses, owner occ., SGD/m2
Concrete (foundations) SGD/m3
Structural steel, UB, UC etc erected GD/t
16 Successive issues of Rawlinson’s “Australian Construction Cost Handbook”, International Construction Costs table
SKM (Singapore) Pte. Ltd. PAGE 23
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
4.3. Initial capital cost
Modifications are applied to make the unit cost applicable to this study reflect different design features for the Singapore plant, and to consider that the plant required for this review is based on shared infrastructure within a multi-unit plant. A two-unit plant is assumed. The modifications applied are:
• Allowances are made for the capital cost of gas compression plant (2 train per unit);
• Civil costs are calculated on a two-unit station basis and then halved;
• Building and structures costs are calculated for a two unit station and then halved;
• The plant is based on a once-through cooling system with the civil costs added separately on a shared (two-unit) basis;
• Allowance for dual fuel systems for the gas turbines and fuel forwarding from the tanks;
• Allowance for a jetty and fuel unloading facilities is added separately on a shared (two-unit) basis; and
• Allowances for fuel tanks are added on a shared (two-unit) basis.
The resulting EPC cost for the plant (excluding external connections) is SGD479.2M per unit as shown in Table 14. This cost is on an "overnight" basis17.
17 That is, excluding Interest during Construction (IDC).
SKM (Singapore) Pte. Ltd. PAGE 24
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Table 14 EPC capital cost summary (per unit) for 2012-13, with comparison against the 2009-2010 review and the 2011-12 review18
Project Cost Summary 2009-2010
review SGD k
2011-12 review SGD k
Current review SGD k
Comments
I Specialized Equipment 345,000 292,400 245,345 II Other Equipment 47,100 9,668 13,334 III Civil 29,106 28,572 Shared IV Mechanical 41,306 32,955 V Electrical Assembly & Wiring 9,546 5,703 VI Buildings & Structures 13,217 11,966 Shared, except
turbine hall VII Contractor's Engineering & commissioning
7,000 19,866 20,679
VIII Contractor's Soft & Miscellaneous Costs (including Contractor's contingencies, margins and preliminaries)
20,000 91,099 78,681
Transport 6,900 Included Included Gas compressors 11,070 9,062 Adjust for OT C/W system 6,700 6,676 6,544 Shared Jetty & unloading 10,000 7,972 7,813 Shared Fuel tanks 19,000 18,933 18,556 Shared EPC equivalent capital cost excluding connections
461,700 550,859 479,212
Note that there may be additional savings if both units of a two unit plant were procured at the same time. A small reduction in the costs of the second (and subsequent units if more than two are procured) which is expected to be of the order of 5% would result due to the sharing of transaction and engineering costs at both the contractor and owner level. Where the plant procurement is phased by more than (say) two years, these savings are less likely to result.
Average load growth is projected to be less than 200MW/year through 2016, and peak demand growth to be 240-254MW/y, and hence it would be expected that additions of base-load plant in nominally 400MW blocks would be spaced 1.5 years apart or more, unless there are retirements from the market.
18 2009 values have been allocated to equivalent categories on an estimated basis
SKM (Singapore) Pte. Ltd. PAGE 25
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
If the plant were not phased then consideration would be given to constructing the plant as a "2+1" block instead of two "1+1" blocks. Technical performance is very similar (including the amount of output lost when one gas turbine trips). The specific capital cost (SGD/MW) is typically materially lower with a "2+1" arrangement than for two "1+1" blocks. However, this depends on the load growth being sufficiently high to justify the additional capacity being constructed immediately after the first unit. This is not included in this analysis.
4.4. Through-life capital costs
Capital costs of plant maintenance through the overhaul cycle of the gas turbine and steam turbine are included in Sections 5.1 and 5.2.
Additional capital costs are incurred through the project's life. Actual costs incurred vary considerably and are based on progressive assessments made of plant condition through the plant's life. Recommended estimates for this review are given in Table 15:
Table 15 Through-life capital expenditure (per unit)
Area Time within project Estimate, per unit Discounted equivalent, SGDM/unit (pre-tax real WACC=3.2%), per unit
Distributed control system (DCS)
15 years 7 SGDM real 4.4
Gas turbine rotor 15 years (100,000 to 150,000 operating hours)
12.7 SGDM real (USD10M)
8.0
Total 12.3The cost of the DCS upgrade depends on the level of obsolescence of related items such as field instrumentation and associated wiring.
Towards the end of the notional technical life of the plant, if market studies indicated that the plant may still be economic, studies would be undertaken to evaluate extending the plant's life. The studies and the resulting costs and resulting life extensions are not included.
4.5. Land and Site Preparation Cost
The land and site preparation cost excludes (i) facility costs (ancillary buildings, demineralisation plant, sea water intake/outfall structures, constructing the jetty for emergency fuel unloading facility and gas receiving facilities) and (ii) emergency fuel facilities. These costs have been included under capital cost for the current review.
SKM (Singapore) Pte. Ltd. PAGE 26
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
The land cost is based on 12.5 Ha of land and 200m of water front for a 2 unit plant. Based on data published by the JTC Corporation’s Land Rents and Prices, for a 30 year lease, the land price at Tuas View is between $245 and $306 per square metre19. This implies that the cost of a 12.5Ha parcel of land is between $30.625 million and $38.25 million. Water frontage fees range from $1,226 to $1,839 per metre per year. For a 200 metre waterfront, the annual cost is between $245,200 and $367,800. Using the average annual cost at a discount rate of 3.2% over 20 years, this gives an equivalent capital cost of $4.48 million. Total capital cost for land assuming a mid-point land cost is thus $38.91 million.
Site preparation cost is relatively minor. In 2010 for the 2011-12 review, this was assessed to be $1.5million. For the current review, we have assumed this to be $2 million. Total land and site preparation costs are thus $40.91million and a per unit cost of SGD$20.46 million.
The land and preparation cost for the 2010-12 review was SGD13.65M/unit.
4.6. Connection Cost
Connection costs exclude civil works for the plant’s, erection and assembly, detailed engineering and start-up costs. These costs have been included under the overall capital cost for the current review.
The electrical connection cost has been estimated using a "bottom-up" approach as shown in Table 16. SKM has taken into consideration in this assessment the cost of connecting two 400MW CCGT units using the configuration shown in Figure 11.
