GE Primary Frequency Regulation (PFR)

J

Thread Starter

JM

What is/are the limitation of the GE PFR? Presently, our GT baseload is around 235 MW with these kind of ambient temp and humidity. Example my GT at 200 MW PSL and a big power station tripped which cause the grid frequency to fall from 50 hz to 48 hz (just to exaggerate things), does my GT will pickup up to baseload only (temp control) am I right? TTRXB/TTRX are 656.6oC and 648.9oC respectively for maximum limit (Isothermal). TTRXB/TTRX is 610oC during baseload.

Why there is no more Peak load selector in our plant? In my previous experience in a 32 MW GE gas turbine Mark IV, we have PSL, Baseload and Peak load selector. During baseload, exhaust temp was 560oC and if I remembered it right at peak load, exhaust temp setpoint was 580oC. (12 years ago).

Another situation, If my GT is at 70 MW PSL, and suddenly too many feeders tripped (again let's exaggerate things), grid frequency will increase too much. Does my GT will trip on reverse power cause it will try to regulate the frequency? Or it will limit only to minimun load of 25 MW? 25 MW is the minimum load (Spinning reserve as per GE definition) during start up. I can see in RUN status: Spin_Res.

What is the meaning of Secondary Frequency Response (Regulation)? Does all GE gas turbine have this kind of control?

Thanks.
 
Primary Frequency Regulation just says that if you are operating the unit on Pre-Selected Load Control with Primary Frequency Response enabled that the unit will respond to frequency excursions by adjusting load in response to the frequency change in order to try to help maintain grid frequency. And, as always the unit can never load past CPR-biased exhaust temperature control limit, whether it's responding to frequency excursions or operator commands.

Peak Load (and Peak Reserve Load) is(are) a purchased option(s) for most machines; you want it, you got to pay for it. Many, but not all, F-class units are not capable of Peak Load; they are already running at the limit of the capability and materials of the machine.

If your unit is operating at Part Load (NOT Pre-Selected Load Control) and the frequency tries to increase, the Speedtronic will decrease the output to try to maintain frequency. But, if it's not the Isochronous machine, it won't control the frequency very well without some manual operator intervention, especially if it's the largest unit remaining on the grid.

If your unit is operating on Pre-Selected Load Control and Primary Frequency Response is *not* enabled, then the output will remain at 70 MW because the operating reference is a load setpoint and even if the frequency changes the Speedtronic will adjust the output to maintain 70 MW, so it will make the situation worse.

If your unit is operating on Pre-Selected Load Control and Primary Frequency Response *is* enabled, then the output will change as the frequency changes to try to maintain a stable frequency. BUT, if it's not the Isochronous unit and it is the largest generator remaining on the grid, then it won't do a very good job of frequency control without some manual operator intervention.

If the load decreases sufficiently to drive the unit to negative real power, the generator breaker will open on Reverse Power. Reverse Power is not generally a turbine trip, just a breaker-opening event. But, it really depends on the stability of the grid, how big the other units on the grid at the time are and if any of the other units are in Isochronous control. If the grid is down to one generator and it's operating in Droop speed control, it's not going to control frequency very well without some manual operator intervention or unless it is switched to Isochronous Speed Control.

Spinning Reserve is usually an adjustable setpoint, and it is usually only displayed immediately after a successful automatic synchronization during a unit start sequence. Unless the unit has some very unique sequencing or unique displays or a display problem, the only time Spinning Reserve will be displayed is after a successful automatic synchronization during a unit start-up.

Secondary Frequency Regulation, as generally interpreted, means some kind of "remote" frequency control being used to adjust a generator's output to control grid frequency. That's usually done with a hardwired analog (4-20 mA) signal or RAISE/LOWER contacts to drive the turbine speed reference, and usually some regulatory body or supervisory agency uses one of these methods to control a unit's output remotely.

No; most GE machines don't have this. It's usually a purchased option, and it's usually called External Load Control or Remote Load Control or AGC (Automatic Governor Control) Control.
 
Thanks CSA.

So, if my GT is operating on 70 MW PSL and frequency really goes high, it will trip on reverse power. There will be no minimum load (or deadband) in operating the unit.

I am thinking that it will just hold to minimum load (lower load limit) of the GT which is 25 MW load and if the frequency really goes high, GT 52G will trip by overfrequency 81O (G60 = 51.5 Hz, ABB REG 316 = pickup1 51.5 Hz and pickup2 52 Hz)as per tripping matrix and unit will stay on FSNL.

If the frequency goes low, the upper limit is CPR biased exhuast temperature limit or they called load limiter to the other control system. Just like in Base load operation, even PFR is ON, GT will not response anymore and in reality it will reduced the actual load due to reduction in speed which cause reduction of airflow, Am I right?

