How can it make the grid disturbance worse?
How can Pre-Selected load control can make system/grid instability worse?
How can Pre-Selected load control can make system/grid instability worse?
During a grid frequency disturbance the problem is either, 1) the load on the system/grid exceeds the amount of generation currently available on the grid; or, 2) the load on the system/grid is less than the amount of generation currently available on the grid. This causes the grid frequency to either, 1) decrease; or, 2) increase (respectively). Droop Speed Control, for machines operating at loads less than Base Load (in other words, Part Load) will cause the machine to respond in an appropriate--and expected (by the system/grid operators) manner. It WILL NOT act to return the system/grid frequency to normal, but it will 1) increase output at the reduced frequency; or, 2) reduce output at the increased frequency. Which is, again, what the system/grid operators expect and which is very helpful in stabilizing the grid frequency until the system/grid operators can respond to return the grid frequency to normal.
Droop Speed Control is the most common method of operation of a prime mover and generator for machines connected to any system/grid with other machines (GE machines; Siemens machines; any manufacturer; steam turbines; gas turbines; hydro turbines; reciprocating engnies). For GE-design heavy duty gas turbines operating at loads LESS THAN Base Load Droop Speed Control is active. Always. When Pre-selected Load Control is active it over-rides Droop Speed Control. Pre-Selected Load Control looks at the Pre-Selected Load Control Reference (setpoint) and adjusts the machine speed reference to make the actual load equal to the Pre-Selected Load Control Reference. Very straightforward--WHEN THE SYSTEM/GRID FREQUENCY IS AT RATED.
When the system/grid frequency deviates from rated, Droop Speed Control will adjust the load as described above to support grid stability until such time as the system/grid operators can respond appropriately and return the frequency to normal. HOWEVER, when Pre-Selected Load Control is active it will sense the change in load and make an adjustment to OVER-RIDE the adjustment that Droop Speed Control is making to try to support grid stability. And, then Droop Speed Control senses the frequency is still not normal and tries again to respond appropriately--but, Pre-Selected Load Control senses the load drifting from the Pre-Selected Load Control Reference (setpoint), then Droop Speed Control tries again to respond appropriately and Pre-Selected Load Control counters the response again, and on, and on, and on. The load is going up and down, or down and up, continuously as long a Pre-Selected Load Control is active or until the system/grid frequency is returned to normal.
So, when the machine is operating at Part Load with Pre-Selected Load Control enabled and active and the system grid frequency goes down and Droop Speed Control increases load Pre-Selected Load Control does the exact opposite, reducing the load after Droop Speed Control raises it, and the Droop Speed Control increases the load again and Pre-Selected Load Control reduces it again, and so on, and so on, until someone either cancels Pre-Selected Load Control OR the system/grid frequency returns to normal. (The opposite happens when the grid frequency increases, Droop Speed Control lowers load and Pre-Selected Load Control raises load, back and forth, back and forth, back and forth.) This oscillating of load makes the grid more unstable at a time when it needs to be stable.
That's how it contributes to grid instability AND it doesn't provide the support of helping with the load/generation imbalance. It doesn't let the machine increase or reduce load as it should to help with maintaining grid stability until the system/grid operators can correct the situation. System/grid operators rely on the proper response of machines operating in Droop Speed Control to help with grid stability; but Pre-Selected Load Control causes machine load to vary instead of being stable at whatever load it needs to be at based on the Droop regulation and frequency deviation.
Another thing that happens is that many (most, actually) Plant Managers and Operations Supervisors believe that during a grid disturbance (frequency or otherwise) that THEIR MACHINE should stay at a constant load regardless of what the grid is doing--which is just completely false, incorrect and indicates how little they actually know and understand AC power generation and machine operation. Which trickles down to the operators, who then think the machine load should be stable during a system/grid instability.
So, that's it in a nutshell. No doubts; just fact.
Droop Speed Control is the most common method of operation of a prime mover and generator for machines connected to any system/grid with other machines (GE machines; Siemens machines; any manufacturer; steam turbines; gas turbines; hydro turbines; reciprocating engnies). For GE-design heavy duty gas turbines operating at loads LESS THAN Base Load Droop Speed Control is active. Always. When Pre-selected Load Control is active it over-rides Droop Speed Control. Pre-Selected Load Control looks at the Pre-Selected Load Control Reference (setpoint) and adjusts the machine speed reference to make the actual load equal to the Pre-Selected Load Control Reference. Very straightforward--WHEN THE SYSTEM/GRID FREQUENCY IS AT RATED.
When the system/grid frequency deviates from rated, Droop Speed Control will adjust the load as described above to support grid stability until such time as the system/grid operators can respond appropriately and return the frequency to normal. HOWEVER, when Pre-Selected Load Control is active it will sense the change in load and make an adjustment to OVER-RIDE the adjustment that Droop Speed Control is making to try to support grid stability. And, then Droop Speed Control senses the frequency is still not normal and tries again to respond appropriately--but, Pre-Selected Load Control senses the load drifting from the Pre-Selected Load Control Reference (setpoint), then Droop Speed Control tries again to respond appropriately and Pre-Selected Load Control counters the response again, and on, and on, and on. The load is going up and down, or down and up, continuously as long a Pre-Selected Load Control is active or until the system/grid frequency is returned to normal.
So, when the machine is operating at Part Load with Pre-Selected Load Control enabled and active and the system grid frequency goes down and Droop Speed Control increases load Pre-Selected Load Control does the exact opposite, reducing the load after Droop Speed Control raises it, and the Droop Speed Control increases the load again and Pre-Selected Load Control reduces it again, and so on, and so on, until someone either cancels Pre-Selected Load Control OR the system/grid frequency returns to normal. (The opposite happens when the grid frequency increases, Droop Speed Control lowers load and Pre-Selected Load Control raises load, back and forth, back and forth, back and forth.) This oscillating of load makes the grid more unstable at a time when it needs to be stable.
That's how it contributes to grid instability AND it doesn't provide the support of helping with the load/generation imbalance. It doesn't let the machine increase or reduce load as it should to help with maintaining grid stability until the system/grid operators can correct the situation. System/grid operators rely on the proper response of machines operating in Droop Speed Control to help with grid stability; but Pre-Selected Load Control causes machine load to vary instead of being stable at whatever load it needs to be at based on the Droop regulation and frequency deviation.
