Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Evaluation of Gas Turbine Performance Alternatives For Indonesia Power

Download as pdf or txt
Download as pdf or txt
You are on page 1of 33

Evaluation of Gas Turbine Performance

Improvement Alternatives For Indonesia Power


in cooperation with the
Indonesian Institute of Sciences (LIPI) Technical Implementation Unit

 
Report No. 1246-001-01
 
   
January 28, 2015
Revision 1

Prepared for:

PowerPHASE LLC
801 Northpoint Parkway
West Palm Beach, FL 33407

Prepared by:

RRT SIGMA Engineering


1 Huntington Quadrangle, Suite 3S01
Melville, New York 11747-4401
631-756-4628
FAX: 631-756-1064
E-mail: info@rrtsigma.com
 

 
RRT SIGMA Engineering 

 
   
 
Evaluation of Gas Turbine Performance Improvement
Alternatives for Indonesia Power

in cooperation with the Indonesian Institute of Sciences (LIPI)


Technical Implementation Unit

Report No. 1246-001-01

REVISION HISTORY
 
Issue Issue Date Summary
Rev 0 01/27/15 Initial issue
Rev 1 01/28/2015 Added Bibliography, typo errors, Operating Costs tables

Report No. 1246-001-01 Page i


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement
Alternatives for Indonesia Power

 
LEGAL NOTICE AND STATEMENT OF CONFIDENTIALITY

This document was prepared by RRT SIGMA Engineering (“SIGMA”) solely for the benefit of
PowerPHASE (“Client”) for Indonesia Power. Neither SIGMA, Client nor their parent
corporations or affiliates, nor any person acting on their behalf: (a) makes any warranty,
expressed or implied, with respect to the use of any information or methods disclosed in this
document; or (b) assumes any liability with respect to the use of any information or methods
disclosed in this document.

Any recipient of this document, by their acceptance or use of this document, releases
Sigma, Client, their parent corporations and affiliates from any liability for direct, indirect,
consequential, or special loss or damage whether arising in contract, warranty, express or
implied, tort or otherwise, and irrespective of fault, negligence, and strict liability.

The information contained in this report is intended for the exclusive use of PowerPHASE
and Indonesia Power, and may contain confidential or privileged information. This
document has been prepared pursuant to Sigma proposal PowerPHASE 14-01 and the
Continuing Services Agreement, therein between PowerPHASE and SIGMA dated October
14, 2014, and is subject to the confidentiality agreement and terms and conditions contained
therein.

REPORT UPDATE

SIGMA HAS NO RESPONSIBILITY TO UPDATE THIS REPORT FOR ANY CHANGES


OCCURRING SUBSEQUENT TO THE FINAL ISSUANCE OF THIS REPORT
 
 
 

Report No. 1246-001-01 Page ii


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement
Alternatives for Indonesia Power

TABLE OF CONTENTS

Section No. Title Page

1.0  EXECUTIVE SUMMARY 1-1 


2.0  DISCUSSION 2-1 
2.3  Gas Turbine Overview............................................................................................................... 2-3 
2.3.1  Factors that Affect Gas Turbine Performance ................................................................. 2-3 
2.3.2  Inlet and Exhaust Losses ................................................................................................ 2-5 
2.3.3  Gas Turbine Performance Degradation .......................................................................... 2-5 
2.4  Gas Turbine Power Improvement Alternatives ......................................................................... 2-6 
2.4.1  Evaporative cooling ......................................................................................................... 2-6 
2.4.2  Fogging ............................................................................................................................ 2-7 
2.4.3  Wet Compression ............................................................................................................ 2-7 
2.4.4  Inlet Chilling ..................................................................................................................... 2-9 
2.4.5  Steam Injection .............................................................................................................. 2-10 
2.4.6  Air Injection .................................................................................................................... 2-10 
2.5  Evaluation of Alternatives for Indonesia .................................................................................. 2-12 
2.5.1  Evaluation Methodology ................................................................................................ 2-12 
3.0  RESULTS 3-1 
3.1  Systems evaluated .................................................................................................................... 3-1 
3.2  Unit 1: GT1-Alstom Atlantique PG5341 ................................................................................... 3-2 
3.3  Unit 2: GE MS5001 ................................................................................................................... 3-3 
3.4  Unit 3: Westinghouse 251-B11 ................................................................................................. 3-4 
3.5  Unit 4, Westinghouse 251-B11 ................................................................................................. 3-6 
4.0  CONCLUSIONS 1 
5.0  RECOMMENDATIONS 1 
6.0  BIBLIOGRAPHY 1 

List of Tables

Table 2-1: Pesanggaran Gas Turbine Installation Summary............................................................... 2-1


Table 2-2: Gas Turbine Performance Summary ................................................................................... 2-2
Table 2-3: Typical Effect of Inlet & Exhaust Losses on Performance (MS7000EA) .......................... 2-5
Table 2-4: GT Pro Cost Factors Recommended for Indonesia .......................................................... 2-13
Table 3-1: Equipment Ratings to Enable GT Uprates .......................................................................... 3-1
Table 3-2: GT2 Performance Enhancement Comparison .................................................................... 3-4
Table 3-3: GT2 Operating Cost Comparison, 2,000 hrs./year ............................................................ 3-4
Table 3-4: GT3 Performance Enhancement Comparison .................................................................... 3-6

Report No. 1246-001-01 Page iii


RRT SIGMA Engineering

TABLE OF CONTENTS

Section No. Title Page

Table 3-5: GT3-Operating Cost Comparison, 2,000 hours/yr. ........................................................... 3-6


Table 3-6 GT4 Performance Enhancement Summary ......................................................................... 3-7
Table 3-7: GT4 Operating Cost Comparison ......................................................................................... 3-7 
 
 
List of Figures

Figure 2-1: Gas turbine Cycle Schematic .................................................................................... 2-3 


Figure 2-2: Effect of Ambient Temperature on MS7000EA Gas Turbine Performance (typical) .. 2-4 
Figure 2-3: Altitude/Inlet Pressure Effect on Gas Turbine Performance for MS7000EA (typical) 2-4 
Figure 2-4: Typical Effects of Humidity on GT Performance ........................................................ 2-5 
Figure 2-5: Evaporative Cooling Components .............................................................................. 2-6 
Figure 2-6: Typical wet compression nozzle that creates ultra-fine water droplets ...................... 2-8 
Figure 2-7: Wet Compression depiction with 100% coating of the compressor components (Al-
Sati, 2014) ..................................................................................................................................... 2-9 
Figure 2-8: Typical wet compression demin pump skid with VFD ................................................ 2-9 
Figure 2-9: Typical turbine air chilling system with water cooling tower mounted above the chiller
& pump house (Turbine Air Systems) ......................................................................................... 2-10 
Figure 2-10: Air injection schematic ............................................................................................ 2-11 
Figure 2-11: Typical Air Injection System (Turbo PHASE)TM provided by PowerPHASE .......... 2-11 
Figure 3-1: Pesanggaran Unit 1, Alstom Atlantique PG 5341 ...................................................... 3-3 
Figure 3-2: Pesanggaran Unit #2, GE MS5000 ............................................................................ 3-3 
Figure 3-3: Pesanggaran Unit 3& 4 (typical) ................................................................................. 3-5 

List of Attachments

Attachment A-1: Unit 1 Performance Figures .............................................................. A-1


Attachment A-2: Unit 2 Performance Figures .............................................................. A-2
Attachment A-3: Unit 3 Performance Figures .............................................................. A-3
Attachment A-4: Unit 4 Performance Figures .............................................................. A-4

Report No. 1246-001-01 Page iv


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

1.0 EXECUTIVE SUMMARY

A study was conducted to compare various alternative gas turbine power enhancements for the
gas turbines located in Indonesia. The Pesanggaran site in Bali, Indonesia, which has four (4)
simple cycle gas turbines was chosen as a sample site to perform the comparison, an Alstom
Atlantique PG5341 rated at 21.35MW installed in 1985, a GE MS500L rated at 20.10 MW
installed in 1993, and two (2) Westinghouse CW-251B11 rated at 42.07 MW each. Turbines 1
and 2 were rated at 30C (86F) and units 3 and 4 (Westinghouse) were rated 27C (80.6F) and
83% relative humidity. or, The Pesanggaran site is located basically at sea level.

