It is GE 9E DLN1.0 gas turbine. I am a question when i am reading the CSP. I hope someone could help me.

I find FSRP is not directly assigned to as fsrout. Instead of that, there are several steps of modification before it is output to Servo Card.
Following is the main line:

FSRP==FFRG1X==FFRG1
FSRP: FUEL STROKE REFERENCE PRIMARY
FFRG1X: FGCV #1 Heat Rate Flow Reference
FFRG1: FGCV #1 Heat Rate Flow Reference

FSRP equals FFRG1X and FFPG1,but the signal name changed from stroke reference to Heat Rate Flow Reference. That is wired.

FQRG1_V1= FQRG1/ FQKCNST1/ FSRG1FP/ FPG2AF

FQRG1: FGCV #1 Flow Reference (#/sec) + Prefill;
FSRG1FP: Fuel Gas Valve G1 Piping geometry factor;
FQKCNST1: FLOW CONTROL EQUATION CONSTANT #1;
FPG2AF :FUEL GAS INTERVALVE PRESSURE FILTERED

To be continued (if is asked, I am willing to finish this)

So these is some modification? and it is very complicated?

My question is why these modifications is considered, as traditional combustor doesn't?
Why so much change?

Best regards!
Neo [email protected]

 

Neo,

This is a great question--I just wish I had an answer as good.

This is done for DLN units which use the IGCV (Independent Gas Control Valve) skid. Because all of the gas control valves are connected to (supplied by) a common header/manifold it was felt it was necessary to do a kind of "sanity check", as it were, on the individual valve position references. And, so the "simple gas law/equation" method was chosen (I think that's what it's called) to do so.

As you mentioned, this isn't done for units with conventional combustors. Well, it sort of is done--but in a very round-about way. The GCV valve internals are known to the control system requisition engineer, and those values were fed into the "GAS program" and knowing the upstream pressure (from the SRV configuration) they would calculate the expected values of GCV position which corresponded to the flow-rates required for light-off (firing), FSNL, Base Load, and Peak- and Peak Reserve Load, if so equipped. The result of those calculations are shown in the Control Specification Drawing, Sect. 05.02.nn (05.02 is the Gas Fuel section of the Fuel portion of the Control Spec; 'nn' stands for the number of the section titled "Expected Fuel Flow Characteristics" (or something similar)). There should be several lines in a table for common conditions (firing; warm-up; FSNL 1/4 Rated Load; 2/4 Rated Load; 3/4 Rated Load; 4/4 Rated Load; etc., if so equipped).

This table in this section of the CSPEC (as it's affectionately known) is based on the expected fuel characteristics supplied with the order for the turbine (or control system if the combustion system is being upgraded) AND presumes the LVDTs for the fuel control valves are calibrated properly AND the P2 pressure is as specified. It's pretty rare that on any given day that actual fuel being burned exactly matches the expected fuel characteristics supplied when the turbine/control system was being built, AND it's not very likely that LVDTs are calibrated properly (remember: there's no written procedure for calibrating LVDTs...!). And, the longer it's been since the unit was commissioned, the more likely it is that the actual fuel characteristics are close to the supplied expected fuel characteristics (unless the turbines are located very close to the well-head(s)). <b><i>However,</b></i> the values should be relatively close to actual values experienced when the turbine is operating. This one section is one of the most important--and useful--in the CSPEC for troubleshooting and understanding expected fuel flows and valve positions. It is a "calculated" table of values, based on some presumptions, but it is a very good reference.

(I've been able to use it to point very clearly to fuel supply problems in several cases when the site was having problems lighting-off and making rated load--because the fuel being burned was very different from the expected fuel characteristics. It's also been very useful in pointing out some very serious problems with LVDT calibrations which had resulted in some pretty wild changes to various Control Constants which were causing some "mysterious" problems--which, of course, were ALL the fault of the Mark V (or the Mark VI), which, of course, should be capable of preventing people from doing not-so-intelligent things, which is basically what I was told by the Plant Managers (true story)).