Another thing that happens is that many (most, actually) Plant Managers and Operations Supervisors believe that during a grid disturbance (frequency or otherwise) that THEIR MACHINE should stay at a constant load regardless of what the grid is doing--which is just completely false, incorrect and indicates how little they actually know and understand AC power generation and machine operation. Which trickles down to the operators, who then think the machine load should be stable during a system/grid instability.
So, that's it in a nutshell. No doubts; just fact.
Same Droop Speed regulation value (say, 4%), and different no load frequencies? I wish I understood how no load frequency became a part of the question....
Frequency when not loaded (when the generator breaker is NOT closed--it's open) doesn't mean anything. 50 Hz generators are 50 Hz generators; if the speed of one generator rotor is such that it's frequency when not loaded is 49.37 Hz, and the frequency of another generator when not loaded is 50.29 Hz, that doesn't mean anything--because when the units are being synchronized the frequency (speed) will have to be matched to the system/grid frequency (whether that's 50.03 Hz, or 49.95 Hz, or 50.00 Hz)--the synchronizing process requires the frequency of the incoming generator(s) to be equal to or greater than the system/grid frequency and the operator or the automatic synchronizer will make that adjustment. (NOTE: While it IS possible to synchronized a generator whose frequency is less than the system/grid frequency that means that immediately after the generator breaker is closed the real power flow will be slightly negative, less than 0.0 watts/kW/MW (negative, or reverse, real power into the generator from the system/grid). It's just that for MOST synchronous generators and their prime movers it's best if they are synchronized when their frequency is at least equal to or slightly greater than 0.0 watts/kW/MW (positive real power flow out of the generator onto the system/grid).
GE's control philosophy is to have the generator at 100.3% of rated speed when synchronizing begins--and the operator (if the unit is being synchronized manually) or the automatic synchronizer (if the unit is being synchronized automatically) will then begin the process of adjusting the machine's speed (frequency) to be equal to or slightly greater than the system/grid frequency before closing the generator breaker. Now, that's their typical FSNL (Full Speed-No Load) speed setpoint, but some sites have changed this value (usually because the system/grid operator asked them to change the value to make for "smoother" synchronizations to the system/grid). 0.3% of 4% Droop regulation for a 25 MW machine is approximately 1.875 MW ((0.3%/4%)*25 MW)=1.875 MW. For a 40 MW machine it's 3.0 MW ((0.3%/4%)*40 MW)=3.0 MW. For a 120 MW machine it's 9.00 MW ((0.3%/4%)*120 MW=9.00 MW). So, the value of real power produced immediately after synchronization can vary greatly depending on the rating of the machine--even if the machines all have the same Droop regulation setpoint (in this case, 4%--which is kind of the standard for heavy duty gas turbines around the world for all manufacturers). Remember, a Droop regulation setpoint of 4% means that when the machine speed reference is 104% the load the machine will be able to produce rated load (the rating of the prime mover) when at rated speed (frequency).
And, also, do not forget: Speed and frequency are directly related.
Energy flow-rate into the machine=((Machine Speed Reference - Actual Machine Speed) * Gain) + Offset (this is GE's basic Droop Speed Control Formula). You should see that for any GE-design heavy duty gas turbine operating on typical fuels at turbine nameplate rated ambient conditions when the inlet air filters and the IGVs are clean, and the axial compressor is clean and the machine back-pressure is normal and the ambient temperature and -humidity are at nameplate rating and the fuel matches what was used to program the machine when it was being built that when the turbine speed reference (TNR) is approximately 104% and the actual speed (TNH) is 100% (for a difference of 4%--the Droop regulation setpoint of MOST GE-design heavy duty gas turbines!) the load will be at or very near the nameplate rated power output for Base Load of the turbine.
[I know; I wrote ((Machine Actual Speed - Machine Speed Reference)*Gain)+Offset in one response; some control system manufacturers use that formula. It's the same with different gain and offset values. GE's formula is ((TNR-TNH)*Gain+Offset).]
I want to mention something that MANY people (including a LOT of commissioning personnel that work on GE-design heavy duty gas turbines--the 100% speed value for ANY machine that has a Load (Reduction) Gear between the turbine and the generator is (should be) the Load (Reduction) Gear nameplate ratio. For example, many GE Frame 5 and Frame 6 machines (both of which have Load (Reduction) Gears) come from the factory with a 100% speed setpoint of 5130 RPM, some are even 5100 RPM. But the person commissioning the machine during start-up should have looked at the Load (Reduction) Gear nameplate to find out exactly what the input speed is for 3000 RPM output (for a 50 Hz machine) or 3600 RPM (for a 60 Hz machine) and then adjusted the Mark* control system 100% speed value to match the Load (Reduction) Gear nameplate. So, if the reduction gear nameplate says 5134 RPM input equals 3000 RPM output, the Mark* 100% speed setpoint should be 5134 RPM, not 5100 RPM. If the nameplate says 5122 RPM input equal 3000 RPM output the Mark* 100% speed setpoint should be 5122 RPM. Does it make a huge difference if the speed setpoint doesn't exactly match the Load (Reduction) Gear nameplate rating? Not really, only that it may take a little longer to automatically synchronize the machine, and when running on-line and the generator frequency is 50.00 Hz but the speed is not 5130 RPM, that kind of bothers some people (it bothers me, anyway). The machine will still produce rated power at ISO conditions it will just not have exact numbers for indication. EVERYTHING else will be normal, and, again, it might take a little longer to automatically synchronize--but it will work adequately. There are LOTS of GE machines out there that don't have the exact proper 100% speed setpoint in the Mark*--and they all work adequately. We're only talking about hundredths of a percent.)
Frequency when not loaded (when the generator breaker is NOT closed--it's open) doesn't mean anything. 50 Hz generators are 50 Hz generators; if the speed of one generator rotor is such that it's frequency when not loaded is 49.37 Hz, and the frequency of another generator when not loaded is 50.29 Hz, that doesn't mean anything--because when the units are being synchronized the frequency (speed) will have to be matched to the system/grid frequency (whether that's 50.03 Hz, or 49.95 Hz, or 50.00 Hz)--the synchronizing process requires the frequency of the incoming generator(s) to be equal to or greater than the system/grid frequency and the operator or the automatic synchronizer will make that adjustment. (NOTE: While it IS possible to synchronized a generator whose frequency is less than the system/grid frequency that means that immediately after the generator breaker is closed the real power flow will be slightly negative, less than 0.0 watts/kW/MW (negative, or reverse, real power into the generator from the system/grid). It's just that for MOST synchronous generators and their prime movers it's best if they are synchronized when their frequency is at least equal to or slightly greater than 0.0 watts/kW/MW (positive real power flow out of the generator onto the system/grid).