Six (6) technologies were review and compared at nine (9) different conditions at the request of
the representatives from the Indonesian Institute of Sciences. The nine comparisons were
1. Evaporative Cooling
2. Fogging
3. Inlet Chilling to 130C (55.40F)
4. Inlet Chilling to 150C (590F)
5. Inlet Chilling to 170C (62,60F)
6. 1% Wet Compression
7. 2% Wet Compression
8. Dry Air Injection
9. Humid Air Injection

Due to consistently high relative humidity of 83%, alternatives 1 and 2 offer no real benefit to
gas turbines installed in the area. Inlet chilling (3, 4, & 5) was found to be very capital intensive
and based on the remaining life of the equipment, not good investments. While the technology
of wet compression (6 & 7) offered the most power enhancement, they require the internal
compressor components to be coated and have a higher risk of compressor damage and
accelerated maintenance that can be extremely risky. Again, the limited remaining life of the
assets may not be a worth the significant capital investment, along with a demineralized water
plant addition to provide the demineralized water.

The analysis showed that dry air injection offered the best alternative for the Pesanggaran
simple cycle gas turbines. The benefits of dry air injection include (a) short downtime to install,
(b) low capital cost per kW, (c) introduces no additional contaminants into the gas turbine (d)
has the portability to relocate the system on any other, or future gas turbine(s) that might be
installed at the site. Humid Air injection would be considerably more expensive to introduce a
steam source at the site, along with high pressure steam piping to move the steam to the each
gas turbine package, as well as the additional cost of demineralized water to make up the
steam (humid air addition) that gets exhausted to the stack.

Report No. 1246-001-01 Page 1-1


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

2.0 DISCUSSION

2.1 Introduction

Indonesia Power (“IP”) supplied approximately 9.2 GW of the 30 GW of capacity


needed in the Java-Bali system of Indonesia during 2012 (Power, 2012).
Approximately 0.7 GW of this installed capacity is composed of simple cycle gas
turbines with a capacity factor of approximately 19% and availability of 90% while
approximately 2.7 GW of this installed base is gas turbine combined cycle with a
capacity factor of approximately 39%. Not all of the gas turbine installed capacity is
capable of being realized due to the prevailing warm weather conditions in Indonesia
and age of the original equipment. The country is only approximately 65% electrified,
with plans to increase electrification to 93% by 2025 (Rose, August 2010). In the
interests of better utilizing the existing gas turbine resource base, Indonesia Power
desires to evaluate gas turbine power output improvement options that can be applied
to the fleet of existing gas turbine assets. As a government owned utility, the Indonesia
power market is a regulated market. While efficiency is important to reduce costs,
capacity increase at the least cost is the primary goal.

The “test case” utilized in this analysis is the Bali Generation Business Unit
(Pesanggaran plant) located in southern Bali, on the east portion of Denpensar, Bali.
It is one of three generation plants owned and operated by PT Indonesia Power. The
other Bali stations are Gilimanuk Gas Power Generation on the West end of Bali
Island, and Pemaron Gas Turbine Power Plants (GTPP) located at Northern Beach of
Bali. Bali is one of the larger islands of Indonesia, located immediately to the East of
the island of Java. Bali’s installed capacity is approximately 433 MW. Supplemental
power is provided to Bali from Java via an underwater cable. Currently the load peaks
on Bali during the early and late evening when the tourist hotels demand the most
electricity. The Pesanggaran facility contains four operating gas turbines as well as
reciprocating engines. A new Wartsila diesel engine based plant totaling
approximately 200 MW is also under construction at the site. Table 1 summarizes the
four gas turbines, manufacturer, year installed, design capacity and design ambient
conditions located at the Pesanggaran site.
Table 2-1: Pesanggaran Gas Turbine Installation Summary
Design
OEM Model COD Capacity [MW] Design Conditions
Unit 1
Alstom-
PG5341 1985
21.35 (base) 30C, n.s. % rh1,
Atlantique 23.05 (peak) 1.013B
20.10 (base)
Unit 2 GE MS500L 1993
23.05 (peak 30C, n.s. % rh2,
Unit 3 Westinghouse CW-251-B11 1994 42.07 27C, 83 % rh,
Unit 4 Westinghouse CW-251-B11 1994 42.07 1013 mbar, 0.8 pf

1 A relative humidity was not specified in the design documents


2 A relative humidity was not specified in the design documents

Report No. 1246-001-01 Page 2-1


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

For each gas turbine at the Pesanggaran Station, Table 2 shows the most recent plant
performance test report date and time, test results and ambient temperature. The gas
turbine performance tests reported by IP were not corrected to the design ambient
conditions of temperature, barometric pressure and humidity. Since a gas turbine’s
power and heat rate are affected by these conditions, to properly compare test results,
performance must be corrected to a consistent set of ambient conditions, usually the
initial design conditions. The manufacturer normally provides these correction curves
with the gas turbine manuals. The plant library data was reviewed but only the
performance correction curves for GT2 were located, which were mostly illegible.

2.2 USE of GT Pro Software

GT Pro, a power cycle thermodynamic computer modeling software program, was


utilized to conduct the performance evaluation of the various alternatives. In order to
determine the baseline power production, a gas turbine of the same type is simulated
using the GTPro power plant modeling software at the design conditions, if known, for
inlet temperature, humidity and barometric pressure. The gas turbine is then run at an
“off-design” case, i.e. at the ambient temperature for that unit’s latest plant
performance test results. Only the temperature correction was applied as the ambient
conditions of humidity and barometric pressure were not recorded during the plant
tests and were not available in any of the site recorded data. The ratio of the
performance test result divided by the GTPro predicted new and clean (N&C) gross
power output indicates the shortfall in gas turbine performance since new. Subtracting
this ratio from 1.0 yields the apparent degradation. The study could be refined in the
future to include the effects of barometric pressure and humidity. IP Staff personnel
were attempting to retrieve test conditions for these parameters, as well as confirm
temperature, from the Airport weather services which is located only a few kilometers
away. Note that all of the devices used to measure power, temperature and fuel flow
at the plant have an inherent uncertainty which has not been analyzed, so an
uncertainty analysis is not available.