Table 16 Electrical connection costs (2 units)
Item Connection Cost Components Cost (SGDM) 1 Standard Connection Charge (to SPPG) SGD
50,000/MW 38.4
2 230kV Switchgear GIS Notes:
breaker and a half configuration
include switch house but exclude generator transformer
GIS 6 off 17.8
3 XLPE Underground Cable (based on 2x 1000MVA circuits of 1 km length, direct burial)
4.51/km 2km 9.0
Total 65.1
19 JTC's Land Rents and Prices with effect from 1 January 2012, http://www.jtc.gov.sg/Pages/JtcIndustrialLand_Price.aspx
SKM (Singapore) Pte. Ltd. PAGE 27
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Based on the standard Power Grid connection charge, the cost of electrical connection including the cost of the typical 230kV switchgear and XLPE underground cable is estimated to be SGD32.6M per unit.
The connection cost in the 2011-12 review was SGD31.6M/unit.
Figure 11 Assumed electrical connection configuration (items per Table 16)
The gas connection costs are escalated from the 2010 report to SGD13.3M or SGD6.65M per unit.
Total connection cost is thus SGD78.4M, or SGD39.2M/unit.
4.7. Owner's costs after financial closure
The Owner's costs incurred from Financial Closure to the Commercial Operation Date of the plant are typically allowed as percentage extra costs on the EPC basis plant costs.
SKM recommends the following allowances as shown in Table 17:
SKM (Singapore) Pte. Ltd. PAGE 28
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Table 17 Owner's costs allowances (after financial closure)
Area Percentage of EPC +
connection cost
Cost, per unit
(SGDM)
Owners Engineering 3% 15.6 Owners "minor items" 3% 15.6 Initial spares 2% 10.4 Start-up costs 2% 10.4 Construction related insurance etc.
1% 5.2
Total 57.0 Note that the capital cost estimates are made at the 50th percentile of expected outcomes as is considered appropriate for this application. The EPC estimate includes the contingency and risk allowances, along with profit margins, normally included in the Contractor's EPC cost estimates. The extra contingency allowances normally included by the owner within investment decision making processes to reduce the risk of a cost over-run below 50% are not included.
Owner's engineering costs are the costs to the owner of in-house and external engineering and management services after financial closure, including inspections and monitoring of the works, contract administration and superintendancy, project management and coordination between the EPC contractor, connection contractors and contractors providing minor services, witnessing of tests and management reporting.
Minor items include all the procurement costs to the owner outside of the primary plant EPC costs and the electricity and gas connections. This includes permits/licences/fees after Financial Closure, connections of other services, office fit-outs and the like. This also reflects any site specific optimisation or cost requirements of the plant above those of a "generic" standard plant covered in Section 4.3.
Start-up costs include the cost to the owner of bringing the plant to commercial operation (noting that the actual commissioning of the plant is within the plant EPC contractor's scope). The owner is typically responsible for fuels and consumables used during testing and commissioning, recruiting, training and holding staff prior to operations commencing, and for establishing systems and procedures.
Note that initial working capital, including initial working capital for liquid fuel inventory and for accounts receivable versus payable, are not included (these are an ongoing finance charge included in the fixed operating costs of the plant in Section 5.1).
SKM (Singapore) Pte. Ltd. PAGE 29
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
4.8. Owner's costs prior to Financial Closure
At the time of Financial Closure, when the investment decision is being made, the costs accrued up to that time against the project are "sunk" and are sometimes not included in a new entrant cost estimate.
Nevertheless, the industry needs to fund the process of developing projects to bring a plant from initial conception up to financial closure. If these are to be added, the costs can be highly variable. The allowances should include both in-house and external costs to the owner/developer from concept onwards including all studies, approvals, negotiations, preparation of specifications, finance arranging, legal, due diligence processes with financiers etc. These would typically be over a 3 to 5 year period leading up to financial close. An example of typical allowances based on percentages of the EPC cost is shown in Table 18.
Table 18 Owner's costs allowances prior to Financial Closure
Area Percentage of EPC +
connection cost
Cost, per unit (SGDM)
Permits, licenses, fees 2% 10.4 Legal & financial advice and costs
2% 10.4
Owner's engineering and in-house costs
2% 10.4
Total 31.1 Permits, licences and fees primarily consist of gaining the environmental and planning consents for the plant.
Legal and financial advice is required for establishing the project vehicle, documenting agreements, preparing financial models and information memoranda for equity and debt sourcing, management approvals and due diligence processes.
Owner's engineering and in-house costs prior to financial closure include the costs of conceptual and preliminary designs and studies (such as optimisation studies), specifying the plant, tendering and negotiating the EPC plant contract, negotiating connection agreements, attending on the feasibility assessment and due diligence processes, management reporting and business case preparation, etc.
Project development on a project financed basis sometimes incurs extra transaction costs, such as swaptions for foreign exchange cover or for forward interest rate cover. These are highly project specific and not always necessary. No extra allowance is included.
SKM (Singapore) Pte. Ltd. PAGE 30
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
5. Operating costs 5.1. Fixed annual running cost
An assessment of the fixed annual cost of operating a CCGT station is shown in Table 19.
Note that we have included the gas turbine and steam turbine Long Term Service Agreement (LTSA) costs as variable costs rather than fixed costs, as LTSA's are normally expressed substantially as variable costs. The EMA Vesting Contract Procedures state that semi-variable maintenance costs should be included with the fixed costs amounts. If calculated correctly with the appropriate plant factor, the same vesting contract LRMC will result. Current LTSA costs for CCGT plants have been expressed as variable costs in this review and hence these costs are included in the variable cost section.
Typically, an LTSA only covers the main gas turbine and steam turbine components. All of the balance of the plant including boilers, cooling system, electrical plant etc are maintained separately by the owner outside of the LTSA. The cost of this maintenance is typically considered to be a fixed cost, and is included in this section.
Table 19 Fixed annual operating cost allowance
Area SGDM for 2 units
Manning 4.53 Allowance for head office services 2.72 Fixed maintenance and other fixed operations20
11.501
Starts impact on turbine maintenance 0.785 Distillate usage impact on turbine maintenance
0.064
EMA license fee (fixed) 0.05778 Working capital (see below) 8.322 Emergency fuel usage 2.448 Property Tax 1.384 Insurance 4.792 Total (for 2 units) per year 36.594 Costs per unit would thus be SGD18.297M per year.
20 Calculated as 3% of the plant capital cost per year excluding the cost attributable to the gas turbine and steam turbine (which are included in the variable operating/maintenance costs below). These costs need to cover non-turbine maintenance, all other fixed costs including fixed charges of utilities and connections, service contracts, community service obligations etc.