GE's control philosophy is to have the generator at 100.3% of rated speed when synchronizing begins--and the operator (if the unit is being synchronized manually) or the automatic synchronizer (if the unit is being synchronized automatically) will then begin the process of adjusting the machine's speed (frequency) to be equal to or slightly greater than the system/grid frequency before closing the generator breaker. Now, that's their typical FSNL (Full Speed-No Load) speed setpoint, but some sites have changed this value (usually because the system/grid operator asked them to change the value to make for "smoother" synchronizations to the system/grid). 0.3% of 4% Droop regulation for a 25 MW machine is approximately 1.875 MW ((0.3%/4%)*25 MW)=1.875 MW. For a 40 MW machine it's 3.0 MW ((0.3%/4%)*40 MW)=3.0 MW. For a 120 MW machine it's 9.00 MW ((0.3%/4%)*120 MW=9.00 MW). So, the value of real power produced immediately after synchronization can vary greatly depending on the rating of the machine--even if the machines all have the same Droop regulation setpoint (in this case, 4%--which is kind of the standard for heavy duty gas turbines around the world for all manufacturers). Remember, a Droop regulation setpoint of 4% means that when the machine speed reference is 104% the load the machine will be able to produce rated load (the rating of the prime mover) when at rated speed (frequency).
And, also, do not forget: Speed and frequency are directly related.
Energy flow-rate into the machine=((Machine Speed Reference - Actual Machine Speed) * Gain) + Offset (this is GE's basic Droop Speed Control Formula). You should see that for any GE-design heavy duty gas turbine operating on typical fuels at turbine nameplate rated ambient conditions when the inlet air filters and the IGVs are clean, and the axial compressor is clean and the machine back-pressure is normal and the ambient temperature and -humidity are at nameplate rating and the fuel matches what was used to program the machine when it was being built that when the turbine speed reference (TNR) is approximately 104% and the actual speed (TNH) is 100% (for a difference of 4%--the Droop regulation setpoint of MOST GE-design heavy duty gas turbines!) the load will be at or very near the nameplate rated power output for Base Load of the turbine.
[I know; I wrote ((Machine Actual Speed - Machine Speed Reference)*Gain)+Offset in one response; some control system manufacturers use that formula. It's the same with different gain and offset values. GE's formula is ((TNR-TNH)*Gain+Offset).]
I want to mention something that MANY people (including a LOT of commissioning personnel that work on GE-design heavy duty gas turbines--the 100% speed value for ANY machine that has a Load (Reduction) Gear between the turbine and the generator is (should be) the Load (Reduction) Gear nameplate ratio. For example, many GE Frame 5 and Frame 6 machines (both of which have Load (Reduction) Gears) come from the factory with a 100% speed setpoint of 5130 RPM, some are even 5100 RPM. But the person commissioning the machine during start-up should have looked at the Load (Reduction) Gear nameplate to find out exactly what the input speed is for 3000 RPM output (for a 50 Hz machine) or 3600 RPM (for a 60 Hz machine) and then adjusted the Mark* control system 100% speed value to match the Load (Reduction) Gear nameplate. So, if the reduction gear nameplate says 5134 RPM input equals 3000 RPM output, the Mark* 100% speed setpoint should be 5134 RPM, not 5100 RPM. If the nameplate says 5122 RPM input equal 3000 RPM output the Mark* 100% speed setpoint should be 5122 RPM. Does it make a huge difference if the speed setpoint doesn't exactly match the Load (Reduction) Gear nameplate rating? Not really, only that it may take a little longer to automatically synchronize the machine, and when running on-line and the generator frequency is 50.00 Hz but the speed is not 5130 RPM, that kind of bothers some people (it bothers me, anyway). The machine will still produce rated power at ISO conditions it will just not have exact numbers for indication. EVERYTHING else will be normal, and, again, it might take a little longer to automatically synchronize--but it will work adequately. There are LOTS of GE machines out there that don't have the exact proper 100% speed setpoint in the Mark*--and they all work adequately. We're only talking about hundredths of a percent.)
The question arose when thinking of how to modify load sharing (after paralleling) when two generators are in droop mode, or when one generator is isochronous and the other is in droop mode.
My understandings:
Two generators in droop mode (after paralleling):
Same droop and same no load frequency --> The generator shares the load equally
Different droop and same no load frequency --> The generator with lesser droop shares more power
Same droop and Different no load frequency --> The generator with higher no load frequency shares more power
Different droop and different no load frequency --> ??
One Generator in Isochronous and one in droop mode:
When the no load frequency (after paralleling) of the droop mode generator is less or equal to the speed reference of the Isochronous generator, the droop mode generator will not accept the load
when the no load frequency (after paralleling) of the droop mode generator is increased gradually to accept the load, the load on the isochronous generator is removed when the no load frequency exceeds the isochronous generators speed reference.
To portray this in picture, same droop but different no load frequency;
My understandings:
Two generators in droop mode (after paralleling):
Same droop and same no load frequency --> The generator shares the load equally
Different droop and same no load frequency --> The generator with lesser droop shares more power
Same droop and Different no load frequency --> The generator with higher no load frequency shares more power
Different droop and different no load frequency --> ??
One Generator in Isochronous and one in droop mode:
When the no load frequency (after paralleling) of the droop mode generator is less or equal to the speed reference of the Isochronous generator, the droop mode generator will not accept the load
when the no load frequency (after paralleling) of the droop mode generator is increased gradually to accept the load, the load on the isochronous generator is removed when the no load frequency exceeds the isochronous generators speed reference.
To portray this in picture, same droop but different no load frequency;
If a machine and its prime mover are synchronized to a system/grid or even just one other machine (generator and its prime mover) BOTH machines run at the same frequency. Period. Full stop. That's how AC power systems work. If every generator synchronized to a system/grid operated at a different frequency, how would one get 50 Hz (or 60 Hz) out of the wall socket or the light fixture mains supply? There's no "smoothing" filtering done on frequency to get the "average" frequency; it just doesn't happen.
That graph is very misleading--as are MOST of the graphs and passages in textbooks and reference books about Droop Speed Control changing the machine speed as the machine is loaded. If the machine is operating independently of any other machine, or if it's being operated in a small system/grid with other machines ALL in Droop Speed Control as load increases the speed of all the machines will decrease, in proportion to the percentage of Droop Speed regulation. HOWEVER, if the system/grid is larger even if ALL of the machines are operating in Droop Speed Control mode the machines are NOT going to behave like this. The more machines (generators and their prime movers) there are the more "inertia" the entire system has.