After annual Combustion Inspection outage, I&C will do tuning, so this PFR will always selected to OFF during tuning to prevent any load variation during frequency disturbance (deviation).

What is the difference between REVERSE POWER during frequency high and REVERSE POWER during normal shutdown? During normal shutdown, 52G will open approx 5 MW. You told (CSA) that 52G will open on reverse power during negative real power.

Can anyone explain to me how to calculate the exact MW load the generator 52G will trip during reverse power. The following data as follows: G60 reverse power (32) PT ratio: 1570/110, CT ratio 15000/1, settings stage 1 min 0.043 pu, delay 10 s (blocks while LCI in service). ABB REG316 (32) PT ratio: 1570/110, CT ratio 15000/1, P settings: -0.07, angle 30, drop ratio 60, delay 10s, PN 0.8.

So, secondary frequency regulation is AGC or Remote Load Control. We have this kind of regulation, it is control by our Toshiba DCS. We already tested this AGC, the Load Dispatch Center will give the load command for our GT GE 9FA Mark VI with some limitation. This AGC only operates during testing (commissioning) and never use again. We still do the basic load change by calling on the phone by the Load Dispatch Center.

Thanks.
 
Many Speedtronic turbine control panels will trip the generator breaker on reverse power before the 32 relay; the 32 relay is used as a back-up to the Mark V or Mark VI. *In general* and unless otherwise directed or requested, the Speedtronic is programmed to open the breaker on negative MW, but on occasion some sites require or request to set the "reverse power" trip to a slight positive value, might even be 5 MW. You should look at your sequencing/application code to be sure.

I have never heard of a "lower load limit", but that doesn't mean one can't be programmed into the unit. You should look at your sequencing/application code to be sure.

The exact, precise details of how your unit operates are detailed in the sequencing or application code which is running in the Speedtronic of your unit. No written description or forum answer can over-ride what's programmed into the Speedtronic panel. I was providing general information; apparently your site is different. You should look at your sequencing/application code to be sure.

If you have some tripping matrix which outlines how your unit operates, then you should have the answer to your questions.

PFR will only adjust load when the unit was operating at a Pre-Selected Load less than Base Load. If it increases power output to CPR-biased exhaust temperature control (Base Load) because of a decrease in frequency, then the reduced power airflow will cause a reduction in power at Base Load.
 
Thanks CSA, I just checked the Mark VI manual today and it states that:

REVERSE POWER SEQUENCING

The generator breaker is tripped on reverse power sensing, in the standard shutdown sequence. Reverse power rather than drop out of 4's alone is necessary because of the risk of coincidental fuel system malfunction that could cause ovespeed. Breaker opening is allowed only with 4 trip or shutdown 94x. The breaker is not allowed to open with other possible control malfunctions which could lead to overspeed.

Time delay on the reverse power relay is provided to allow for load transients while synchronizing. Turbine trips cause motoring of the generator for the time delay on reverse power sensing.

On new units, a megawatt setting is compared with the DWATT signal in place of the reverse power relay. This allows smooth shutdown with minimun draw on system megawatts.

Now, it is clear to me that during shutdown sequence, there is a MW setting to open the 52G which is 5 MW load. Before, my understanding the 32 relay is giving the command to open the 52G.

Today, we are on the process in testing the AGC control. All load changes was done by LDC (Load Dispatch Center). GT load High/Low Limit are available. High limit (224.2 MW) was based on the actual ambient and humidity while Low limit (140 MW) was selected to prevent from removing from premix mode.

I trend the system frequency during our GT#2 tripping, frequency fell from 50 Hz to 49.87 Hz. GT#3 pick up from 200 MW to 210 MW only. GT#1 on annual outage. At 4% droop setting, GT#3 should pick up around 15.275 MW (based on 235 MW full load calculation) but it pick up 10 MW only. LDC asked me why GT#3 only pick up 10 MW? Frequency still not reached 50 Hz. Is there any deadband/limit in Mark VI to prevent to increase to CPR-biased exhaust temperature control limit? I hope somebody can answer this question.

Before, I also work in 647 MW Coal Fired Thermal Plant with Alspa P320 DCS control but later it migrates to Emerson Ovation. During Alspa P320, Frequency compensation is applied in the load control loop, when there is a change in grid frequency the Auto Load control set point is adjusted by application of the frequency compensation term KDeltaF MW (up to a maximum of ±50 MW) with a droop setting of 5%. The Upper Level Load Controls and the Base Level Regulator both support the free governing action.

For very large frequency fluctuations, if the Upper Level Load Control logic detects a large error (>=100 MW) in the Auto Load Control(ALC) loop, it will switch from Auto Load Control into Direct Load Control(DLC). The actual MW on the Unit will depend on the actual Load set point at the time of engaging Direct Load Control, and the application of KdeltaF in the Base Level Regulator.