Anyway, back to the original question, there should still be a Section 05.02.nn, 'Expected Fuel Flow Characteristics,' in every CSPEC. And, the CSPEC is the place I'm going to refer you to to get more information about these "modifications" to the various GCV position references. That's where I've found information--albeit it brief and nebulous--related to the DLN gas control valve position reference. If you really want to dig into it, you're going to need to do some World Wide Web-searching about the "simple gas law/equation" (again, I think that's what it's called) to piece together a better and more comprehensive understanding of these "modifications. I've done this many years ago, but I didn't keep my notes. And, in any case, I've never found that it didn't work. (It may not have worked as well as it could have, but that's one of the beauties of the Speedtronic turbine control system--it has some "flexibility" in that configuration and programming and parameters and calibrations don't have to be spot-on all the time. And, it's one of the maddening things for people trying to learn the Speedtroinic--because everyone expects digital control systems to always be 100% spot-on and perfect.)

Anyway, that's about as much as I can offer. I have accepted that this scheme works well enough to meet the requirements of the process--at least, it's never given me any trouble that required a deep-dive into the guts of the control scheme. So, other than what I've written, there's not too much more I can offer. There are those in GE, mostly younger more theoretical types, who have some very serious strong objections to the way things have always been done in Speedtronic control systems and philosophies, and they're driven to put more science and discipline into the control schemes without understanding the nuances and complexities--and realities--of real-world operations and maintenance. One has to remember that "back in the day" when gas turbine control systems were being developed many field devices and instruments (I'm referring to transmitters in particular) were extremely expensive both to purchase, to maintain, to replace and to integrate into control systems. They required power supplies and signal conditioning and all of these things were expensive and costly--and in some cases, not as reliable as required. Always remember that reliability is king in GE heavy duty gas turbine control philosophy. Today, such devices are "cheap" in relative terms, and more easily integrated into control systems without too much expense or without detracting from reliability. So, you're going to continue to see more and more sensors of various types used to "enhance" heavy duty gas turbine control and reliability. (My fear is that this OEM is or will approach or outdo the Germans in their use of instrumentation and control system complexity--which is seriously "over the top" and approaching the limit of reliability.) So, as things get less expensive and people determine (rightly or wrongly) that more instrumentation is better things are going to get more complicated and involved. Unfortunately, this OEM isn't one to document their intent or philosophy or control schemes or algorithms--and is even becoming more protective of their "intellectual property" as it pertains to controls and operation. This is the work we live in.

Hope this helps!

 

CSA:

Thank you so much for your elaborate reply, and it helps as it used to be.

After reading your reply several times, it comes to me that it is done the same in steam turbine control.

The steam mass flow rate is first calculated, and then it is converted to valve position command based on valve mass flow characteristic curve. In the situation of gas turbine, the valve mass flow characteristic curve is replaced by complicated equations. Is it right?<pre>
This is the mass flow rate equation:
FQRG1=(FFRG1/100)*(FJKGHR/FHKG_LHV)*FQKCNST4+FQRG1PFC
FJKGHR:Fuel Gas Heat Rate Gain (MBTU/HR)</pre>

CSA,can you give some explanation on FJKGHR. I have no idea about that. What is the physical meaning of FJKGHR?
And i find it is also used in other places,for example:<pre>
FSKSU2_FI=FJKSU2_FI/<b>FJKGHR</b>*100</pre>
That is also different from which i know. In the CSP i read before,FSDSU2 is one of the control constants, not calculated by the above equation.

I don't know why this difference is made. By the way i doesn't find <b>Expected Fuel FlowCharacteristics</b> in control specification.

Looking forward to your reply.
Best regards.
Neo [email protected]

 

Neo,

What is the value of <b>FJKGHR</b>? I suspect it's a ratio, with a value near 1.000--but that's just a SWAG (Scientific Wild-Arsed Guess).

What is the longname description for FJKGHR in Toolbox?

What is the longname description for FJKGHR in the CSPEC?

You're going to see more instances of "modifications" like this in the future. There is a disruption going on in the GE heavy duty gas turbine control philosophy realm, and a lot of it is not going to be documented very well.

 

Anytime a small adjustment or "tweak" is needed. Always "tweak" on a percentage basis. 1.0000.... 0.9999.....1.0001

 

I have seen this "tweak" utilized on heat rate calculation on many occasions.

On coal plants; adjust the moisture content of the coal to fit the recent model.