If you're referring to "no load" frequency as being the frequency of the machine when the generator breaker is closed and the load of the machine is 0 watts/kW/MW, if it's a single machine, then, yes, the frequency can be whatever it is (normal or abnormal) when the machine is at 0 watts/kW/MW. And, yes, if two machines were BOTH operating in Droop Speed Control while synchronized together then it's possible that the frequency of the system/grid they are powering could be higher or lower than normal--but then that's not a well-regulated system/grid and the frequency of the wall outlet and the light fixture(s) IS NOT normal. Do you consider 61.2 Hz to be normal? or 64 Hz? Not in my neighborhood they're not normal. And if you're powering refineries, or a cement plant, or a food processing plant they're going to be VERY upset if you supply them with 61.2 Hz or 64 Hz. Do it on a regular basis, and there's going to be some contract re-negotiations going on pretty quick.
I think you're not reading the graph correctly. Take the G2 line, for example. It crosses the P(MW) axis at 3 MW. That means the frequency of the output is 60.0 Hz (normal) for that particular Droop Speed Control setting. For a machine with 4% Droop regulation and a power rating of 6 MW supplying power to a system/grid with no generator operating in Isochronous Speed Control mode and NO automatic control of system frequency that would equate to a Machine Speed Reference of 102%.
G1's line "crosses" the P(MW) axis at 6 MW, meaning that it's rating is 6 MW. For that machine, it is operating at 104% Machine Speed Reference and is producing rated power at rated frequency, 60.0 Hz.
That means the load on the system is 9 MW, with G1 supplying 6 MW and G2 supplying 3 MW. The two machines, together, can provide 12 MW, but at this time the load is only 9 MW. BOTH machines are rated at 6 MW each, and BOTH machines have 4% Droop regulation setpoints. There is no automatic frequency control in this power plant, and the operators are busy eating some very biryani and reading newspapers.
Let's say that the load on the system drops by 3 MW (a very large pump motor somewhere is stopped). The load is now 6 MW. And NO ONE or no thing changes the Machine Speed References. In this case since the machines have the same rating (6 MW) and the same Droop Regulation setpoint (4%) each machine will decrease its output by the same amount: 1.5 MW, for a total of 3 MW. G1's load decreases to 4.5 MW AND it's frequency increases from 60.0 Hz to 60.6 Hz. G2's load decreases by 1.5 MW to 1.5 MW AND it's frequency increases to 60.6 Hz. Droop Speed Control caused the load being powered by the two generators to decrease to match the load being powered by the two generators. Since the two machines are rated equally (6 MW) AND both machines have the same Droop Regulation setpoint (4%) they split the change in load equally between them--but the result is that even though the load on the two machines decreased the frequency of BOTH machines increased, and by the same amount (0.6 Hz from 60.0 Hz to 60.6 Hz). The Machine Speed References of BOTH machines did not change to cause the load to increase. However, the Machine Actual Speeds of BOTH machines DID change due to the decrease in the system load. Per the Droop Speed Control formula
Energy flow-rate into Machine=(Machine Speed Reference - Machine Actual Speed)*Gain+Offset
here's how the machines' energy flow-rates changed:
G1@6 MW=((104%-(60.0Hz/60.0 HZ*100%))*Gain+Offset=(104%-100%)*Gain+Offset=(104)*Gain+Offset
[email protected] MW=((104%-(60.6 Hz/60 Hz)*100%))*Gain+Offset=((104%-101%)*Gain+Offset=(103)*Gain+Offset
G2@3 MW MW=((102%-(60.0Hz/60.0 HZ*100%))*Gain+Offset=(102%-100%)*Gain+Offset=(102)*Gain+Offset
[email protected] MW=((102%-(60.6 Hz/60 Hz)*100%))*Gain+Offset=((102%-101%)*Gain+Offset=(101)*Gain+Offset
The difference between Machine Speed Reference and Machine Actual Speed for each machine changed by 1%.
[NOTE: The values of both Offset and Gain for the two machines are probably the same (if they are similar machines made by the same manufacturer burning the same fuel with the same auxiliaries (fuel control valves). Their actual values aren't important because they don't change as the machine is operating; they remain the same from start-up to shutdown, every day of every year, especially if they are GE-design heavy duty gas turbines with Mark* turbine control systems.]
You can use the "dotted" lines to see what's happening, also. But, it takes some explanation about the circumstances to understand how to read and understand the graph by explaining the details of the conditions the graph is explaining.
BOTH machine's actual speeds increased by the SAME AMOUNT because of the decrease in LOAD and the fact that both machines have the same Droop Regulation setpoint. NEITHER machine's speed reference changed (G1 remained at 104% and G2's remained at 102%), but since BOTH machine's actual speeds increased the amount of energy flowing into the two machines also decreased by the same amount (1%*Gain+Offset). It's impossible for the machines to continue to produce the same power as before the system/grid load changed by 3MW; that would violate immutable laws of physics. BUT, because the machine's speed references DIDN'T change the fuel didn't change by the amount required to maintain 60.0 Hz; a conscious, well-trained operator has to do that, or a well-programmed frequency control system has to do that. (Or, if one of the machines was operating in Isochronous Speed Control it would do that...!)
Now, let's say the load dropped by another 3 MW, from 6 MW to 3 MW. Because the machines have the same Droop Regulation setpoint (4%) and the same rating they are going to split the load change again; G1's load will decrease from 4.5 MW to 3MW, and G2's load will decrease from 1.5 MW to 0 MW. The system/grid frequency--and that of BOTH machines--will increase to 61.2 Hz.
Here's how the machines' energy flow-rates changed:
[email protected] MW=((104%-(60.6 Hz/60 Hz)*100%))*Gain+Offset=((104%-101%)*Gain+Offset=(103)*Gain+Offset
G1@3 MW=((104%-(61.2Hz/60.0 HZ*100%))*Gain+Offset=(104%-102%)*Gain+Offset=(102)*Gain+Offset
[email protected] MW=((102%-(60.6Hz/60.0 HZ*100%))*Gain+Offset=(102%-101%)*Gain+Offset=(101)*Gain+Offset
G2@0 MW=((102%-(61.2 Hz/60.0 Hz)*100%))*Gain+Offset=((102%-102%)*Gain+Offset=(100)*Gain+Offset
Again, the load is what the load is and the machines CAN'T produce any more than the load they are supplying requires.