SKM (Singapore) Pte. Ltd. PAGE 31
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Manning costs have been estimated based on 42 personnel covering 2 units at SGD107,700/person/year. The unit rate considers the cost allowed in 2010 for the 2011-12 review indexed using a factor produced from average remuneration changes in a “chemicals” manufacturing environment in Singapore (in the absence of a power generation industry index being available). This index is shown in Figure 12.
The personnel include shift operators/technicians and shift supervision as well as day shift management, a share of trading/dispatch costs if this is undertaken at the station (versus head office), engineering, chemistry/environmental, trades supervision, trades and trades assistants, stores control, security, administrative and cleaning support. The cost per person is intended to cover direct and indirect costs.
Figure 12 Labour cost index21
0%
20%
40%
60%
80%
100%
120%
1998 2000 2002 2004 2006 2008 2010 2012 2014
Labo
ur cost ind
ec (relatie to
2010)
21 Indexed produced using SingStats “Yearbook of statistics Singapore 2011 Table 10.7 and 10.8 "Chemical and chemical products" manufacturing” average remuneration.
SKM (Singapore) Pte. Ltd. PAGE 32
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Head office costs would be highly variable and depend on the structure of the business and the other activities the business engages in. Only head office support directly associated with power generation should be included as part of head office costs. The allowance for head office costs is a nominal allowance (60% of manning cost allowance) for services that might be provided by head-office that are relevant to the generation services of the plant. These would include (for example):
• Support services for generation such as trading etc;
• Corporate management and governance;
• Human Resources and management of group policies (such as OH&S, training etc);
• Accounting and legal costs at head office; and
• Corporate Social Responsibility costs.
The manning and head office costs are bundled (with non-fuel working capital costs) are based on 60% of the manning cost allowance per the 2011-12 review.
The starts impact on turbine maintenance costs accounts for the fact that some gas turbine OEM's add an Equivalent Operating hours (EOH) factor for starts and this impacts on the costs under the LTSA.
EOH costs are based on 1.75 EUR/CCGT-MWh at nominal full load based on discussions with the OEMs. Allowing for part load adjustments the equivalent cost is EUR469.7/EOH. Note that the LTSA is based on the gas and steam turbine only rather than maintenance of the whole plant. The starts factor only impacts on the gas turbine component however. Based on 50 starts/unit and 10 EOH/start, the cost is SGD392,604/unit/year.
Additionally, the distillate usage (discussed below) also has an impact on turbine EOH consumption. Based on 1.5 EOH/hour when operating on distillate, the additional EOH consumption over natural gas fuel operation is 0.5 EOH/hour. This equates to an impact on maintenance of SGD32,053/unit/year.
Calculation of the working capital cost and the emergency fuel usage cost below requires an estimate of the costs of distillate and natural gas. For the purposes of the draft report assumed prices of 29.25 SGD/GJ and 22.10 SGD/GJ for distillate and gas, respectively are applied.
This distillate cost assumption is based on USD977.6/t (USD128.29/bbl) for this draft report based on the average of daily rates for Gasoil (0.5%) from Dec 2011 through Feb 2012. A handling and delivery cost based on the allowance of USD6.02/bbl is added to give a delivered distillate cost of USD134.31/bbl, or SGD29.25/GJ.
SKM (Singapore) Pte. Ltd. PAGE 33
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Working capital costs are the annual costs of the financial facilities needed to fund working capital. This comprises two components:
• Emergency fuel inventory: 90 days (per 2 units), 8.8PJ at a distillate cost of SGD29.25/GJ and a pre-tax real WACC of 3.2% gives a working capital cost of SGD8.211M/year; and
• Working capital against the cash cycle (timing of receipts from sales versus payments to suppliers) based on a net timing difference of 30 days and excluding fuel costs (based on the short settlement period in the market of 20 days from the time of generation). For two units the working capital requirement on this basis is SGD3.46M and the working capital cost (using a pre-tax real WACC of 3.2%) is SGD0.11M/year.
Emergency fuel usage is a notional amount of emergency fuel usage for testing, tank turnover etc. Calculated as 1% of the annual fuel usage and using a cost based on the extra cost of distillate over natural gas (SGD29.25/GJ vs SGD22.10/GJ).
Property tax has been estimated based on 10% per year of an assumed Annual Value of 5% of the land, preparation and buildings/structures cost22. Note is also made of the IRAS circular regarding property taxes on plant and machinery23. The value of certain fixed plant and machinery items must be included within the property valuation when calculating property taxes. However an appended list of exemptions exempts most of the principal plant items of a CCGT plant including turbines, generators, boilers, transformers, switchgear etc. To allow for the extra value of the portion of the plant that is included, 10% of the cost of the plant is included in the property tax valuation calculation (except where already included). The total value included for calculation of property tax is thus SGD276M (2 units).
Insurance has been estimated based on 0.5% of the capital cost. This is considered to cover property, plant and industrial risks but would not cover business interruption insurance or the cost of hedging against plant outages.
A comparison with the values shown in the 2011-2012 review is shown in Table 20.
22 Following http://www.business.gov.sg/EN/Government/TaxesNGST/TypesofTaxes/taxes_property.htm 23 IRAS circular: "TAX GUIDE ON NON-ASSESSABLE PLANT AND MACHINERY COMPONENTS FOR PETROCHEMICAL AND POWER PLANTS", 16 Nov 2006.
SKM (Singapore) Pte. Ltd. PAGE 34
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Table 20 Fixed annual operating cost allowance comparison, SGD Millions for 2 units
Area 2011-12 review Current review
Manning 4.20 4.53 Allowance for head office services 2.52 2.72 Fixed maintenance and other fixed operations 15.631
11.501
Starts impact on turbine maintenance 0.935 0.785 Distillate usage impact on turbine maintenance 0.0763
0.064
EMA license fee (fixed) 0.05 0.058 Working capital 13.526 8.322 Emergency fuel usage 1.497 2.448 Property Tax 1.037 1.384 Insurance 5.509 4.792 Total (for 2 units) per year 44.981 36.594
5.2. Variable non-fuel cost
It is assumed a Long Term Service Agreement (LTSA) would be sought for the first one to two overhaul cycles of the gas turbine and steam plant (typically 6 to 12 years). These are typically structured on a "per operating hour" or "per MWh" basis and hence are largely variable costs.
An assessment of the variable, non-fuel, costs is given in Table 21.
SKM (Singapore) Pte. Ltd. PAGE 35
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Table 21 Variable non fuel costs
Area SGD/MWh Notes Gas turbine 3.42 Based on approximately EUR1.75/MWh of total plant
output, adjusted for part load factor Steam turbine 0.5 Balance of plant, chemicals, consumables
0.5
Town Water 0.052 For a salt water cooled plant the town water costs are typically small. Based on 0.1t/MWh usage and a cost of 0.52 SGD/t24.