Table 2-2: Gas Turbine Performance Summary


N&C
Last Test Aux Test Capacity at Apparent
SFC
GT Capacity Results Loads Conditions Test Degradation
[l/kWh]
Test Date [MW] [kW] [°C] Temperatur [%]
e [MW]
17DEC201
1 17.8 TBD 0.453 TBD 21.35 TBD
15JAN2014
2 19:00-20:00
16.30 220 0.422 30C 20.10 23.6%
06JAN2014
3 19:00-20:00
39.80 161 0.339 30C 40.77 2.4%
19JAN2014
4 19:00-20:00
35.40 98 0.404 30C 40.77 13.2%

Report No. 1246-001-01 Page 2-2


RR
RT SIGMA Engineering
E g Evalu
uation of Gas
s Turbine
Performance Impro
ovement Alte ernatives
for Indones
sia Power

2.3 Ga
as Turbine Overview
O

The basic gaas turbine ope erating cycle is also calledd the Brayton Cycle and is depicted belo ow. It
consists of a compression stage, a he eat addition (ccombustion) stage, and an expansion stage
(turbine). Th
he turbine (exxpander) thru ust (power) iss a function o of mass flow w. Air enterinng the
compressor at point 1 is s compressed d to some hig gher pressure e. No heat iss added; how wever,
compression n raises the air temperature e so that the aair at the disccharge of the compressor iis at a
higher temperature and pressure. Up pon leaving tthe compresssor, air enters the combu ustion
system at point
p 2, where fuel is inje ected and co ombustion occcurs. The co ombustion process
occurs at esssentially connstant pressure. Although high local te emperatures a are reached within
the primary combustion
c zone (approac ching stoichioometric condittions), the com mbustion systtem is
designed to provide mixiing, burning, dilution and cooling. Thu us, by the timme the combu ustion
mixture leavees the combu ustion system m and enters th he turbine at point 3, it is a
at a mixed avverage
temperature. In the turbinne section of the gas turbin ne, the energ gy of the hot ggases is convverted
into work. Soome of the wo ork developed by the turb ine is used to o drive the co
ompressor, an nd the
remainder iss available for useful work k at the outpu ut flange of thhe gas turbin ne. Typically, more
than 50% of the work developed
d by the turbine section(s) iss used to pow wer the axiaal flow
compressor. In the ideal cycle
c (a) the compression
c is performed adiabaticallyy (no heat trannsfer),
(b) the comb bustion is performed with no change i n pressure (iisobaric) and (c) the expa ansion
occurs adiab batically as well.
w In reality friction an
nd heat loss//gain results in less than ideal
performance e.

Figure 2-1: Gas turbinee Cycle Schem


matic

2.3.1 Facttors that Affe


ect Gas Turb
bine Perform
mance

Gas turbines s catalog performance is normally ratted at Interna ational Standdards Organizzation
(ISO) conditions of 59F (15 C), 14.7 psigp (1.013Baar) and 60% rrelative humid dity. Since th
he gas
turbine is essentially a constant volume air-brea athing machine, its perfo ormance varie es by
anything that affects the density
d and/oor mass flow o mpressor. Figure 2-
of the air intake to the com
2 shows how w ambient tem mperature affeects the outpuut, heat rate, heat consum mption, and exxhaust
flow of a sing
gle-shaft MS7 7001 (Frank J.
J Brooks, 20 000). When th e inlet temperature
he gas turbine
increases, thhe air is less dense,
d this re
esults in redu
uced turbine o output. Each gas turbine m model
has its own temperature--effect perforrmance curve es, driven by the Brayton cycle param meters,
hardware effficiencies and d air mass flow
w.

Report No. 1246-001-01 Page 2-3


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

Figure 2-2: Effect of Ambient Temperature on MS7000EA Gas Turbine


Performance (typical)

Correction for altitude or barometric pressure is more straightforward. The air density reduces
as the site elevation increases. While the resulting airflow and output decrease proportionately,
the heat rate and other cycle parameters are not affected. A standard altitude/inlet pressure
correction curve is presented in Figure 2-3. The reduced performance due to inlet pressure
drop can also be estimated using the same curve.

Figure 2-3: Altitude/Inlet Pressure Effect on Gas Turbine Performance for


MS7000EA (typical)
Similarly, humid air, which is less dense than dry air, also affects output and heat rate, as
shown in Figure 2-4.

Report No. 1246-001-01 Page 2-4


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

Figure 2-4: Typical Effects of Humidity on GT Performance

In the past, this effect was thought to be too small to be considered. However, with the
increasing size of gas turbines and the utilization of humidity to bias water and steam injection
for NOx control, this effect has greater significance.

2.3.2 Inlet and Exhaust Losses

Inserting air filtration, silencing, evaporative coolers or chilling coils into the inlet or adding heat
recovery devices in the exhaust ductwork causes permanent pressure losses in the system.
The effects of these pressure losses are unique to each gas turbine design. Table 2-1 below
shows typical inlet and outlet pressure losses for a GE Frame 7EA. These pressure losses
occur regardless of whether the performance enhancement is operational or not, for example
part load conditions.

Table 2-3: Typical Effect of Inlet & Exhaust Losses on Performance (MS7000EA)
4 inch H2O (10mBar) 4 inch H2O (10mBar)
inlet Loss outlet Loss
Power Output
1.42% 0.42%
Reduction
Heat Rate Increase 0.45% 0.42%
Exhaust Temperature
1.9F (1.1C) 1.9F (1.1C)
increase

2.3.3 Gas Turbine Performance Degradation

All turbo-machinery experience losses in performance with time. Gas turbine performance
degradation can be classified as either recoverable or non-recoverable loss. Recoverable loss
is usually associated with compressor fouling and can be partially rectified by water washing or,
more thoroughly, by mechanically cleaning the compressor blades and vanes after opening the
unit. Non-recoverable loss is due primarily to increased turbine and compressor clearances

Report No. 1246-001-01 Page 2-5


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

and changes in surface finish and airfoil contour. Because this loss is caused by reduction in
component efficiencies, it cannot be recovered by operational procedures, external
maintenance or compressor cleaning, but only through replacement of affected parts at
recommended inspection intervals.

2.4 Gas Turbine Power Improvement Alternatives

Various gas turbine performance enhancements exist that attempt to negate or minimize the
negative performance impact of hot ambient conditions. Some options have secondary
maintenance, operations, performance and support equipment requirements which are
described below.

2.4.1 Evaporative cooling

Evaporative cooling is a passive process. In an evaporative cooling system a wet


media is installed in the cross-section of the gas turbine filter house. The media is kept
wet using high quality water, such as that from a reverse osmosis unit. The air entering
the filter house passes over the saturated media, and the water contained in the media
evaporates into the air stream on its way to the gas turbine. The extent of the
evaporation is inversely proportional to the percent humidity in the air stream.

Figure 2-5: Evaporative Cooling Components

Report No. 1246-001-01 Page 2-6


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

The lower the humidity the more easy it is for the water to evaporate. The heat to
vaporize the water is taken from the ambient air, leaving the air stream cooler. Cooler
air results in more dense air, which corresponds to more mass flow, which results in
greater output in gas turbines (recall power is function of mass flow). The higher the
humidity, the harder it is for the water to evaporate. However to arrive at 100%
saturation technically requires an infinite contact time with the saturated media. The
evaporative cooling effectiveness is the percentage of water expected to be evaporated
divided the water needed to achieve saturation. A typical effectiveness is 85% to 95%.
In the climate of Indonesia, the humidity is high, averaging 83% -. With this humidity the
dry bulb depression achievable using evaporative cooling is very low, approximately
40F at the design condition of 900F /85% relative humidity.