I given this information (maximum of ±50 MW load frequency compensation in Alspa P320), just in case there is a similar limitation from Mark VI to increase to a certain load and prevent to reach the CPR-biased exhaust temperature control limit. I also thinking that there might be a deadband/limit in reducing the GT load during high frequency.





 
Any information you find in a GE manual is generic and should be considered as a guide for how the Speedtronic might be programmed. Consult your application code in the .m6b file running in the Mark VI at your site for the exact details of how the Mark VI is programmed and will operate at your site.

We don't have any details about how the units were connected to the grid, what other units were operating at the time; nothing. For Droop speed control to increase or decrease load it is necessary for there to be a difference in frequency, and we don't know what load, if any, other units might have picked up during this testing. Many times, these "tests" are deemed to be failures because they are improperly conceived and executed. If there were other turbines operating at the time in parallel with your units, and they were being operated in Droop mode, then it's likely they picked up some load as well.

We can't comment on how any other control system was programmed in relation to how the Speedtronic control system was programmed. Please consult the OES for these units for any details and assistance with the results of your testing. Please don't assume that every turbine control system will respond similarly to load or frequency disturbances. Please take all factors of operation into account when making any decisions about the success or failure of any testing. Please remember: Droop speed control is straight proportional control.
 
When the GT is in PSL when frequency raises GT will start unloading if and only if the rate of raise of frequency is more than the TNKR1_4 value which is 0.33%/min. In the above condition your speed error signal will become negative, which will reduce FSRN inturn your load will come down so if the frequency raise is more 52G may open on reverse(not on low forward power). This can be prevented either by raising TNKR1_4 value or by reducing FSKNG value.

In the same way when frequency drops load on your machine will increase but restricted by TTRXB.

regards
venu
 
Hi CSA,
Our grid operator always checks the Reserve Effectiveness (Ratio between actual MW to demanded MW ) in the event of a frequency disturbance. RE =1 is ideal however there is some tolerance within which machine have to respond. RE is calculated by unit specific mathematical model. At Some frequency disturbance the RE is not with in limit (Meaning Actual MW is much lower than Demanded in our case) . I know the droop setting make the GT response better however the DROOP can not be changed just like that without detailed study by OEM. What else could cause this RE failure ? Do GT performance indicators like air inlet filters , fuel gas compositions , part load operations or ambient conditions can affect this ?
Can you please help advise on this ? This machine is GE Alstom legacy F class machine.
Thank You
 
Hi CSA,
Our grid operator always checks the Reserve Effectiveness (Ratio between actual MW to demanded MW ) in the event of a frequency disturbance. RE =1 is ideal however there is some tolerance within which machine have to respond. RE is calculated by unit specific mathematical model. At Some frequency disturbance the RE is not with in limit (Meaning Actual MW is much lower than Demanded in our case) . I know the droop setting make the GT response better however the DROOP can not be changed just like that without detailed study by OEM. What else could cause this RE failure ? Do GT performance indicators like air inlet filters , fuel gas compositions , part load operations or ambient conditions can affect this ?
Can you please help advise on this ? This machine is GE Alstom legacy F class machine.
Thank You
Hi

I join this thread to mention that you experiencing grid frequency instability ...so how these questions regarding GT performance come to you...
Gas turbine generator should operate as smooth as possible even during grid frequency instability ...there are kind of PSS (to be tuned on AVR /DIGITAL EXCITER ) that can help to support smooth operations in transients modes..

Any time
Cheers..
 
mariselvan,

Hmmm..... Gas turbines and power output during grid frequency disturbances.

As we know (because we all know this) generator speed is a function of grid frequency when synchronized to an "infinite" grid with other generators and their prime movers. So, as grid frequency decreases the speed of the generator decreases, and since the gas turbine and generator are coupled together so, too, must the speed of the gas turbine (I am presuming a single-shaft gas turbine; whether or not the gas turbine is coupled to the generator through a Load (Reduction) Gear or not, it is still "directly" coupled. (And, vice versa--as grid frequency increases so too does the generator, and turbine, speed.) And, if the grid frequency is not only high or low but is also unstable--SO TOO will the generator and turbine speed be unstable. That's physics--and it ain't never going to change. No matter how much anyone wants it too, or thinks it should--if the grid frequency is high or low or unstable--ALL the generators and their prime movers are also going to have high- or low frequency or unstable frequency; it's unavoidable. Full stop. Period. (Think of it this way--how can one single generator remain at stable frequency and load when the grid frequency and load is high or low or unstalbe--AND the power coming out of the electrical outlet on the wall or supplying a motor be normal or unstable? There is no "smoothing" device that will allow one, or any number of, generators to operate at any other frequency than the grid frequency (the grid they are SYNCHRONIZED to (a VERY important and powerful word, "synchronized") so that one or any number of generators can operate at different frequencies than the grid--and still have 50.0 or 60.0 Hz coming out of the wall or the wire at the consumer's end? It's just NOT POSSIBLE. If the grid frequency is high or low or unstable then EVERY generator SYNCHRONIZED to that grid is also going to be experiencing high or low or unstablel frequency. Again, full stop. Period. End of discussion.)