THAT is how the graph is to be read. NOT that the "no load frequency" of the two machines are different. Now, you didn't provide any supporting documentation for the graph--BUT based on decades of reading textbooks and reference books on the subject of Droop Speed Control I can bet that there was very little in the way of the details concerning the operating conditions of the scenario, or how to read the graph. These graphs are just usually inserted in the document/book/manual with nothing to help explain how to read the graph. The graph isn't telling the reader that there is nothing controlling frequency (that's NOT Droop Speed Control's job--to control frequency!). It doesn't say that G1's Machine Speed Reference is 104% and G2's Machine Speed Reference is 102%. It doesn't say these are the only two machines supplying the load at the time. It doesn't say the two machines are operating on Droop Speed Control.
THIS is ONE aspect of Droop Speed Control--the part where the machine responds to changes in frequency to try to support grid stability. Pre-Selected Load Control, when active, acts in EXACTLY THE OPPOSITE MANNER and actually fights Droop Speed Control, with Droop Speed Control fighting to try to do its job (change load as frequency changes.)
That graph is very misleading--as are MOST of the graphs and passages in textbooks and reference books about Droop Speed Control changing the machine speed as the machine is loaded. If the machine is operating independently of any other machine, or if it's being operated in a small system/grid with other machines ALL in Droop Speed Control as load increases the speed of all the machines will decrease, in proportion to the percentage of Droop Speed regulation. HOWEVER, if the system/grid is larger even if ALL of the machines are operating in Droop Speed Control mode the machines are NOT going to behave like this. The more machines (generators and their prime movers) there are the more "inertia" the entire system has.
If you're referring to "no load" frequency as being the frequency of the machine when the generator breaker is closed and the load of the machine is 0 watts/kW/MW, if it's a single machine, then, yes, the frequency can be whatever it is (normal or abnormal) when the machine is at 0 watts/kW/MW. And, yes, if two machines were BOTH operating in Droop Speed Control while synchronized together then it's possible that the frequency of the system/grid they are powering could be higher or lower than normal--but then that's not a well-regulated system/grid and the frequency of the wall outlet and the light fixture(s) IS NOT normal. Do you consider 61.2 Hz to be normal? or 64 Hz? Not in my neighborhood they're not normal. And if you're powering refineries, or a cement plant, or a food processing plant they're going to be VERY upset if you supply them with 61.2 Hz or 64 Hz. Do it on a regular basis, and there's going to be some contract re-negotiations going on pretty quick.
I think you're not reading the graph correctly. Take the G2 line, for example. It crosses the P(MW) axis at 3 MW. That means the frequency of the output is 60.0 Hz (normal) for that particular Droop Speed Control setting. For a machine with 4% Droop regulation and a power rating of 6 MW supplying power to a system/grid with no generator operating in Isochronous Speed Control mode and NO automatic control of system frequency that would equate to a Machine Speed Reference of 102%.
G1's line "crosses" the P(MW) axis at 6 MW, meaning that it's rating is 6 MW. For that machine, it is operating at 104% Machine Speed Reference and is producing rated power at rated frequency, 60.0 Hz.
That means the load on the system is 9 MW, with G1 supplying 6 MW and G2 supplying 3 MW. The two machines, together, can provide 12 MW, but at this time the load is only 9 MW. BOTH machines are rated at 6 MW each, and BOTH machines have 4% Droop regulation setpoints. There is no automatic frequency control in this power plant, and the operators are busy eating some very biryani and reading newspapers.
Let's say that the load on the system drops by 3 MW (a very large pump motor somewhere is stopped). The load is now 6 MW. And NO ONE or no thing changes the Machine Speed References. In this case since the machines have the same rating (6 MW) and the same Droop Regulation setpoint (4%) each machine will decrease its output by the same amount: 1.5 MW, for a total of 3 MW. G1's load decreases to 4.5 MW AND it's frequency increases from 60.0 Hz to 60.6 Hz. G2's load decreases by 1.5 MW to 1.5 MW AND it's frequency increases to 60.6 Hz. Droop Speed Control caused the load being powered by the two generators to decrease to match the load being powered by the two generators. Since the two machines are rated equally (6 MW) AND both machines have the same Droop Regulation setpoint (4%) they split the change in load equally between them--but the result is that even though the load on the two machines decreased the frequency of BOTH machines increased, and by the same amount (0.6 Hz from 60.0 Hz to 60.6 Hz). The Machine Speed References of BOTH machines did not change to cause the load to increase. However, the Machine Actual Speeds of BOTH machines DID change due to the decrease in the system load. Per the Droop Speed Control formula
Energy flow-rate into Machine=(Machine Speed Reference - Machine Actual Speed)*Gain+Offset
here's how the machines' energy flow-rates changed:
G1@6 MW=((104%-(60.0Hz/60.0 HZ*100%))*Gain+Offset=(104%-100%)*Gain+Offset=(104)*Gain+Offset
[email protected] MW=((104%-(60.6 Hz/60 Hz)*100%))*Gain+Offset=((104%-101%)*Gain+Offset=(103)*Gain+Offset
G2@3 MW MW=((102%-(60.0Hz/60.0 HZ*100%))*Gain+Offset=(102%-100%)*Gain+Offset=(102)*Gain+Offset
[email protected] MW=((102%-(60.6 Hz/60 Hz)*100%))*Gain+Offset=((102%-101%)*Gain+Offset=(101)*Gain+Offset
The difference between Machine Speed Reference and Machine Actual Speed for each machine changed by 1%.
[NOTE: The values of both Offset and Gain for the two machines are probably the same (if they are similar machines made by the same manufacturer burning the same fuel with the same auxiliaries (fuel control valves). Their actual values aren't important because they don't change as the machine is operating; they remain the same from start-up to shutdown, every day of every year, especially if they are GE-design heavy duty gas turbines with Mark* turbine control systems.]
You can use the "dotted" lines to see what's happening, also. But, it takes some explanation about the circumstances to understand how to read and understand the graph by explaining the details of the conditions the graph is explaining.
BOTH machine's actual speeds increased by the SAME AMOUNT because of the decrease in LOAD and the fact that both machines have the same Droop Regulation setpoint. NEITHER machine's speed reference changed (G1 remained at 104% and G2's remained at 102%), but since BOTH machine's actual speeds increased the amount of energy flowing into the two machines also decreased by the same amount (1%*Gain+Offset). It's impossible for the machines to continue to produce the same power as before the system/grid load changed by 3MW; that would violate immutable laws of physics. BUT, because the machine's speed references DIDN'T change the fuel didn't change by the amount required to maintain 60.0 Hz; a conscious, well-trained operator has to do that, or a well-programmed frequency control system has to do that. (Or, if one of the machines was operating in Isochronous Speed Control it would do that...!)