EMC fees 0.343 Based on EMC's Admin Fees of S$29.027 million / 2, and a forecast wholesale volume of 42,257 GWh.
PSO 0.221 From EMC website25 for FY2010-11 EMA license fee (variable)
0.179 Advised by EMA
Total 5.21 Note the MWh in the above are those of the overall CCGT plant unit, not the individual turbine output.
If the alternative treatment of the LTSA had been adopted the variable operating cost would reduce by approximately SGD3.92/MWh and the fixed operating cost would increase by approximately SGD19.16M/y (for 2 units). This would not change the LRMC value calculated.
A comparison with the values shown in the 2011-2012 review is shown in Table 22.
Table 22 Variable operating cost allowance comparison, SGD/MWh
Area 2011-12 review
Current review
Gas turbine 4.64 3.42 Steam turbine 0.5 0.5 Balance of plant, chemicals, consumables 0.5 0.5 Town Water 0.2 0.052 EMC fees 0.3343 0.343 PSO 0.2205 0.2212 EMA license fee (variable) 0.155 0.179 Total 6.55 5.21
24 http://www.pub.gov.sg/general/Pages/WaterTariff.aspx for “Industrial Water Tariff” 25 http://www.emcsg.com/psobudgetandfees
SKM (Singapore) Pte. Ltd. PAGE 36
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
6. Other parameters 6.1. Build duration
Current expected build duration for this type of plants is 30 months. This is unchanged from the 2011-2012 review.
6.2. Economic life
The technical life of this type of plant is considered to be approximately 30 years.
The economic life has been assessed at 20 years as discussed in Appendix B (versus 24 years in the 2011-12 review).
6.3. Average expected utilisation factor
In the 2011-12 review the plant load factor of the new plant was determined from the average historical capacity factor of the existing Class F plant (Senoko Energy's CCP 3 to 5, YTL PowerSeraya's CCP 1 and 2 and Tuas Power Generation's CCP1 to 4) for the 12 months leading up to the base month. .
EMA has advised that for consistency with the previous reviews, the actual historic capacity factor for the previous 12 months should again be applied. This value has been advised by EMA to be 72.8%.
SKM (Singapore) Pte. Ltd. PAGE 37
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
7. Results – vesting contract parameters 7.1. Introduction
The LRMC resulting from the inclusion of the parameters considered in this report along with the financial parameters that are to be determined by others will be calculated by EMA.
For the purposes of comparing the impacts of the changes in technical parameters, a calculation is included of the LRMC, made using assumptions for financial parameters where necessary pending their calculation by others.
7.2. Summary of technical parameters
Table 23 Summary of recommended technical parameters and previous values
Item Parameter 2011-12 Value 2013-14 Value 6 Economic capacity of the most economic
technology in operation in Singapore (MW) 381 383.47MW net at 32oC
7 Capital cost of the plant identified in item 6 ($US/kW)
1053 1004.21 USD/kW
8 Land, infrastructure and development cost of the plant identified in item 6 ($Sing million)
152.0M SGD 147.81M
11 HHV Heat Rate of the plant identified in item 6 (Btu/kWh)
7010 6886 btu/kWh net HHV
12 Build duration of the plant identified in item 6 (years)
2.5 2.5 years
13 Economic lifetime of the plant identified in item 6 (years)
24 20 years
14 Average expected utilisation factor of the plant identified in item 6, i.e. average generation level as a percentage of capacity (%)
74.9% 72.8%
15 Fixed annual running cost of the plant identified in item 6 ($Sing)
22.49 18.297 M SGD
16 Variable non-fuel cost of the plant identified in item 6 ($Sing/Mwh)
6.55 5.21 SGD/MWh
The significant differences from the previous review are considered to be primarily attributable to:
• A reduction in the estimated EPC cost of large CCGT plants in the region;
• A reduction in WACC, which reduces the capital contribution, working capital costs and other minor parameters;
• An increase in the SGD / USD exchange rate; and
SKM (Singapore) Pte. Ltd. PAGE 38
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
• Recommended removal of the allowances for economic shutdowns from the calculation of the utilisation factor.
7.3. Calculated LRMC
Table 24 Assumed financial parameters for the LRMC calculation
Parameter Value Notes WACC 5.26% post-tax, nominal
3.2% pre-tax, real Post-tax nominal value from EMA
CPI 3.03% Dec 2011-Feb2012 Gas price $22.10 SGD/GJ Assumption based on oil price Exchange rates 1.2764 SGD/USD
Table 25 Calculated LRMC for 2012-13
Parameter Value SGD/MWh Notes Fuel component 160.55 Capital component 24.74 See note below Fixed opex 7.49 Variable opex 5.21 Total 197.98 Note that in accordance with the Vesting Contract formulae and the treatment in previous years, the WACC applied in the calculation of the LRMC is the nominal WACC. Comparison with previous estimates
Table 26 Comparison of the calculated LRMC with the previous estimate, SGD/MWh
Parameter 2011-12 review Current review WACC 8.43% post-tax, nominal
6.37% pre-tax, real 5.26% post-tax, nominal
3.2% pre-tax, real CPI 3.56% 3.03% Gas price $17.22 $22.10 SGD/GJ Exchange rates 1.393 1.2764 SGD/USD Fuel component 127.48 160.55 SGD/MWh Capital component 34.80 24.74 SGD/MWh Fixed opex 8.99 7.49 SGD/MWh Variable opex 6.55 5.21 SGD/MWh Total 177.82 19.98 SGD/MWh
SKM (Singapore) Pte. Ltd. PAGE 39
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
8. Review of Scale Factor Indices 8.1. Introduction
In the terms of reference, the Authority requested that a review be made on the method of calculating the LRMC Scale Factor Indices (as set out in Section 3.8 of the Procedures paper) and recommend improvements to the method with the aim of reducing the quarterly volatility of the vesting price. The Vesting Contract Procedures require that adjustment indices be determined each quarter to scale the allocated Vesting Price. These indices are the:
• Capital Cost Index to reflect movements in the capital cost for the current quarter relative to the base quarter; and
• Overhead Cost Index which is mainly an inflation index and seeks to reflect the overhead cost for the current quarter relative to the base quarter.