Evaporate cooling is inexpensive in a new installation as it requires only the cooling


media section to be added to the inlet filter house and a water circulation pump. In a
retrofit, the installation requires the modification of the inlet filter house to accommodate
the evaporative cooling media. This requires re-engineering, and service interruption.
The application of evaporative cooling system is well suited for climates with low
humidity. A drawback of the system is when the ambient humidity is close or at
saturation (raining) the system is totally ineffective at improving gas turbine
performance.

Another important consideration is the permanently added pressure drop addition to the
inlet filter house. If at any time the operation of the evaporative cooler is not warranted,
the air will still flow over the permanently installed media, and impose a pressure drop
at all times, even during part load operation. Evaporative cooling systems require a
few minutes to achieve 100% saturation of the inlet media.

2.4.2 Fogging

Fogging is an active form of an evaporative cooling system. With fogging pressurized


water is sprayed into the air stream in ultra-fine droplets of approximately 10 microns in
size. The amount of water spray is typically more than 100% of that needed to achieve
saturation. The excess water compensates for the residence time so that 100%
effectiveness can be achieved. Fogging has the same attributes and limitations as
evaporative cooling but virtually no additional inlet pressure drop since wet media is not
used. A small amount of parasitic load is required to create the pressure water (2000
psig/138 bar). Since water is forced into the air stream, demineralized water is
recommended as normal water would enable deposits to enter the gas turbine which
would harm the turbine internal components. While the overspray conditions of fogging
are more likely to approach saturation, the extra water can present problems,
especially in retrofit installations. The extra water will drain on the inside walls, collect
dirt which will get sucked into the gas turbine fouling the gas turbine. Since fogging
systems are water droplets, the water must be demineralizer water or they stand to
damage the compressor with dissolved minerals. Fogging systems can be online very
quickly.

2.4.3 Wet Compression

Report No. 1246-001-01 Page 2-7


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

In an application using wet compression, demineralized pressurized water is injected


close to the inlet of the gas turbine and/or in various stages of the gas turbine
compressor. The water nozzles create an ultra-fine droplet size water mist.

Figure 2-6: Typical wet compression nozzle that creates ultra-fine water droplets

At the inlet some of the water evaporates quickly cooling the inlet air similar to
evaporative cooling or fogging. The water injected above that needed to achieve
saturation is drawn into the compressor blade path and evaporates during the various
stages of compression. Wet compression has the advantage of providing the effect of
interstage cooling as water evaporates in each stage. Evaporation of the water
droplets in the compressor blade path causes the air temperature to drop and thereby
reduces the power consumption of the compressor because less energy is required to
compress cool air compared to warm air with the same mass. This translates into a
decrease in turbine work because one-half to two-thirds of a turbine’s output is typically
used to drive the compressor. The result is more turbine power for the generation of
electricity and improved gas turbine efficiency.

Water that does not completely evaporate impinges on the compressor blades and can
easily cause pitting and premature damage to the compressor blades, particularly the
leading edges. When an early stage of a compressor fails, the broken pieces go
downstream resulting in a multi-million dollar repair bill and at least 8 – 12 weeks out of
service. Therefore proper applications require the compressor blades and stators to
be coated to protect the surfaces for the steam and water impact. Siemens uses an
Advanced Compressor Coating on gas turbine hardware to minimize the damage of
wet compression (Crampsie, March-April 2012). Considerable downtime would be
required to coat the compressor components. The compressor rotor and stator
components would have to be shipped off site to an authorized service shop to apply
the advanced coating. This would results in a few weeks of downtime. Alternatively this
work could be conducted during a scheduled major overhaul.

Wet compression systems can be on line very quickly.

Report No. 1246-001-01 Page 2-8


RR
RT SIGMA Engineering
E g Evalu
uation of Gas
s Turbine
Performance Impro
ovement Alte ernatives
for Indones
sia Power

Figure 2-7: Wet


W Compresssion depictiion with 100%% coating of tthe compresssor
componen nts (Al-Sati, 22014)

Main
ntenance is also
a he droplet sizze increases which
critical. As water nozzzles foul, th
resu
ult in water dro
oplets impacting the comp ressor bladess.

Figure
e 2-8: Typicall wet compre ssion demin pump skid w
with VFD

The wet compression system m consists of a skid moun nted water p


pump with va ariable
frequuency drive, using demin neralized wate er supplied ffrom a waterr treatment fa acility.
The pump discha arge is connec cted to a commpressor inlet spray systemm. The gas tu urbine
conttrol logic is adjusted
a to accommodate e the wet com mpression opperating limitss and
integ
grated with th he balance ofo plant logicc. Installation n and commissioning of a wet
compression sys stem usually requires an outage of 2--4 weeks as well as a fu ull site
survvey (Crampsie e, March-Aprril 2012). The e installation cost of a deemineralized water
systeem, deminera alized plant operation
o (wa
ater, chemica als and labor)) and maintennance
cost also need to be included in i a system e evaluation.

2.4.4 Inlett Chilling

With oil is installed in the cross section of th


h inlet chilling a cooling co he gas turbine e inlet
filterr plenum. Co old water (generated from m a chilled wa ater source) is pumped through
the coil
c as the ga et air flows ovver the coil. The advantag
as turbine inle ges of inlet cchilling

Report No. 1246-001-01 Page 2-9


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

system is the ability to reduce the inlet of the gas turbine to a desired temperature,
generally ISO conditions or slightly lower during all times of operation. A source of the
chilled water is required. The chilling system can be an electric driven chiller(s) or
absorption chiller(s). The chilling system will require a way to reject heat, which, in
most cases is a wet cooling tower. The wet cooling tower uses the vaporization of
water to the atmosphere to take away the heat of the cycle driver and the cooling
effect. A disadvantage to inlet chilling is the permanent pressure drop in the inlet filter
house that will permanently reduce the GT output year round, regardless of the
whether the inlet chilling system is operational, or not, even during part load operation.
Inlet chilling system are generally very slow at responding, due to the thermal inertia of
the warm water sitting in the piping before startup. One method to avoid this is to add a
very small chiller to run 100% of the time to simply keep the chilled water piping system
cold.

Figure 2-9: Typical turbine air chilling system with water cooling tower mounted
above the chiller & pump house (Turbine Air Systems)
2.4.5 Steam Injection

In this system steam is injected at the combustor. The pressure of the steam must be
high enough to match the compressor discharge pressure. A source of the steam is
required, usually from the HRSG or steam turbine extraction of a combined cycle plant.
The steam is 100% lost to the atmosphere as it proceeds through the turbine and up
the stack. Therefore use of steam injection requires a demineralizer system adequate
to provide for the makeup water volume. Steam (as with air) controls the flame
temperature and results in lower NOx emissions. Steam injection normally is only
attractive when (a) it is being generated in a combined cycle or HRSG (b) free steam is
available, or (c) required for emissions control. The introduction of water or steam in
the blade path of the gas turbine will impact the chemistry of turbine components to
various degrees. Depending on the source of the steam, steam injection can be slow in
reaching full benefit, sometime as much as 1 hour. Steam injected is not considered at
the Pesanggaran site because there is no steam capability.