Now, let's say the load dropped by another 3 MW, from 6 MW to 3 MW. Because the machines have the same Droop Regulation setpoint (4%) and the same rating they are going to split the load change again; G1's load will decrease from 4.5 MW to 3MW, and G2's load will decrease from 1.5 MW to 0 MW. The system/grid frequency--and that of BOTH machines--will increase to 61.2 Hz.
Here's how the machines' energy flow-rates changed:
[email protected] MW=((104%-(60.6 Hz/60 Hz)*100%))*Gain+Offset=((104%-101%)*Gain+Offset=(103)*Gain+Offset
G1@3 MW=((104%-(61.2Hz/60.0 HZ*100%))*Gain+Offset=(104%-102%)*Gain+Offset=(102)*Gain+Offset
[email protected] MW=((102%-(60.6Hz/60.0 HZ*100%))*Gain+Offset=(102%-101%)*Gain+Offset=(101)*Gain+Offset
G2@0 MW=((102%-(61.2 Hz/60.0 Hz)*100%))*Gain+Offset=((102%-102%)*Gain+Offset=(100)*Gain+Offset
Again, the load is what the load is and the machines CAN'T produce any more than the load they are supplying requires.
THAT is how the graph is to be read. NOT that the "no load frequency" of the two machines are different. Now, you didn't provide any supporting documentation for the graph--BUT based on decades of reading textbooks and reference books on the subject of Droop Speed Control I can bet that there was very little in the way of the details concerning the operating conditions of the scenario, or how to read the graph. These graphs are just usually inserted in the document/book/manual with nothing to help explain how to read the graph. The graph isn't telling the reader that there is nothing controlling frequency (that's NOT Droop Speed Control's job--to control frequency!). It doesn't say that G1's Machine Speed Reference is 104% and G2's Machine Speed Reference is 102%. It doesn't say these are the only two machines supplying the load at the time. It doesn't say the two machines are operating on Droop Speed Control.
THIS is ONE aspect of Droop Speed Control--the part where the machine responds to changes in frequency to try to support grid stability. Pre-Selected Load Control, when active, acts in EXACTLY THE OPPOSITE MANNER and actually fights Droop Speed Control, with Droop Speed Control fighting to try to do its job (change load as frequency changes.)
Thank you for the detailed explanation!
If I understand correctly, when the 3MW load drops, either from the grid or from the parallel droop generators, the speed(frequency) of the grid increases or the frequency of the parallel running droop generators increases, with the explanation being that the fuel that was previously used to generate the now removed 3MW is the cause of the increase in speed(frequency).
If I understand correctly, when the 3MW load drops, either from the grid or from the parallel droop generators, the speed(frequency) of the grid increases or the frequency of the parallel running droop generators increases, with the explanation being that the fuel that was previously used to generate the now removed 3MW is the cause of the increase in speed(frequency).
You are correct. And, the speed (frequency) of BOTH the generators and the system/grid increase when load is reduced.
If you can post any supporting documentation (paragraphs?) that accompanied the graph in your source, it would be very helpful. I presumed it was trying to show two generators, synchronized together and powering a small system/grid, with both generator prime movers operating in Droop Speed Control with no manual or automatic correction. But, it may have been trying to show two generators, both with 4% Droop Regulation, each one operating independently of the other (not synchronized to the same system/grid. It kind of works both ways.
I want to caution you, and anyone reading this thread--When a GE-design heavy duty gas turbine is operating with BASE LOAD enabled and active if the grid frequency decreases from normal the machine CANNOT INCREASE its load!!! It's already operating at the maximum, optimal power output for the given ambient and machine conditions. AND, IN FACT the load of the machine will DECREASE (it won't maintain the Base Load output) because as the system/grid frequency decreases the speed of the axial compressor decreases which reduces air flow through the machine which causes the exhaust temperature to increase which causes the Mark* to reduce fuel which reduces the load being produced by the machine. This is a dirty little secret of gas turbines--of any manufacturer.
If a machine is operating on Droop Speed Control very near Base Load Rating and the grid frequency decreases the load will increase until the Base Load power output for the given operating conditions is reached and if the grid frequency decreases any further the load will decrease as the axial compressor speed decreases.
Conversely, if a machine is running with BASE LOAD active and the grid frequency increases the power output of the machine will increase--because the axial compressor speed increases which pushes more air through the turbine and exhaust sections of the machine which cools the exhaust temperature which causes the Mark* to increase fuel to maintain the allowable exhaust temperature. This is part of the dirty little secret of gas turbines. GE sells software modifications to allow a machine running with BASE LOAD active to properly respond to grid frequency disturbances, but it has to be purchased and the software modifications installed and tested. GE also sells a software modification which allows the machine to properly respond to grid frequency disturbances while Pre-Selected Load Control is active for as long as the grid frequency is outside of an allowable range, which means the machine load will change appropriately and will not remain stable during a grid frequency disturbance, but once the grid frequency disturbance has abated the Mark* will automatically return the machine to the Pre-Selected Load Control Reference. Paying for either or both of these options is like paying the OEM to fix their problems.... It kind of irks me, personally.
Anyway, if you can provide the accompanying paragraphs describing the graph I would very much appreciate it.
I didn't really respond to all of your questions in Response #25 because I didn't really understand what you were asking/stating, and I think your thoughts were based on an incorrect reading of the graph. If that's true, I'd just as soon forget trying to answer those questions.
When there is a large system/grid with many prime movers and generators all synchronized together to power the system/grid, it's not typical for there to be one machine operating in Isochronous Speed Control. Large systems can have very large grid disturbances which a single machine operating in Isochronous Speed Control cannot respond to, and because there are SO MANY machines operating in Droop Speed Control and all responding to grid frequency disturbances (if they are not operating at rated output or near zero output) that large disturbances can usually be limited and system/grid operators will have a chance to take appropriate action to restore grid frequency to normal. In fact, when a power plant to be synchronized to a system/grid is being built it's very common for the plant owner/operator to be required to tell the system/grid operators what the Droop Regulation of the machines in the plant are, and software used by the grid operators is used to try to predict if there is enough capacity on the system/grid (from machines operating at Part Load, hopefully WITHOUT Pre-Selected Load Control enabled and active!) to respond to anticipated grid disturbances. It's when machines have had their Droop Regulation setpoints changed so that the machines don't respond as expected (or when machines are operated with Pre-Selected Load Control enabled and active!) that grid disruptions can become very problematic. (This was a HUGE problem in Texas a few years ago when it was discovered during Droop Speed Control testing of all power plants that some plants had Droop Regulation setpoints of 7- or 8%, and higher, when they were supposed to have either 4% or 5% Droop Regulation setpoints. That meant they wouldn't change load by the expected amounts during grid frequency disruptions.... Some older analog systems had drifted over time and had not been calibrated in many years; older governors used links and levers which had worn linkages and were misadjusted. Other control systems had been "adjusted" to change loading/unloading rates without understanding knock-on effects.)
