Evaluating HRSG Field Data    V.Ganapathy

Gas Turbine HRSGs like other waste heat boilers are designed for a particular set of conditions but often operate at slightly or even significantly different gas/steam parameters due to various reasons. The steam pressure could be different due to plant header conditions;feed water temperature may be 5 to 15 F off due to deaerator operation;more steam could be taken off the HRSG for deaeration; exhaust gas flow ,temperature and analysis could be different due to changes in ambient conditions and gas turbine load and so on..So how does then one reconcile the actual operating data with the HRSG design data or guarantees?
HRSG simulation or the HRSGS program could help in determining if the HRSG is performing as designed even under different gas/steam parameters.This will be explained using the case of a simple single pressure HRSG.
Example:A gas turbine HRSG is designed for the parameters shown in Table 1. When put into operation,the field data was slightly different as shown. The steam temperature is uncontrolled.
Gas analysis used :% vol co2=3,h2o=7,n2=75,o2=15.use a heat loss=1% in both cases.The plant engineer wants to know if the HRSG is doing fine or not and whether the lower steam generation is solely due to the lower gas flow/inlet temperature conditions and steam pressure.

Table of design and operating data for HRSG 
 condition Design/Guarantee Operation/test
gas flow,lb/h 550,000 ?
exhaust gas temperature,F 1000 970
exit gas temperature,F 371 380
steam pressure,psig 600 500
steam temperature,F 700 690
feed water temperature,F 230 230
blow down,% 2 0
steam flow,lb/h 79,400 68,700
The above situation is not uncommon in cogeneration plants.Let us use HRSGS,the HRSG simulation program, to see how the HRSG is doing. (Visit the HRSG Simulation web site for a Free Demo and information on HRSG simulation)
The HRSGS program is used to firm up or establish the HRSG 'design'. Using the gas parameters given in the design case ,the pinch and approach points are varied till we get the exit gas temperature close to the design value;the design pinch and approach points are then frozen.The design is also indirectly frozen or established.If the water temperature at the exit of the economizer is known,the approach point used can also be verified.The output from the HRSGS program is shown below.
(The temperatures are in F,gas/steam flows in lb/h and the duty in MM Btu/h and steam pressure in psia.)
  Design Case: Establishing the HRSG Design basis: gas flow=550,000 lb/h
surface gas in F gas out stm in stm out duty press flow % stm pinch apprch
suph 1000 921 492 700 11.71 615 79440 100
evap 921 517 472 492 59.36 635 79440 100 25 20
econ 517 371 230 472 20.77 645 81029
In order to evaluate the off-design or operating condition,the gas flow is first determined from the operating data using heat balance methods. Gas flows are difficult to measure,while gas temperatures,steam/water flows can be measured more accurately.The energy absorbed by steam is first computed and using the inlet and exit gas temperatures and the gas analysis,the gas specific heat or enthalpy is computed from which the gas flow is obtained. In this case,the gas flow can be shown to be 501,300 lb/h.With this gas flow,the HRSGS program is again run in the off-design mode  and the results are shown below.Note that the steam pressure is different in the off-design case.

Off-Design performance:Verifying the operating data: gas flow=501,300 lb/h
surface gas in F gas out stm in stm out duty press flow % stm pinch apprch
suph 970 894 473 691 10.26 515 69524 100
evap 894 495 461 473 52.97 533 69524 100 22 12
econ 495 364 230 461 16.93 543 69524
In order to arrive at some conclusion,the HRSG supplier and plant owner should agree to some tolerances in instrument readings,errors in measurements,procedures used for gathering data etc.The ASME PTC 4.4 gives some information on these issues.
From the off-design run,It can be seen that the HRSG should be generating 69,524 lb/h and that the gas exit temperature should be 364 F,while that measured is higher and the steam flow is also lower.Hence there is reason to investigate further as even after correcting for the different gas inlet temperature,flow and steam pressure,the actual performance is off from that suggested by the simulation process.It is possible that the HRSG is not adequetely sized or that the steam flow readings,gas temperatures have to be checked again.In any case,with the HRSGS program,plant engineers have a tool for evaluating HRSG performance under different parameters and operating conditions.One cannot simply say that due to differences in steam pressure or gas flow or temperatures,the HRSG performance is different from that guaranteed and walk away.HRSGS program also helps the plant engineer to perform what-if analysis. Using one set of operating data,the performance at other conditions may be obtained.Say in the future,the feed water temperature or steam pressure is likely to be different or if the gas turbine is going to be replaced,the HRSG performance can be easily evaluated using HRSGS program.One can also see what happens if supplementary firing were to be done.What is most important is that he can perform these studies himself without contacting the HRSG supplier or use the HRSG supplied information to verify his calculations and thus become a knowledgeable customer! Note that the simulation process is explained at my web site above and also in my books. One of the major features,which makes this a simple process is that we don't have to know the tube/fin configuration,geometry,tube lengths,spacings etc to perform the above analysis! Hence anyone familiar with heat balances can do it! [Note that we could have also established the design using the operating data and then checked the off-design performance using the guarantee gas inlet conditions and checked for the steam generation].
Note that this is a simple  HRSG configuration.Similar analysis could be performed for complex,multi-pressure, fired HRSGs. The HRSG supplier and Plant engineers have to work together and agree to several aspects discussed above before reaching any drastic conclusions as there are lot of variables involved.

 Books,software,papers on Boilers,HRSGs,Steam Plant calculations
 email Ganapathy