2.4.6 Air Injection

Air injection has a separate, stand-alone, air compressor designed to match the gas
turbine’s compressor discharge conditions. The system can either be a humid air
injection (HAI) or dry air injection (DAI) system. In general DAI provides greater cycle

Report No. 1246-001-01 Page 2-10


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

efficiency.. The DAI system uses atmospheric air less the moisture that condenses in
the compressor intercooler. The intercooler substantially improves the compressor and
reduces the compressor size requirements. Dry air introduces no additional chemistry
concerns as with water or steam injection, and therefore the life of the gas turbine
components is not impacted whatsoever. The DAI system can respond quickly to
power demands, coming on line in less than a few minutes. DAI systems are external
to a gas turbine package and the compressed air is single line to each gas turbine.
Therefore a DAI system for multiple gas turbines at the same site can be provided from
one, DAI system “manifold”. The DAI System manifold can then take advantage of
economies of scale, for example using five (5) DAI air compressors serving four (4) gas
turbines, with one air compressor serving as a backup to any of the other four
compressors. Also since they are external to the package, the DAI System can be
applied to any future replacement gas turbines, or relocated to another gas turbine
installation if the Pesanggaran site becomes no longer viable.

Figure 2-10: Air injection schematic

Figure 2-11: Typical Air Injection System (Turbo PHASE)TM provided by


PowerPHASE

Report No. 1246-001-01 Page 2-11


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

The HAI system is essentially a combination of high pressure steam injection diluted
with dry air injection. HAI can provide greater output than DAI, but at less efficiency.
HAI is most common in a combined cycle where steam is already being produced. HAI
has the added cost of demineralized water make-up, water chemistry, steam piping,
and of course make up water cost. HAI introduces the potential concerns for
water/steam chemistry impacting the life of the gas turbine components. HAI systems
can take a while to be fully functional, depending on the source of the high pressure
water vapor.

Manufacturer’s published general limitations on excess air or steam flow is 5%.


However on older gas turbines, the surge margins are generally more conservative and
therefore higher DAI and HAI mass flows are likely possible.

2.5 Evaluation of Alternatives for Indonesia

2.5.1 Evaluation Methodology

GT Pro software, under license from ThermoFlow of Massachusetts, was utilized to develop
comparative performance and cost estimates for the each of the above technologies on the
family of Bali gas turbines installed at Pesanggaran.

The following methodology was employed:

1) The first step is to replicate the New and Clean (N&C) performance of the each gas
turbine at the design ambient conditions when the unit was installed using the GTPro
computer performance modelling software. Site specific conditions are entered such as
generator voltage, line voltage, site specific fuel composition, GT starter mechanism, and
others. For the case of GT3, this value is 42.07 MW at 27C (80.6F). Using 4 inches inlet
pressure drop and 4 inches outlet pressure drop the GTPro performance calculates
41.743 MW output which is close enough.

2) Using GTPro, the model arrived at in step (1) is run at the ambient conditions of the most
recent gas performance test provided by the plant. For GT3, this was 30C (86F) on
06JAN2014. Unfortunately, the plant only records the ambient temperature and does not
record the atmospheric pressure or humidity, both of which also affect the gas turbine
performance, albeit in a secondary effect. Therefore any off-design performance impacts
due to barometric pressure and humidity could not be corrected for. Historically, the
Plant does not correct for the off-design conditions of temperature, atmospheric pressure
or humidity when conducting any of the performance tests. While the pressure and
humidity have less influence on power and heat rate, all performance test results should
be corrected for all three parameters. In the N&C case, the predicted performance is at
30C is 40.768 MW.

3) By comparing the corrected performance in step (2) divided by the performance in step
(1), a degradation factor is determined. For GT3, the test value was 39.8 MW indicating
a degradation of 2.4%. The calculated degradation is then permanently applied to the
GTPro Model which establishes a performance baseline to which to compare each
performance enhancement being considered.

Report No. 1246-001-01 Page 2-12


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

4) Using the GTPro capital cost estimating segment of the computer program (PEACE), an
estimate of the cost to install the gas turbine in current year dollars is determined. It was
agreed to use the default cost multipliers for Indonesia provided in GTPro for
commodities, equipment, labor and materials. The cost (“PEACE”) factors are listed in
the table below. The costs of the baseline GT determined establishes the new and clean
cost to install the gas turbine without any options, as constructed. The baseline cost
prediction in GTPro is $US 33,379k for GT3.

Table 2-4: GT Pro Cost Factors Recommended for Indonesia


Category Factor for Indonesia
Specialized Equipment 1.05
Other Equipment 0.75
Commodities 0.65
Labor 0.54

5) Each GT will then be modelled in GTPro with each of the enhancements under
consideration. For each enhancement, the parameters of (a) Power Output, (b) Fuel
Consumption and (c) total installed cost of the GT are recorded. The respective values
are compared to the baseline GT (a) Power Output with apparent degradation, (b) fuel
consumption and (c) total installed cost. The difference between the respective
parameters is the estimated first cost to install the enhancement if new. Some initial
costs require additional adjustments not modelled in GTPro including demineralized
water plant, compressor coatings, source of steam or air injection which must be added.
The initial cost will be slightly higher since GTPro is comparing total plant cost, not
calculating the incremental cost for a retrofit. Since all of the installations considered are
retrofits, an adjustment cost to mobilize, demobilize, demolishment etc. of 10% has been
added to each enhancement option. Other less tangible aspects of the alternatives are
also shown. These include
a. Lead time: the time from purchase order to receipt of equipment for installation at
the project site.
b. Installation time: the time the installing contractor will on the site to install the
equipment
c. Downtime: the time the gas turbine must be off-line and unavailable for dispatch
to complete the installation.
d. Foot print: the land area required to install a typical system. This does not include
interconnecting piping routes which may be installed overhead or underground.
e. O&M Costs: the additional O&M water costs estimated to support the system

Report No. 1246-001-01 Page 2-13


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

3.0 RESULTS

3.1 Systems evaluated

Based on the aforementioned information and procedure outlined, the following are the results
of the GTPro simulations conducted based on the test data and calculations described in
Table 1-2. For Unit 3, the evaluation procedure was conducted at the Pesanggaran site office
with the representatives from Indonesia Institute of Sciences Technical Implementation Unit
for Instrumentation Development, Mr. Hilman Syaeful Alam M.Eng, and Mr. Imam Djunaedi.
Unit 1 test information was not provided with the test temperature. Once the ambient
temperature conditions are known, the table will be provided for Unit 1.

Each unit was checked for generation constraints to ensure it can accommodate additional
capacity. Below is a table which compiles the manufacturer’s data for the main components of
the system that would be used to deliver the incremental power, namely the gas turbine, the
load gear, the electric generator and the transformer. For each application, further study
would require confirmation that the electrical bus, 11kV bus, circuit breaker and HV cables are
sufficiently sized to handle the extra power. Based on this research, none of the technologies
evaluated pose a threat to exceeding any of the major equipment ratings on any of the gas
turbines installed.

Table 3-1: Equipment Ratings to Enable GT Uprates


UNIT 1 UNIT 2 UNIT 3 UNIT 4
base peak base peak
GAS TURBINE MW 21.35 23.05 20.1 23.05 42.07 42.07
LOAD GEAR MW 27 31.5 80.46 80.46
ELECTRIC GENERATOR MVA 37 37 28.358 30.4 61.375 61.375
BUS 1 (a) (a) (a) (a) (a) (a)
BREAKER (a) (a) (a) (a) (a) (a)
BUS 2 (a) (a) (a) (a) (a) (a)
TRANSFORMER (Manufacturer) MVA 27 (Alstom) 27 (Hunday) 60 (Hunday) 60 (Hunday)
(a) Not confirmed

Evaporative Cooling: Installing an in-line evaporative cooling media would require significant
downtime, and because of the prevalent humidity in the region, it does not produce much
incremental power. A major drawback is (a) permanent pressure drop of the evaporative
cooling media (even during part load GT operation) and (b) output is inversely proportional to
the humidity, which is high in the country.