You had mentioned modifying load sharing after paralleling (synchronizing) and that's not something you really want to do. OEMs have a LOT of experience with Droop Speed Control and know how their machines should react and respond. And that may be how you were interpreting the graph in the post.
Many people think the way to change the loading/unloading rate of a generator prime mover in a plant (for example, when changing load while operating at Part Load, or when an automatic load change (such as to respond to some alarm condition) is being done, or during normal shutdowns from load) is to change the Droop Regulation setpoint--and that's false and represents a complete misunderstanding of Droop Speed Control. There are usually loading/unloading rates which can be adjusted for this purpose, but people think loading/unloading rates are controlled by the Droop Regulation setpoint--which, again, is incorrect. Many times people say Droop Speed Control is how machines "share load" and controls how much load machines will take as the error between Machine Speed Reference and Machine Actual Speed changes it IS NOT INTENDED to be used to control loading/unloading rates in a plant. Droop Speed Control does allow load sharing--by stably controlling load when the system/grid frequency is stable, so the error between the Machine Speed Reference and the Machine Actual Speed is stable. Droop Speed Control DOES control how much load is added or removed when the error between the Machine Speed Reference and the Machine Actual Speed changes. So, the way most OEMs control loading/unloading rates is to have different rates at which the error between Machine Speed Reference and Machine Actual Speed changes the energy flow-rate into the prime mover, which is different from how much load is changed for a given error between Machine Speed Reference and Machine Actual Speed.
I'm thinking of making some tables to try to explain things a little better, but I have other tasks and responsibilities I need to attend to so it may be a while in the making before I can post them.
If you can post any supporting documentation (paragraphs?) that accompanied the graph in your source, it would be very helpful. I presumed it was trying to show two generators, synchronized together and powering a small system/grid, with both generator prime movers operating in Droop Speed Control with no manual or automatic correction. But, it may have been trying to show two generators, both with 4% Droop Regulation, each one operating independently of the other (not synchronized to the same system/grid. It kind of works both ways.
I want to caution you, and anyone reading this thread--When a GE-design heavy duty gas turbine is operating with BASE LOAD enabled and active if the grid frequency decreases from normal the machine CANNOT INCREASE its load!!! It's already operating at the maximum, optimal power output for the given ambient and machine conditions. AND, IN FACT the load of the machine will DECREASE (it won't maintain the Base Load output) because as the system/grid frequency decreases the speed of the axial compressor decreases which reduces air flow through the machine which causes the exhaust temperature to increase which causes the Mark* to reduce fuel which reduces the load being produced by the machine. This is a dirty little secret of gas turbines--of any manufacturer.
If a machine is operating on Droop Speed Control very near Base Load Rating and the grid frequency decreases the load will increase until the Base Load power output for the given operating conditions is reached and if the grid frequency decreases any further the load will decrease as the axial compressor speed decreases.
Conversely, if a machine is running with BASE LOAD active and the grid frequency increases the power output of the machine will increase--because the axial compressor speed increases which pushes more air through the turbine and exhaust sections of the machine which cools the exhaust temperature which causes the Mark* to increase fuel to maintain the allowable exhaust temperature. This is part of the dirty little secret of gas turbines. GE sells software modifications to allow a machine running with BASE LOAD active to properly respond to grid frequency disturbances, but it has to be purchased and the software modifications installed and tested. GE also sells a software modification which allows the machine to properly respond to grid frequency disturbances while Pre-Selected Load Control is active for as long as the grid frequency is outside of an allowable range, which means the machine load will change appropriately and will not remain stable during a grid frequency disturbance, but once the grid frequency disturbance has abated the Mark* will automatically return the machine to the Pre-Selected Load Control Reference. Paying for either or both of these options is like paying the OEM to fix their problems.... It kind of irks me, personally.
Anyway, if you can provide the accompanying paragraphs describing the graph I would very much appreciate it.
I didn't really respond to all of your questions in Response #25 because I didn't really understand what you were asking/stating, and I think your thoughts were based on an incorrect reading of the graph. If that's true, I'd just as soon forget trying to answer those questions.
When there is a large system/grid with many prime movers and generators all synchronized together to power the system/grid, it's not typical for there to be one machine operating in Isochronous Speed Control. Large systems can have very large grid disturbances which a single machine operating in Isochronous Speed Control cannot respond to, and because there are SO MANY machines operating in Droop Speed Control and all responding to grid frequency disturbances (if they are not operating at rated output or near zero output) that large disturbances can usually be limited and system/grid operators will have a chance to take appropriate action to restore grid frequency to normal. In fact, when a power plant to be synchronized to a system/grid is being built it's very common for the plant owner/operator to be required to tell the system/grid operators what the Droop Regulation of the machines in the plant are, and software used by the grid operators is used to try to predict if there is enough capacity on the system/grid (from machines operating at Part Load, hopefully WITHOUT Pre-Selected Load Control enabled and active!) to respond to anticipated grid disturbances. It's when machines have had their Droop Regulation setpoints changed so that the machines don't respond as expected (or when machines are operated with Pre-Selected Load Control enabled and active!) that grid disruptions can become very problematic. (This was a HUGE problem in Texas a few years ago when it was discovered during Droop Speed Control testing of all power plants that some plants had Droop Regulation setpoints of 7- or 8%, and higher, when they were supposed to have either 4% or 5% Droop Regulation setpoints. That meant they wouldn't change load by the expected amounts during grid frequency disruptions.... Some older analog systems had drifted over time and had not been calibrated in many years; older governors used links and levers which had worn linkages and were misadjusted. Other control systems had been "adjusted" to change loading/unloading rates without understanding knock-on effects.)
