In our power plant, GE gas turbines 9FA are liquid fuel operated. For all units, the water injection is in service @ 150 MW but only one unit the water injection is in service @ 75 - 80 MW. That difference happened after the performance test for thos unit. We compared between the operating parameters @ 150 MW and 75 MW for this unit and another unit. TTXM, IGV position, IBH, SRF1, water injection demand all this parameters are constant for the same operating conditions and loads. Both of water injection combustion reference constant and fuel flow sufficient constant are the same. No alarms announced during water injection enabling.

Any ideas about this situation?

 

Mahmoud Alshimy,

Water injection is usually scheduled (calculated) as a function of fuel flow-rate. It starts when the fuel flow-rate gets above the "enabling" fuel flow-rate, and varies as a function of fuel flow-rate.

The amount of water required to meet a certain NOx output (in ppm) can be related to the fuel flow-rate. Most regulatory agencies accept the practice of calculating the water injection flow-rate required to meet a particular NOx emissions guarantee on fuel flow-rate. Most emissions monitoring systems are relatively slow to provide an output (they have to take a sample from the stack, draw the sample into the analyzer, analyze the stack gas sample, then provide a measurement output, then purge the sampling line and analyzer before drawing another sample to begin the analysis process again. So, using the emissions monitor output to control water injection flow-rate is not very responsive to changing operating conditions. Some newer emissions monitors are faster, but still not "real time."

Most (but not all) sites with water- (or steam-) injection for NOx reduction also have emissions monitors. So, if your site is equipped with emissions monitors you can tell if the water injection flow-rate is properly controlling the NOx emissions for the turbines.

If your site doesn't use fuel flow-rate for calculating (scheduling) the water injection flow-rate but uses the output from an emissions monitor to control (including starting and stopping) water injection flow-rate then it's very likely that there is some problem with the emissions monitor output.

I have also seen a couple of GE-design heavy duty gas turbines which uses emissions monitor output to bias water injection flow-rates so as to keep the water consumption to a minimum. If your site uses a scheme like this then it's very possible that there is something amiss with the emissions monitor output to the Speedtronic turbine control system.

Without being able to see the sequencing/application code running in the Speedtronic turbine control panels at your site and without lots more information about the equipment and configuration in use at your site it's very difficult to make any other concise comments than to say that something is definitely amiss because, while the two turbines will NOT always inject the same amount of water to achieve the same emissions, they should be relatively close--much closer than you are describing. Most water injection systems start (enable) water injection at about 30% of rated load (sometimes a little more; sometimes a little less). So, if one turbine isn't enabling water injection until 150 MW something is definitely amiss.

Please write back with more information or to tell us how you resolve the problem.

 

CSA,

Let me first tell you that I have a mechanical background not I&C. But let me redraft because I think you didn't get my point of view. Here we are not talking about "the amount of water required for NOx control" but talking about when NOx control be in service about "Water Injection enabling load" for the same frame, site , control and -I guess- same configuration. We reviewed the control logic and we found out that liquid fuel flow and combustion reference temperature _CRT_ should be higher that a certain value to enable water injection. comparing to the other unit the differences in constants values are not enough to make this big difference in water injection enabling loads. Also we compared TTXM, IGV position, IBH, SRF1, water injection demand all this parameters are the same for this unit and the others for the same load and operating conditions. We already emissions monitoring systems but it is installed and commissioned by HRSG contractor not by GE.

 

Mahmoud Alshimy,

You are 110% correct. I don't understand your query.

If water injection is being used for emissions reduction on the units at your site then it's likely that the emissions on the unit that starts (enables) water injection at 150 MW are higher than they should be at loads below 150 MW. That's why I recommended checking the emissions monitors of the two units (at similar loads, say, 50, 75, 100, 125, 150, 175 and 200 MW); that would tell you if the proper amount of water was being injected for the two units if the emissions were not the same at the same loads.

In some parts of the world, if water injection for NOx emissions reduction is not running at the required flow-rate to limit NOx emissions the operators and their supervisors can be arrested, prosecuted and imprisoned. Water injection is usually started at about 30-50% load, and is increased as fuel flow-rate increases in order to limit NOx emissions to the required value at all loads above the point at which water injection is started.

Water injection isn't just something that is optional for most turbines--it's the law (for emissions control). If it's not at your site, why would you buy the water, treat the water (it basically has to be boiler-quality water--demineralized at a minimum) to inject it and lose it out the exhaust stack where it can't be recuperated and reused?

I still suggest that something is amiss if one unit starts water injection at 50 MW and the other one starts it at 150 MW (or whatever the exact numbers are). The actual emissions of the two units at similar loads would confirm that. Exactly what's wrong--well, without being able to see the logic/sequencing/application code running in the turbine control panels at your site it's almost impossible to say based on the information provided. You haven't told us what the liquid fuel flow-rate is for the two units at identical loads, or what the combustion reference temperatures are for the two units at the same identical loads.

Give us actionable data, and we can probably help you.

But, you're right--I do NOT understand your query.

Best of luck in solving your issue or getting your question answered. Any engineer (mechanical or I&C) should be able to logically think through a problem, understand what's supposed to happen, and then be able to get to the cause of why what's supposed to happen isn't happening--and from there, what to do about it. Engineering is nothing more than a logical thought process--regardless of the discipline (mechanical or I&C).

Please write back to tell us how the "problem" is solved (because, based on the information provided, there is a problem).

While it's very uncommon, water injection has been used for power augmentation (increasing power output). But it's not usually started (enabled) until the unit is already at Base Load. And, again, the cost of water and water treatment are usually more than the revenue generated by the increased output--so it's not used very often. It's entirely possible you are referring to water injection for power augmentation. Again, it's not clear from the information provided--which may be why I'm unable to understand your query.

Again, best of luck! It would be nice to hear how the issue is resolved.