The Capital Cost Index is applied to the Total Capital Cost and takes into consideration changes to the cost of acquiring and building a new plant and the impact of exchange rate movements on the cost of a new plant. To calculate the Capital Cost Index, the Domestic Supply Price Index (DSPI) published by the Singapore Department of Statistics is used. Using the most recent DSPI (DSPID), and the DSPI from two quarters before the most recent (DSPID-2), the DSPI for the next quarter is calculated as:
DSPID+1 = DSPID + (DSPID – DSPID-2)/2
and for the next quarter Capital Cost Index, CI = DSPID+1/DSPIB where
DSPIB is the DSPI at the Base date.
Similarly, the Overhead Cost Index is applied by the Authority to Total Overhead Cost and takes into consideration the movements in general inflation in the Singapore economy relative to the base quarter and is applied to non-fuel overhead costs including wages and imported spare parts. The Consumer Price Index (CPI) published by the Singapore Department of Statistics is used for the purpose of calculating the Overhead Cost Index in a manner similar to the Capital Cost Index. Using the most recent CPI (CPID), and the CPI from two quarters before the most recent (CPID-2), the CPI for the next quarter is calculated as:
CPID+1 = CPID + (CPID – CPID-2)/2
and for the next quarter Overhead Cost Index, OCI = CPID+1/CPIB where
CPIB is the CPI at the Base date.
SKM (Singapore) Pte. Ltd. PAGE 40
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
8.2. SKM Consideration
The formulation of the index forecast for quarter D+1 using the trend from quarter D-2 to D does potentially introduce volatility to the calculation of the cost indices. Where the index growth of two consecutive quarters exhibit either high positive or high negative numbers, the forecast for D+1 quarter will exacerbate the trend. Also the use of quarterly data that constantly moves while the base quarter is fixed means that the trend since the base quarter is not necessarily reflected. Using a longer period between quarters however is likely to reduce volatility. Fixing the index at the base quarter plus using the trend going forward from the base quarter is likely to both reduce volatility as well as reflect the trend in cost movements since the base quarter. The forecast indices may be reflected as:
DSPIB+n+1 = DSPIB + (DSPIB+n – DSPIB)/n
and for the next quarter Capital Cost Index, CI = DSPIB+n+1/DSPIB where:
DSPIB is the DSPI at the Base date
DSPIB+n is the most recent quarter DSPI and
DSPIB+n+1 is the forecast DSPI for the next quarter
n is the number of quarters since the base quarter
Similarly,
CPIB+n+1 = CPIB + (CPIB+n – CPIB)/n
and for the next quarter Overhead Cost Index, OCI = CPIB+n+1/CPIB where
CPIB is the CPI at the Base date
CPIB+n is the most recent quarter CPI and
CPIB+n+1 is the forecast CPI for the next quarter
n is the number of quarters since the base quarter
Based on this alternative formulation, an illustrative graph of the index is shown in Figure 13. Volatility has reduced compared to the original formulation.
SKM (Singapore) Pte. Ltd. PAGE 41
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Figure 13 Illustrative index – current formulation vs alternative formulation
0.8
0.9
0.9
1.0
1.0
1.1
1.1
1.2
1.2
1 2 3 4 5 6 7 8 9 10 11 12
CI/O
CI
Quarter B+
Current
Alternative
The illustrative indices used for the graph are shown in Table 27.
SKM (Singapore) Pte. Ltd. PAGE 42
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Table 27 Illustrative index calculation
Quarter DSPI/CPI Original CI/OCI Alternative CI/OCI
-1 100 Base 104
1 107 2 110 1.06 1.06 3 115 1.09 1.09 4 108 1.14 1.14 5 103 1.03 1.05 6 105 0.93 0.99 7 108 1.00 1.01 8 110 1.06 1.04 9 116 1.08 1.06
10 108 1.15 1.13 11 113 1.03 1.04 12 116 1.07 1.09
8.3. Other alternatives
Some other alternatives to the current indices used and formulation have been suggested. These are considered below:
Use if a single data point (CPI/DSPI for the latest month)
SKM understands that the existing formula attempts to project the recent inflation trend into the next quarter. Using a single point data does not enable any trend to be projected. In effect, it would be a backward looking number – which by itself is not unreasonable as many other inflation adjustments are backward looking and seeks to compensate for price changes that has happened. However, this does not appear to be the aim of the current indexation formulation which is forward looking. We also need to recognize that volatility is not necessarily addressed as the single point data is likely also be volatile.
SKM (Singapore) Pte. Ltd. PAGE 43
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Removal of indexation or reduce the frequency of the adjustments (eg. annual adjustments)
Removing indexation altogether ie no adjustments over the two year vesting price period is a radical departure from the current method and may lead to significant divergence between the price and the cost of generation especially in an environment where inflation has been fairly high. If the aim is to reduce volatility, a longer term trend for inflation may be estimated and applied on an annual basis. It can then be set in place for the coming year regardless of the actual CPI/DSPI outcome. Annual adjustments can then be made to reconcile actual to the projected CPI/DSPI outcomes. So if the EMA takes the view based on historical trends that the likely inflation rate for the coming year is say 4% pa, it could decide that quarterly inflation adjustment of 1% would be applied. At the end of 4 quarters, this indexation will result in prices increasing by just over 4.06%. If actual annual inflation were higher or lower than this value, a once off adjustment would be implemented to bring vesting prices back into line with the general price movement. This is likely to smooth out the volatility from the quarterly projections of CPI/DSPI.
Alternatively, if annual adjustment is preferred to quarterly indexation of the overhead and/or capital cost components, it may be accomplished by making the adjustments to the costs based on the expected inflation rate for the year. In practice, EMA’s vesting prices are due to be determined in September 2012 based on mid-year 2012 prices. These prices are to be applied on 1 January 2013, adjusted for the expected inflation rate for 2013. These prices will apply till the next adjustment on 1 January 2014 where the prices will be adjusted to reflect expected 2014 inflation.
As indicated earlier, annual regulatory adjustments are not unusual. Australian regulators do not usually allow quarterly adjustments to prices paid by customers. Retail electricity, gas and water prices are adjusted annually to reflect both cost changes as well as inflation. In period adjustments are usually only allowed due to the pass through of an unusual event like the introduction of a new tax or levy. Similarly network revenue which are set on a five year basis are adjusted annually to reflect the impact of inflation even though inflation indices like the CPI is published quarterly.
SKM (Singapore) Pte. Ltd. PAGE 44
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Alternatives to using headline CPI/DSPI
We understand that the MAS core inflation index is a less volatile measure of inflation. The major difference in the core inflation index and the CPI is the exclusion of accommodation and private transport from the core inflation index. We understand that the cost for accommodation and private transport in Singapore is impacted significantly by administrative policies ie the size of government rebates for Service and Conservancy Charges and rental and the supply quota size of COE releases. It is not clear that the costs of the “new entrant” generation company would be exposed to the accommodation market and private transport. Thus, it could be expected that the overall costs movements of the new entrant is likely to be reflected in the core inflation measure rather than the headline inflation measure. Over the longer term, it appears that CPI is generally higher than the MAS core inflation index. Thus it can be expected that changing from the CPI to the MAS index would result in not only in lower volatility but also lower vesting contract prices.