The amount of air that flow through a single shaft heavy duty gas turbine is a function of the speed of the axial compressor. And, if the speed of the axial compressor is higher than it should be, or lower than it should be, or unstable--because the generator the turbine (and its axial compressor) is coupled to is SYNCHRONIZED to a grid who's frequency is high or low or unstable then the amount of air flowing through the axial compressor and into the combustor and through the turbine and into the exhaust is also going to be higher than normal or lower than normal or unstable. Full stop. Period.

Also, gas turbines (of any manufacturer) are termed "mass flow machines" which means that the more mass (fuel AND air) that can flow through the machine the more power it can produce. Limit either--the fuel flow or the air flow--and the power output of the turbine (and therefore the generator) is going to decrease. Full stop. Period.

Now, if the unit is a GE-design heavy duty gas turbine and it is operating at Base Load ("rated" output for the ambient and machine conditions) then it's exhaust temperature is already at maximum allowable. If the grid frequency decreases--TWO (count 'em--two) things happen. First, the turbine control system has to reduce the fuel in order to prevent the exhaust temperature from increasing (which it already is simply because the air flow through the machine is decreasing). Less fuel means lower turbine power output, which means lower generator output. Full stop. Period.

The second thing that's going to happen is that because the air flow through the machine is decreasing as the turbine-generator speed is decreasing because the grid frequency is decreasing--the power being produced by the turbine (and generator) is also going to decrease.

So--when the grid frequency decreases the turbine-generator power output is going to decrease for two reasons (if the unit is operating at Base Load!!! (Or Peak- or Peak Reserve Load)): the decrease in fuel, required to prevent exhaust temperature from increasing, AND the reduction in mass flow because the axial compressor speed is not at rated because the grid frequency is not at rated.

The opposite happens when the grid frequency is above normal and the turbine-generator is operating at Base Load. The turbine actually produces MORE power because of the increased air flow AND the fact that the turbine control system can actually increase the fuel because the exhaust temperature will be lower than allowable.

BOTH of these things are BAD for the grid frequency. When the grid frequency decreases, the grid regulators and operators (and Customers) want the power outputs of the generators and their prime movers to INCREASE--but if the generators, or the majority of generators, are gas turbines operating at Base Load their power output is doing to DECREASE!!! This is the dirty little secret of gas turbine-generators.

AND, when grid frequency increases the grid regulators and operators and Customers want generators operating at Base Load to DECREASE their power output. But, if the generators are drive by gas turbines, or the majority of generators SYNCHRONIZED to a grid are driven by gas turbines operating at Base Load, they will do the exact opposite and increase their power output.

And, if the grid frequency is unstable, meaning the speeds of the generators and their prime movers are unstable, then the power outputs of the generators are also going to be unstable. And, if the generators or the majority of generators are driven by gas turbines then the outputs of the machines are also going to be unstable. Full stop. Period. It doesn't make any difference what control system is in use or how it's tuned--if the grid frequency is high or low or unstable, the generator output is also going to be unstable. That's the nature of the business--and of AC power generation.

So, mariselvan, you really haven't provided much in the way of details of the grid frequency disturbances or how the unit(s) are being operated at your site (at Base Load; at 90% of Base Load; etc.). So, there's really not much more we can say.

BUT I can add this: If the units are your site are being operated with Pre-Selected Load Control enabled and active--even with PFR (Primary Frequency Response) and the units are at or very near Base Load, then even the vaunted Mark* turbine control system isn't going to make your RE any closer to 1. It's just not going to happen. Pre-Selected Load Control--even with PFR--is an extremely poor way to operate a GE-design heavy duty gas turbine. Period. Full stop. AND, it contributes to grid instability.

Hope this helps! You want more information? We need more details.
 
It’s entirely possible that the unit you are describing doesn’t have a Mark* turbine control system. Further, if that’s true, then the programming is most likely not going to be exactly as I tried to describe. The fact remains, gas turbines running at or near rated load for a given ambient and set of machine conditions do not always respond as they should during off-frequency conditions.

AND, if Belfort, France, was responsible for the programming of the turbine control system then one can safely bet there will be some unique programming as they are known for being a little experimental in their control philosophies.

My sincere apologies for any confusion I may have caused. Without being able to examine and review the programming in the turbine control system I can’t say with any degree of certainty how it works or responds. The concept of “demand” and RE is actually very unusual for me.As ControlsGuy25 said, you need to review the programming in the turbine control system at your site.
 
Top