Fogging System: As with the evaporative cooling system, installing a fogging system would
require significant downtime, and because of the prevalent humidity in the region, it does not
produce much incremental power. Fogging also requires the use of demineralized water, an
added cost for its production, storage and piping. The Plant provided the cost to generate
demineralized water in its current plant as is 576.07 Rupias/liter.

Chilling System: Three cases are evaluated for inlet chilling systems, chilling to 130C, 150C and
170C gas turbine inlet temperature. The resulting air conditioning system load for the respective
system is shown. The system would use package chilling systems with “roof mounted” water

Report No. 1246-001-01 Page 3-1


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

cooling towers to reject the heat. No redundancy of equipment is included. The cooling tower
makeup requirements are also provided.

Wet Compression System: Two cases are modeled, 1% and 2 % wet compression. In addition
to the skid cost for wet compression, the cost of a demineralized water plant, storage tanks for
raw and demin water for 0.5 days of operation, plus the estimated cost to coat the compressor
is added to the implementation cost. The demineralized water required for each wet
compression systems is provided. Considerable downtime would be required to coat the
compressor components, which would have to be shipped off-site to an authorized service
shop to add the advanced coating. This would results in a few weeks of downtime. Alternatively
this work could be conducted during a scheduled major overhaul.

Dry Air Injection (DAI): The dry air injection is modelled based on 5% of ISO air flow conditions,
e.g. 0.05 x 374.6 lb/sec = 18.73 lb/sec of air injection for G3/G4. The air is assumed to be
available from the skid at 220 psig and 600 F. Greater flow can be achieved with the older gas
turbines which were design with more conservative stall margins.

Humid Air Injection (HAI): A performance was evaluated assuming 10% of the dry air flow used
above in DAI is provided with steam at similar conditions. The performance improvement is
shown in the last column. Since there is no steam system at the site, heat recovery steam
generator (HRSG) or steam turbine to extract steam from, a 90% efficient (LHV) boiler is
assumed to be used as a source of the steam flow. The air flow is taken as 90% of the DAI
compressor module heat input for providing the air flow. Additional parasitics for the new boiler
was also added. As can be seen while the power boost is greater for HAI than DAI, the
incremental heat rate is not as attractive as dry air. The incremental cost estimate for a boiler,
installation, steam piping, makeup water piping and fuel piping is included in the capital cost
along with 90% of the DAI cost for the air system. Other sites that have gas turbines in
combined cycle may be able to avoid the cost of the boiler. However note, steam removed
from a steam turbine (closed cycle) to be injected into a gas turbine (open cycle) adds
additional make-up water cost to the steam turbine’s closed system.

3.2 Unit 1: GT1-Alstom Atlantique PG5341

Performance test data pending for comparison purposes.

Report No. 1246-001-01 Page 3-2


RR
RT SIGMA Engineering
E g Evalu
uation of Gas
s Turbine
Performance Impro
ovement Alte ernatives
for Indones
sia Power

Fig
gure 3-1: Pesa
anggaran Uniit 1, Alstom A
Atlantique PG
G 5341

3.3 Unit 2: GE MS
S5001

Figure 3-2:
3 Pesanggarran Unit #2, G
GE MS5000

Below is a summary tab


ble comparing ement technollogies to the Unit 2
g the various GT enhance
gas turbine
e.

Report No. 1246-001-01 Page 3-3


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

Table 3-2: GT2 Performance Enhancement Comparison


water/wet water/wet
Evap Chilling, Chilling, Chilling, compressio compressio Dry Air Humid Air
No Parameter Units Cooling Fogging to 13 C to 15 C to 17 C n n Injection Injection
1 Cost New k.USD 22,656 22,577 25,224 24,958 24,700 23,040 23,295 22,253 22,253
Demin Plant/RO k.USD 35 75 405 492 n/a 175
Compressor component recoating k.USD 700 700 n/a n/a
2 Cost Base Plant k.USD 22,232 22,232 22,232 22,232 22,232 22,232 22,232 22,232 22,232
3 Technology Estimated Cost, USD k.USD 505 462 3,291 2,999 2,715 2,104 2,481 1,171 1,629
(includes 10% for retrofit vs. new premium)

4 New Power kW 16,548 16,615 18,579 18,271 17,991 18,648 20,420 18,365 18,430
5 Base Power kW 16,300 16,300 16,300 16,300 16,300 16,300 16,300 16,300 16,300
6 New Parasitic kW 7 9 1,311 1,140 966 34 99 20 40
7 Additional Net Power Output kW 241 306 968 831 725 2,314 4,022 2,045 2,090

8 Capacity cost USD/kWh $ 2,094 $ 1,511 $ 3,399 $ 3,608 $ 3,742 $ 910 $ 617 $ 573 $ 779

9 Lead time, month months 12 6 12 12 12 16 16 6-8 10-12


10 Installation Time, days weeks 6-8 4-6 5-7 5-7 5-7 12-16 12-16 6-8 10-12
11 Down time, days days 12-16 6 15-20 15-20 15-20 14-28 14-28 2-3 2-3

New GT Fuel Consump. [LHV] kBTUY/hr 224,509 225,024 241,720 239,291 236,869 241,197 284,236 227,846 228,103
Non-GT Fuel consump [LHV] kBTUY/hr - 16,719 20,903
Base Fuel Consump. [LHV] kBTUY/hr 221,898 221,898 221,898 221,898 221,898 221,898 221,898 221,898 221,898
12 Incremental Fuel Cons, Btu/hr [LHV] kBTUY/hr 2,611 3,126 19,822 17,393 14,971 19,299 62,338 22,667 27,108
13 Incremental heat rate, Btu/kWh [LHV] BTU/kWh 10,830 10,222 20,469 20,930 20,638 8,342 15,501 11,084 12,970
Specific Fuel Consump.
[asssumes 19,500 btu/lb] litre/kWh 0.300 0.283 0.568 0.580 0.572 0.231 0.430 0.307 0.360
14 Additional O/M cost, USD/kWh US$/kWh - 0.068 - - - 0.091 0.099 - 0.048
Demin Water Cost (Rp 576.07/liter)
15 Water needs lbs/sec 0.2546 0.2659 6.654 5.838 5.18 2.672 5.078 0 1.273
16 Cooling Capacity AC Tons n/a n/a 1627 1435 1233 n/a n/a n/a n/a
17 Permanent Land requirement ft2 70 500 124 100 90 1649 2401 285 713
Permanent Land requirement m2 7 48 12 10 9 159 231 27 69

When operated for a hypothetical 2,000 hours/year, the incremental power, fuel and
demineralized water costs are tabulated below.