Annual review of capital cost
Information sources like the Gas Turbine World Handbook are not published annually and is usually out of date by the time they are published and thus of little use for annual review purposes. Also the values expressed in the Handbook not necessarily consistent across editions and thus cannot be used as indications of price changes across time. As an alternative, it may be possible to undertake yearly reviews of just the capital cost component, however such an exercise is likely to be time consuming with uncertain gain. While an annual review would ensure that the capital cost component would be more reflective of the prevailing costs faced by a new entrant it should be recognised that information to support annual reviews may not necessarily be available given that only a limited number of comparable projects are likely to be developed annually.
To decide if an annual review of capital costs is justified, an assessment thus has to be made that the cost of engaging in an annual review is outweighed by the cost to consumers in the event that the capital cost is unnecessarily high due to the use of an unsuitable index.
Recommendation
Given the dissatisfaction expressed by various parties regarding the volatility of vesting prices due to the indexation process, SKM recommends that changes be made to the method of determining the LRMC Scale Factor Indices within the two year price period. We recommend that for the overhead cost component, quarterly indexation be removed and replaced with an annual adjustment to reflect forecast inflation over the year.
For capital costs, if the EMA is of the opinion that the cost of undertaking an annual review is justified by the likely savings due to the use of an unsuitable index, SKM recommends that an annual review be made of the capital cost components.
SKM (Singapore) Pte. Ltd. PAGE 45
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Appendix A Prescribed procedures Table 28 Excerpt from Vesting Contract Procedures (parameters for this review in bold)
No. Parameter Description Method of Determination
1 Determination Date Date on which the calculations of the LRMC, which is to apply at the Application Date, are deemed to be made
Determined by EMA
2 Base Month Cut-off month for data used in determination of the LRMC base parameters. For the following base parameters which tend to be volatile in nature, the data to be used for estimating each of them shall be based on averaging over a three month leading up to and including the Base Month: 1. Exchange rate denominated in
foreign currencies into Singapore dollars
2. Diesel price to calculate cost of carrying backup fuel
3. Risk-free rate 4. Debt premium to calculate cost
of debt 5. Consumer price index 6. Domestic supply price index 7. Imported iron and steel index
Determined by EMA
3 Application Date Period for which the LRMC to apply Determined by EMA 4 Current Year Year in which the Application Date
falls Determined by EMA
5 Exchange Rate ($US per $Sing)
The exchange rate is that as determined in Section 3.8
Determined by EMA (in consultation with finance experts)
6 Economic capacity of the most economic technology in operation in Singapore (MW)
The size of the most thermally efficient unit taking into account the requirements of the Singapore system, including the need to provide for contingency reserve to cover the outage of the unit and the fuel quantities available. It is acknowledged that this value may depend on the manufacturer. (For CCGT technology the size of the unit is expected to be around 370MW)
Determined by EMA (in consultation with the engineering and power systems experts)
SKM (Singapore) Pte. Ltd. PAGE 46
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
No. Parameter Description Method of Determination
7 Capital cost of the plant identified in item 6 ($US/kW)
Capital cost includes the purchase and delivery cost of the plant in a state suitable for installation in Singapore and all associated equipment but excludes switchgears, fuel tanks, transmission and fuel connections, land, buildings and site development included in item 8. Where more than one unit is expected to be installed that will share any equipment, the costs of the shared equipment should be prorated evenly to each of the units
Determined b EMA (and in consultation with the engineering and power systems experts)
8 Land, infrastructure and development cost of the plant identified in item 6 ($Sing million)
Where more than one unit is expected to be installed that will share any equipment or facilities, the costs of the shared equipment or facilities should be prorated evenly to each of the units. These costs should include all capital, development and installation costs (excluding all costs included in item 7 and financing costs during the build period). These costs should include the following specific items:
Acquisition costs of sufficient land to accommodate the plant defined above in item 6 (alternatively land may be included as annual rental cost under Fixed Annual Running Costs)
Site development Buildings and facilities Connectors to gas pipelines Switchgear and connections to
transmission Emergency fuel facilities Project management and
consultancy
Determined by EMA, (a) In consultation with the engineering and power systems experts in relation to the following values:
size of site required site development buildings and
facilities connections to
pipelines switchgear
connections to transmission
emergency fuel facilities
project management and consultancy; and
(b) In consultation with real estate experts in relation to land value
9 HSFO 180 CST Oil Price (US$/MT)
The HSFO 180 CST Oil Price is that as determined in Section 3.7
Determined by EMA
SKM (Singapore) Pte. Ltd. PAGE 47
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
No. Parameter Description Method of Determination
10a Gas Price ($Sing/GJ)
The current most economic generating technology in Singapore uses natural gas. This is the Singapore price for gas delivered to electricity generating companies as calculated by the Authority using existing Singapore pipeline gas contracts based on the HSFO price or any other method as determined by the Authority and announced to the gas industry.
Determined by EMA
10b LNG Price ($Sing/GJ)
This is the Singapore regasified LNG price as determined by the Authority. The LNG price is used in place of 10a for the LNG Vesting Quantities under the LNG Vesting Scheme.
The LNG Price includes: the LNG hydrocarbon charge the LNG terminal tariff the transportation tariff applicable
to regasified LNG the LNG Aggregator’s margin
Determined by EMA
11 HHV Heat Rate of the plant identified in item 6 (Btu/kWh)
The high heat value heat rate of the plant specified under item 6 that this expected to actually be achieved, taking into account any improvement or degradation in efficiency from installation in Singapore and other reasonable factors
Determined by EMA (in consultation with the engineering and power systems experts)
12 Build duration of the plant identified in item 6 (years)
The time from the commencement of the major cost of development and installation being incurred up to the time of the plant commissioning. This parameter is used to calculate the financing cost over the duration of the building period and assumes that the development costs are incurred evenly across this period. The build duration should be specified to reflect this use and meaning as opposed to the actual time from the commencement of site development to the time of plant commissioning.
Determined by EMA (in consultation with the engineering and power systems experts)
13 Economic lifetime of the plant identified in item 6 (years)
The expected time from commissioning to decommissioning of the plant. This number is used to amortise the capital cost of the plant, and of installation and development.