Table 3-3: GT2 Operating Cost Comparison, 2,000 hrs./year

Assumptions
Fuel Cost [US$/gal] $ 2.75
Fuel Oil Heating Value[BTU/gal HHV 132,000
LHV/HHV ratio 0.96
Annual hours of operation [hours/year] 2,000
1% 2%
water/wet water/wet
Evap Chilling, Chilling, Chilling, compressio compressio Dry Air Humid Air
Parameter Units Cooling Fogging to 13 C to 15 C to 17 C n n Injection Injection
Electric Energy Generated [kWh] 482,200 611,600 1,936,800 1,662,000 1,450,800 4,627,200 8,043,000 4,090,000 4,180,000

Fuel Cost [US$] $ 113,325 $ 135,677 $ 860,330 $ 754,905 $ 649,783 $ 837,630 $ 2,705,642 $ 983,799 $ 1,176,548
Demin water Cost [US$] $ - $ 41,776 $ - $ - $ - $ 419,798 $ 797,805 $ - $ 200,001
Operating Cost [US$] $ 113,325 $ 177,453 $ 860,330 $ 754,905 $ 649,783 $ 1,257,428 $ 3,503,447 $ 983,799 $ 1,376,549

average power cost [$/kW] $ 0.235 $ 0.290 $ 0.444 $ 0.454 $ 0.448 $ 0.272 $ 0.436 $ 0.241 $ 0.329

3.4 Unit 3: Westinghouse 251-B11

Report No. 1246-001-01 Page 3-4


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

Figure 3-3: Pesanggaran Unit 3& 4 (typical)

Below is a summary table comparing the various GT enhancement technologies to the Unit 3
gas turbine. The apparent degradation of Unit 3 based on the last performance test of 39.8 MW
(06JAN2014) is 2.4%. The ambient test conditions indicated 30C. This degradation is better
than expected.

Report No. 1246-001-01 Page 3-5


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

Table 3-4: GT3 Performance Enhancement Comparison

1% 2%
Evap Chilling, Chilling, Chilling, water/wet water/wet Dry Air Humid Air
b
No Parameter Units Cooling Fogging to 13 C to 15 C to 17 C compression compression Injection Injection
1 GT Cost New w/Enhancement k.USD 33,905 33,793 36,788 36,385 35,990 34,426 34,761 36,041 36,225
Demin/RO Plant 50 100 276 412 n/a 250
Compressor component recoating 750 750 n/a n/a
2 Cost Base Plant w/o enhancement k.USD 33,379 33,379 33,379 33,379 33,379 33,379 33,379 33,379 33,379
3 Technology Estimated Cost, USD k.USD 634 565 3,750 3,307 2,872 2,280 2,798 2,662 3,096
(includes 10% for retrofit vs. new premium)

4 New Power kW 41,316 41,522 44,830 44,179 43,498 46,219 50,244 45,413 45,553
5 Base Power kW 40,743 40,743 40,743 40,743 40,743 40,743 40,743 40,743 40,743
6 New Parasitic kW 10 16 1,394 940 894 53 92 20 40
7 Additional Net Power Output kW 563 763 2,693 2,496 1,861 5,423 9,409 4,650 4,770

8 Capacity cost USD/kWh $ 1,125 $ 741 $ 1,392 $ 1,325 $ 1,543 $ 420 $ 297 $ 573 $ 649

9 Lead time, month months 12 6 12 12 12 16 16 6-8 10-12


10 Installation Time weeks 6-8 4-6 5-7 5-7 5-7 12-16 12-16 6-8 10-12
11 Down time, days days 12-16 6 15-20 15-20 15-20 14-28 14-28 2-3 2-3

New GT Fuel Consump. [LHV] kBTUY/hr 455,563 456,981 481,008 476,365 471,316 489,995 520,587 467,084 467,855
Non-GT Fuel consump [LHV] kBTUY/hr 24,599 30,755
Base Fuel Consump. [LHV] kBTUY/hr 451,042 451,042 451,042 451,042 451,042 451,042 451,042 451,042 451,042
12 Incremental Fuel Consumption [LHV] kBTUY/hr 4,521 5,939 29,966 25,323 20,274 38,953 69,545 40,641 47,568
13 Incremental heat rate [LHV] BTU/kWh 8,026 7,784 11,126 10,145 10,894 7,183 7,391 8,740 9,972
Specific Fuel Consumption
[asssumes 19,500 btu/lb] litre/kWh 0.223 0.216 0.309 0.281 0.302 0.199 0.205 0.242 0.277
14 Additional O/M water cost US$/kWh - 0.037 - - - 0.057 0.063 - 0.031
Demin Water Cost (Rp 576.07/liter)
15 Water needs lbs/sec 0.33 0.363 7.2 5.9 4.73 3.952 7.54 n/a 1.873
16 Cooling Capacity AC Tons n/a n/a 1780 1410 1060 n/a n/a n/a n/a
17 Permanent Land requirement ft2 80 600 144 108 91 1940 2825 297 742.5
Permanent Land requirement m2 8 58 14 10 9 187 272 29 71

When operated for a hypothetical 2,000 hours/year, the incremental power, fuel and
demineralized water costs are tabulated below.

Table 3-5: GT3-Operating Cost Comparison, 2,000 hours/yr.


Assumptions
Fuel Cost [US$/gal] $ 2.75
Fuel Oil Heating Value [BTU/gal HHV] 132,000
LHV/HHV ratio 0.96
Annual hours of operation [hours/year] 2,000

1% 2%
Evap Chilling, Chilling, Chilling, water/wet water/wet Dry Air Humid Air
Parameter Units Cooling Fogging to 13 C to 15 C to 17 C compression compression Injection Injection
Electric Energy Generated [kWh] 1,126,600 1,526,000 5,386,800 4,992,000 3,722,000 10,845,800 18,818,200 9,300,000 9,540,000

Fuel Cost [US$] $ 196,224 $ 257,769 $ 1,300,608 $ 1,099,089 $ 879,948 $ 1,690,668 $ 3,018,446 $ 1,763,921 $ 2,064,569
Demin water Cost [US$] $ - $ 57,031 $ - $ - $ - $ 620,899 $ 1,184,609 $ - $ 294,267
Operating Cost [US$] $ 196,224 $ 314,800 $ 1,300,608 $ 1,099,089 $ 879,948 $ 2,311,567 $ 4,203,056 $ 1,763,921 $ 2,358,836

average power cost [$/kW] $ 0.174 $ 0.206 $ 0.241 $ 0.220 $ 0.236 $ 0.213 $ 0.223 $ 0.190 $ 0.247

3.5 Unit 4, Westinghouse 251-B11

The apparent degradation of Unit 4 based on the last performance test of 35.4 MW
(19JAN2014) is 13.2%. The ambient test conditions indicated 30C. This degradation is high
than expected and the cause should be investigated.

Report No. 1246-001-01 Page 3-6


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

Table 3-6 GT4 Performance Enhancement Summary


1% 2%
water/wet water/wet
Chilling, Chilling, Chilling, compressio compressio Dry Air Humid Air
No Parameter Units Evap Cooling Fogging to 13 C to 15 C to 17 C n n Injection Injection
1 Cost New k.USD 33,878 33,760 36,698 36,326 35,986 34,367 34,699 35,719 33,385
Demin Plant k.USD 50 100 441 660 n/a 250
Compressor component recoating k.USD 750 750 n/a n/a
2 Cost Base Plant k.USD 33,347 33,347 33,347 33,347 33,347 33,347 33,347 33,347 33,347
3 Technology Estimated Cost, USD k.USD 639 564 3,686 3,277 2,903 2,432 3,038 2,372 2,835
(includes 10% for retrofit vs. new premium)

4 New Power kW 36,722 36,977 39,825 39,304 38,758 41,154 44,739 40,413 40,527
5 Base Power kW 36,249 36,249 36,249 36,249 36,249 36,249 36,249 36,249 36,249
6 New Parasitic kW 10 13 1,358 1,130 934 53 92 20 40
7 Additional Net Power Output kW 463 715 2,218 1,925 1,575 4,852 8,399 4,144 4,238