Determined by EMA (in consultation with the engineering and power systems experts)
SKM (Singapore) Pte. Ltd. PAGE 48
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
No. Parameter Description Method of Determination
14 Average expected utilisation factor of the plant identified in item 6, i.e. average generation level as a percentage of capacity (%)
The utilisation factor is the expected annual proportion of plant capacity that will be used for supplying energy for sale. It should exclude station usage, expected maintenance and forced outages and the expected time spent providing reserve capacity. The determination of the factor should assume that the plant is efficiently base-loaded
Determined by EMA (in consultation with the engineering and power systems experts)
15 Fixed annual running cost of the plant identified in item 6 ($Sing)
These costs are the fixed operating and overhead costs that are incurred in having the plant available for supplying energy and reserves but which are not dependent on the quantity of energy supplied. It is acknowledged that some costs are not easily classified as fixed or variable. The costs expected to be included in this parameter are:
Operating labour cost – it is expected that the plant will be running for three shifts per day and seven days per week so all operating labour cost is likely to be a fixed annual cost
Direct overhaul and maintenance cost, with any semi-variable costs treated as annual fixed costs
Generating license Insurance Property tax Costs of emergency fuel Other charges Other overhead costs
(a) Determined by EMA, in consultation with engineering and power systems experts in relation to the following values:
Operating labour Direct overhaul and
maintenance cost Costs of emergency
fuel Other overhead
costs; and (b) Determined solely by EMA
Generating license Insurance Property tax Other charges
16 Variable non-fuel cost of the plant identified in item 6 ($Sing/Mwh)
Any costs, other than fuel costs, that vary with the level of energy output for a base-load plant and are not covered by item 15
Determined by EMA (in consultation with the engineering and power systems experts
17 Proportion of debt by assets
The proportion of debt to total assets at market value. It is an estimate of the industry standard ratio for private sector generators in an economic environment similar to Singapore. The ratio is used to calculate the weighted average cost of capital (“WACC”)
Determined by EMA (in consultation with the finance experts)
SKM (Singapore) Pte. Ltd. PAGE 49
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
No. Parameter Description Method of Determination
18 Risk free Rate (%) The risk-free rate plus a premium as determined by the Authority. The total cost of debt will comprise the base lending rate, the loan margin and upfront and other fees
Determined by EMA (in consultation with the finance experts)
19 Cost of Debt (%) Risk-free rate plus a premium as determined by the Authority. The total cost of debt will comprise the base lending rate, the loan margin and upfront and other fees
Determined by EMA (in consultation with the finance experts)
20 Market Risk Premium (%)
The market risk premium represents the additional return over investing in risk-free securities that an investor will demand for investing in electricity generators in Singapore, as determined by the Authority
Determined by EMA (in consultation with the finance experts)
21 Beta Parameter of scaling the market risk premium for calculating the cost of equity as determined by the Authority. Beta is a measure of the expected volatility of the returns on a project relative to the returns on the market, that is, the systematic risk of the project
Determined by EMA (in consultation with the finance experts)
22 Tax rate (%) Corporate tax rate applicable to generating companies in Singapore at the base date. This rate should include any applicable tax rebates or tax incentives available to generating companies, and be consistent with the gearing and interest rates defined in 17 &18
Determined by EMA
23 Cost of equity (%) The return of equity for the business as calculated from the previous data. It is calculated as item 18+ (item 20) (item 21) + item 22
Calculated by EMA (in consultation with the finance experts)
SKM (Singapore) Pte. Ltd. PAGE 50
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Appendix B Market modelling Market modelling has been undertaken on the Singapore market using the PLEXOS model.
This is to support the estimation of the economic life of a notional new entrant CCGT. The results of the modelling also support the assessment of the part load factor and utilisation factor for the notional new entrant.
The new entrant CCGT plant was modelled with the heat rate identified in this report and based on being fuelled with the same gas cost as other CCGT units in Singapore. This means that the new entrant unit dispatches before the other (non-cogen) CCGT units in the market. The calculated parameters for the new entrant CCGT unit are shown in Figure 14.
Figure 14 Market modelling results for a notional new entrant CCGT
Weighting each year by the real WACC factor, the weighted averages are:
ACF
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2010 2015 2020 2025 2030 2035 2040
Factor
ACF (averaged)
Load factor
CF
93.2% Load factor 86.9% CF 81.1%
SKM (Singapore) Pte. Ltd. PAGE 51
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
The economic life of the new entrant is dictated by the rate of development of the heat rate of newer plants and real reductions in capex of newer plants.
Based on the parameters in Gas Turbine World Handbooks of 1994 and 2012, and applying “E” class CCGT’s in 1994 and the latest “F/H” class units in the 2012 Handbook, the average improvement in heat rate per year was assessed as -0.0077 GJ/MWh/y. The real rate of reduction in specific capital cost was assessed as 2.29% per year.
Applying these rates of change to the new entrant parameters it is calculated that the LRMC of a newer unit would become lower than the SRMC of an incumbent after 20 years. Thus the economic life of the new entrant plant is assessed to be 20 years.