8 Capacity cost USD/kWh $ 1,380 $ 789 $ 1,662 $ 1,702 $ 1,843 $ 501 $ 362 $ 573 $ 669

9 Lead time, month months 12 6 12 12 12 16 16 6-8 10-12


10 Installation Time, days weeks 6-8 4-6 5-7 5-7 5-7 12-16 12-16 6-8 10-12
11 Down time, days days 12-16 6 15-20 15-20 15-20 14-28 14-28 2-3 2-3

New GT Fuel Consump. [LHV] kBTUY/hr 417,773 419,410 439,202 435,575 431,636 447,981 474,658 424,825 425,427
Non-GT Fuel consump [LHV] kBTUY/hr - 24,599 30,755
Base Fuel Consump. [LHV] kBTUY/hr 414,077 414,077 414,077 414,077 414,077 414,077 414,077 414,077 414,077
12 Incremental Fuel Cons, Btu/hr [LHV] kBTUY/hr 3,696 5,333 25,125 21,498 17,559 33,904 60,581 35,347 42,105
13 Incremental heat rate [LHV] BTU/kWh 7,983 7,460 11,327 11,167 11,148 6,988 7,213 8,530 9,935
Specific Fuel Consumption
[asssumes 19,500 btu/lb] litre/kWh 0.221 0.207 0.314 0.310 0.309 0.194 0.200 0.237 0.275
14 Additional O/M cost, USD/kWh US$/kWh - 0.040 - - - 0.064 0.071 - 0.035
Demin Water Cost (Rp 576.07/liter)
15 Water needs lbs/sec 0.349 0.3624 7.083 5.983 4.856 3.953 7.544 0 1.873
16 Cooling Capacity AC Tons n/a n/a 1796 1524 1235 n/a n/a n/a n/a
17 Permanent Land requirement ft2 80 600 144 108 91 1940 2825 297 742.5
Permanent Land requirement m2 8 58 14 10 9 187 272 29 71

When operated for a hypothetical 2,000 hours/year, the incremental power, fuel and
demineralized water costs are tabulated below.

Table 3-7: GT4 Operating Cost Comparison


Assumptions
Fuel Cost [US$/gal] $ 2.75
Fuel Oil Heating Value [BTU/gal HHV] 132,000
LHV/HHV ratio 0.96
Annual hours of operation [hours/year] 2,000
1% 2%
water/wet water/wet
Chilling, Chilling, Chilling, compressio compressio Dry Air Humid Air
Parameter Units Evap Cooling Fogging to 13 C to 15 C to 17 C n n Injection Injection
Electric Energy Generated [kWh] 926,000 1,429,800 4,436,200 3,850,200 3,150,200 9,704,000 16,797,000 8,288,000 8,476,000

Fuel Cost [US$] $ 160,417 $ 231,467 $ 1,090,495 $ 933,073 $ 762,109 $ 1,471,528 $ 2,629,384 $ 1,534,146 $1,827,459
Demin water Cost [US$] $ - $ 56,937 $ - $ - $ - $ 621,056 $ 1,185,238 $ - $ 294,267
Operating Cost [US$] $ 160,417 $ 288,404 $ 1,090,495 $ 933,073 $ 762,109 $ 2,092,584 $ 3,814,622 $ 1,534,146 $2,121,726

average power cost [$/kW] $ 0.173 $ 0.202 $ 0.246 $ 0.242 $ 0.242 $ 0.216 $ 0.227 $ 0.185 $ 0.250

Report No. 1246-001-01 Page 3-7


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

4.0 CONCLUSIONS

From the discussions of the various gas turbine performance enhancements options,
certain systems can be dismissed from consideration due to the humid climate of
Indonesia, namely evaporative cooling and fogging systems. Of the remaining systems,
the inlet cooling system is by far the most expensive and can also be dismissed due to the
infrequent hours of operation of the gas turbines, the significant downtime to install the
chilling coils, and large capital investment for these turbines with few remaining plant life
years.

The wet compression alternative requires gas turbine specific installation of compressor
coatings, nozzles, and demin water usage. Again, because the gas turbines are relatively
old, the additional cost is significant to invest in these units. The most significant cost of
wet compression, application of protective coatings, is not portable to other gas turbines
and must be applied to each individual gas turbine.

This leaves DAI (Dry Air Injection) and HAI (Humid Air Injection). Currently there is no
steam source of approximately 10 bar pressure at the Pesanggaran facility to utilize HAI.
Therefore the cost to install a heat recovery device to use the steam intermittently is not
profitable. While other gas turbine sites may have steam available, if steam is being
removed from a steam turbine it is merely displacing the power created in the steam turbine
with power created in the gas turbine. The steam chemistry would also need to be
reviewed carefully to minimize the effects on the gas turbine components.

It is apparent that the most cost effective, ($/kW) and least effect on O&M is dry air
injection. The dry air injection

O Has the least down time


O Has one of the lowest capital cost per kW
O introduces no additional contaminants into the gas turbine
O Has the portability to relocate the system on any other or future gas turbine(s)
that might be installed at the site

In the GTPro model only 5% incremental air flow was evaluated. Due to the vintage of the
gas turbines at Pesanggaran, the stall margins are much more conservative, and therefore
it is highly likely additional performance can be achieved by exceeding the 5% flow
threshold.

Report No. Page A - 1


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

5.0 RECOMMENDATIONS

For each of the 4 gas turbines evaluated, the dry air injection presents the least cost alternative
to performance enhancements of gas turbines install at the Pesanggaran Station in Bali. The
dry air injection

o Has the least down time

o Has one of the lowest capital cost per kW

o introduces no additional contaminants into the gas turbine

o Has the portability to relocate the system on any other, or future gas turbine(s)
that might be installed at the site or in the IP system

o Requires no water use

o Can be brought on line extremely quickly

o Can be manifolded to serve multiple gas turbines achieving economies of


scale, and can easily be fitted with a system redundancy with an extra module
that can serve any of the four gas turbines.

Other sites in Indonesia may have other parameters which may alter the DAI recommendation.

Report No. Page A - 1


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

6.0 BIBLIOGRAPHY
Al-Sati, Z. (2014). Wet Compression. Dubai: Siemens AG.
Crampsie, S. (March-April 2012). Wet Compression Boost for Power Output and Efficiency. Gas
Turbine World.
Frank J. Brooks, G. E. (2000). GE Gas Turbine Performance Characteristics, GER 3567H.
Schenectedy, NY: General Electric.
Power, I. (2012). Statistic Report, 2012. Indonesia Power.
Rose, N. (August 2010). Indonesian Energy Report. Australia: Norton Rose Group.
Turbine Air Systems. (n.d.). Retrieved January 2015, from Turbine Air Systems:
http://www.tas.com/energy-efficiency/turbine-air-systems/turbine-inlet-chilling-overview.html

Report No. Page A - 1


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

ATTACHMENT A-1

Unit 1 Performance Figures

Report No. Page A - 1


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

ATTACHMENT A-2

Unit 2 Performance Figures

Report No. Page A - 2


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

ATTACHMENT A-3

Unit 3 Performance Figures

Report No. Page A - 3


RRT SIGMA Engineering Evaluation of Gas Turbine
Performance Improvement Alternatives
for Indonesia Power

ATTACHMENT A-4

Unit 4 Performance Figures

Report No. Page A - 4

You might also like