SKM (Singapore) Pte. Ltd. PAGE 52
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
SKM (Singapore) Pte. Ltd. PAGE 53
Appendix C Technical performance data Thermodynamic analysis Performance analysis of new entrant "F" class CCGT units has been undertaken using the GTPro and GTMaster software suite Version 20. Analyses have been made based on optimisation at the site average ambient and cooling water conditions. Representative performance parameters as calculated are shown in the following figures:
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Figure 15 Performance analysis - Alstom "F" class CCGT, clean-as-new
Ambient1.013 P29.5 T85% RH
GT PRO 22.0 Desktop Administratorp [bar] T [C] M [kg/s], Steam Properties: IAPWS-IF97464 03-29-2012 22:21:53 file=D:\DOCUMENTS AND SETTINGS\RZAUNER\MY DOCUMENTS\SKM PROJECTS\EMA VESTING CONTRACTS 2012\GTPRO\ALSTOM GT26\ALSTOM GT26 CCGT.GTP
GT PRO 22.0 Desktop AdministratorGross Power 416120 kWNet Power 407620 kWAux. & LossesAux. & Losses 8500 kWLHV Gross Heat Rate 6135 kJ/kWhLHV Net Heat Rate 6263 kJ/kWhLHV Gross Electric Eff. 58.68 %LHV Net Electric Eff. 57.48 %Fuel LHV Input 709122 kWthFuel HHV Input 785074 kWthNet Process Heat 0 kWth
SKM (Singapore) Pte. Ltd. PAGE 54
1.013 p87.78 T609 M
41.51 T111.5 M
HP
HPB130.3 p331 T76.38 M
470.2 T344 T
IP
IPB31.33 p236.3 T11.9 M
281 T249.3 T
LP
LPB3.799 p141.8 T10.71 M
187.2 T152.5 T
Cold Reheat
31.94 p369 T87.13 M
1.216 M
1.044 p631.9 T609 M
Natural gas 15.32 M709122 kWth LHV
ALS GT26(Curve Fit OEM Data Model #460)
@ 100% load
1.013 p29.5 T593.7 M85% RH
1.003 p29.5 T593.7 M
0.0778 p40.98 T110.3 M
Stop Valve
124 p566 T90.13 M
27.6 p566 T99.03 M
3.381 p288.5 T8.472 M
Hot Reheat
to HRSG
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Figure 16 Performance analysis - GE "F" class CCGT, clean-as-new
Ambient1.013 P29.5 T85% RH
GT PRO 22.0 Desktop Administratorp [bar] T [C] M [kg/s], Steam Properties: IAPWS-IF97464 03-29-2012 23:02:59 file=D:\DOCUMENTS AND SETTINGS\RZAUNER\MY DOCUMENTS\SKM PROJECTS\EMA VESTING CONTRACTS 2012\GTPRO\GE FRAME 9FB\GE FRAME 9FB CCGT.GTP
GT PRO 22.0 Desktop AdministratorGross Power 406010 kWNet Power 397499 kWAux. & LossesAux. & Losses 8511 kWLHV Gross Heat Rate 6163 kJ/kWhLHV Net Heat Rate 6295 kJ/kWhLHV Gross Electric Eff. 58.41 %LHV Net Electric Eff. 57.19 %Fuel LHV Input 695108 kWthFuel HHV Input 769559 kWthNet Process Heat 0 kWth
SKM (Singapore) Pte. Ltd. PAGE 55
1.013 p90.36 T611.8 M
41.51 T106.1 M
HP
HPB130.3 p331 T89.01 M
490.2 T344 T
IP
IPB31.33 p236.3 T7.594 M
269.5 T249.3 T
LP
LPB3.799 p141.8 T10.96 M
187.1 T151.7 T
Cold Reheat
31.94 p369 T86.03 M
1.301 M
1.044 p656.3 T611.8 M
Natural gas 15.02 M695108 kWth LHV
GE 9371FB(Curve Fit OEM Data Model #394)
@ 100% load
1.013 p29.5 T596.7 M85% RH
1.003 p29.5 T596.7 M
0.0778 p40.98 T104.8 M
Stop Valve
124 p566 T89.01 M
27.6 p566 T93.62 M
3.381 p288.5 T8.428 M
Hot Reheat
to HRSG
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
Figure 17 Performance analysis - Mitsubishi "F" class CCGT, clean-as-new
SKM (Singapore) Pte. Ltd. PAGE 56
Ambient1.013 P29.5 T85% RH
GT PRO 22.0 Desktop Administratorp [bar] T [C] M [kg/s], Steam Properties: IAPWS-IF97464 03-29-2012 23:25:14 file=D:\DOCUMENTS AND SETTINGS\RZAUNER\MY DOCUMENTS\SKM PROJECTS\EMA VESTING CONTRACTS 2012\GTPRO\MITSU 701F\MITSU 701F CCGT.GTP
GT PRO 22.0 Desktop AdministratorGross Power 432429 kWNet Power 423683 kWAux. & LossesAux. & Losses 8746 kWLHV Gross Heat Rate 6215 kJ/kWhLHV Net Heat Rate 6344 kJ/kWhLHV Gross Electric Eff. 57.92 %LHV Net Electric Eff. 56.75 %Fuel LHV Input 746587 kWthFuel HHV Input 826552 kWthNet Process Heat 0 kWth
1.013 p95.05 T675.2 M
41.51 T108.1 M
HP
HPB130.3 p331 T83.89 M
469.4 T344 T
IP
IPB31.33 p236.3 T11.47 M
277 T249.3 T
LP
LPB3.799 p141.8 T14.22 M
193.4 T151.7 T
Cold Reheat
31.94 p369.3 T81 M
1.314 M
1.044 p615.5 T675.2 M
Natural gas 16.13 M746587 kWth LHV
Mtsb 701 F(Physical Model #355)
@ 100% load
1.013 p29.5 T659.1 M85% RH
1.003 p29.5 T659.1 M
0.0778 p40.98 T106.8 M
Stop Valve
124 p566 T83.89 M
27.6 p566 T92.47 M
3.381 p288.5 T11.62 M
Hot Reheat
to HRSG
Long Run Marginal Cost (LRMC) parameters, 2013-2014 – Draft report
SKM (Singapore) Pte. Ltd. PAGE 57
Ambient1.013 P29.5 T85% RH
GT PRO 22.0 Desktop Administratorp [bar] T [C] M [kg/s], Steam Properties: IAPWS-IF97464 03-29-2012 23:43:51 file=D:\DOCUMENTS AND SETTINGS\RZAUNER\MY DOCUMENTS\SKM PROJECTS\EMA VESTING CONTRACTS 2012\GTPRO\SIEMENS 4000F\SIEMENS 4000F CCGT.GTP
GT PRO 22.0 Desktop AdministratorGross Power 389522 kWNet Power 381727 kWAux. & LossesAux. & Losses 7795 kWLHV Gross Heat Rate 6148 kJ/kWhLHV Net Heat Rate 6274 kJ/kWhLHV Gross Electric Eff. 58.55 %LHV Net Electric Eff. 57.38 %Fuel LHV Input 665263 kWthFuel HHV Input 736517 kWthNet Process Heat 0 kWth
Figure 18 Performance analysis - Siemens "F" class CCGT, clean-as-new
1.013 p97.28 T633.9 M
41.53 T97.33 M
HP
HPB130.3 p331 T72.18 M
459.3 T344 T
IP
IPB31.33 p236.3 T12.18 M
280.7 T249.3 T
LP
LPB3.8 p141.8 T14.31 M
196.5 T151.8 T
Cold Reheat
31.94 p369.9 T69.5 M
1.189 M
1.044 p595.6 T633.9 M
Natural gas 14.37 M665263 kWth LHV
SGT5-4000F(Curve Fit OEM Data Model #397)
@ 100% load
1.013 p29.5 T619.5 M85% RH
1.003 p29.5 T619.5 M
0.0779 p41 T96.14 M
to HRSG
3.382 p288.5 T11.98 M
27.6 p566 T81.68 M
Hot Reheat
Stop Valve
124 p566 T72.19 M