You had mentioned modifying load sharing after paralleling (synchronizing) and that's not something you really want to do. OEMs have a LOT of experience with Droop Speed Control and know how their machines should react and respond. And that may be how you were interpreting the graph in the post.
Many people think the way to change the loading/unloading rate of a generator prime mover in a plant (for example, when changing load while operating at Part Load, or when an automatic load change (such as to respond to some alarm condition) is being done, or during normal shutdowns from load) is to change the Droop Regulation setpoint--and that's false and represents a complete misunderstanding of Droop Speed Control. There are usually loading/unloading rates which can be adjusted for this purpose, but people think loading/unloading rates are controlled by the Droop Regulation setpoint--which, again, is incorrect. Many times people say Droop Speed Control is how machines "share load" and controls how much load machines will take as the error between Machine Speed Reference and Machine Actual Speed changes it IS NOT INTENDED to be used to control loading/unloading rates in a plant. Droop Speed Control does allow load sharing--by stably controlling load when the system/grid frequency is stable, so the error between the Machine Speed Reference and the Machine Actual Speed is stable. Droop Speed Control DOES control how much load is added or removed when the error between the Machine Speed Reference and the Machine Actual Speed changes. So, the way most OEMs control loading/unloading rates is to have different rates at which the error between Machine Speed Reference and Machine Actual Speed changes the energy flow-rate into the prime mover, which is different from how much load is changed for a given error between Machine Speed Reference and Machine Actual Speed.
I'm thinking of making some tables to try to explain things a little better, but I have other tasks and responsibilities I need to attend to so it may be a while in the making before I can post them.
I’m confused now. Why should it increase the load? Isn’t the function of BASE LOAD is to maintain the load constant regardless of the frequency changes? If the grid frequency decreases, the fuel that was previously used to maintain the frequency (before it got decreased) will be used to maintain the kW it was supplying.
(or)
Am I confusing it with function of Droop?
Please find the paper in the below link (from where I got the graph);
IEEE Paper
Regarding the loading/unloading rate - It's over my head; will need multiple readings and once I get it, will come back.
What factors influence the exact droop setting? (Usually between 2 and 5%) But which criteria decides exact setting? either 2%, or 3%, or 4%, or 5%?
Selk, Droop Speed Control and Base Load control are two different types of fuel control. While it’s possible to have both at the same time it requires software modifications to the Mark* and careful tracking of over-firing time and magnitude to protect against catastrophic turbine parts failure by planning for more frequent maintenance outages and parts replacement (less time between maintenance outages). And that hasn’t historically been part of GE’s typical turbine control system software.
The purpose of Base Load is to produce as much power as possible for the present machine and ambient conditions. Machine conditions include turbine air inlet filter cleanliness, IGV cleanliness and IGV LVDT calibration, axial compressor cleanliness and clearances, exhaust duct back pressure, etc. Ambient conditions include barometric pressure, axial compressor inlet temperature, relative humidity, etc. All of these conditions affect the internal firing temperature which is what Base Load is trying to control to optimize power output (torque) while protecting machine internals. Base Load is only about putting as much fuel into the machine as possible while protecting the machine from excessive wear and thermal stress. it does this by maximizing fuel flow while maintaining a constant internal firing temperature, not by maintaining a constant load. (When Base Load is active the Load varies throughout the day and week and year as ambient conditions vary.) Nothing more, and nothing less. Frequency control and response are not Base Load’s job.
I thought it was clear that to maintain grid stability during a frequency decrease (which means too much load for the current power generation) that Droop Speed Control would increase load as system/grid frequency decreased. But Base Load doesn’t care about frequency—it only cares about maximizing torque production while protecting machine internals based on operating conditions, that’s all. And when the machine is at Base Load it’s already making as much power as it safely can.
When the speed of the prime mover and generator decreases (because system/grid frequency decreased) while Base Load is active air flow through the machine is reduced (because the axial compressor slows down) that decreases axial compressor discharge pressure AND increases exhaust temperature if fuel flow remained constant. The Mark* senses the change in axial compressor discharge pressure and exhaust temperature and reduces fuel flow to protect the turbine. Base Load doesn’t care about frequency, and Droop Speed Control is out of the control loop when Base Load is active (as defined by the Mark* Minimum Select function for fuel control and protection). The gas turbine could be over-fired to produce more power when system/grid frequency decreases while Base Load is active, but that increases thermal stress on turbine internal components which decreases parts life which can result in serious damage to machine in the worst case, and contributes to increased parts consumption (cost) and decreased time between maintenance outages (which reduces generation revenue and increases maintenance costs).
The purpose of Base Load is to produce as much power as possible for the present machine and ambient conditions. Machine conditions include turbine air inlet filter cleanliness, IGV cleanliness and IGV LVDT calibration, axial compressor cleanliness and clearances, exhaust duct back pressure, etc. Ambient conditions include barometric pressure, axial compressor inlet temperature, relative humidity, etc. All of these conditions affect the internal firing temperature which is what Base Load is trying to control to optimize power output (torque) while protecting machine internals. Base Load is only about putting as much fuel into the machine as possible while protecting the machine from excessive wear and thermal stress. it does this by maximizing fuel flow while maintaining a constant internal firing temperature, not by maintaining a constant load. (When Base Load is active the Load varies throughout the day and week and year as ambient conditions vary.) Nothing more, and nothing less. Frequency control and response are not Base Load’s job.
I thought it was clear that to maintain grid stability during a frequency decrease (which means too much load for the current power generation) that Droop Speed Control would increase load as system/grid frequency decreased. But Base Load doesn’t care about frequency—it only cares about maximizing torque production while protecting machine internals based on operating conditions, that’s all. And when the machine is at Base Load it’s already making as much power as it safely can.
When the speed of the prime mover and generator decreases (because system/grid frequency decreased) while Base Load is active air flow through the machine is reduced (because the axial compressor slows down) that decreases axial compressor discharge pressure AND increases exhaust temperature if fuel flow remained constant. The Mark* senses the change in axial compressor discharge pressure and exhaust temperature and reduces fuel flow to protect the turbine. Base Load doesn’t care about frequency, and Droop Speed Control is out of the control loop when Base Load is active (as defined by the Mark* Minimum Select function for fuel control and protection). The gas turbine could be over-fired to produce more power when system/grid frequency decreases while Base Load is active, but that increases thermal stress on turbine internal components which decreases parts life which can result in serious damage to machine in the worst case, and contributes to increased parts consumption (cost) and decreased time between maintenance outages (which reduces generation revenue and increases maintenance costs).
Yes, but the droop speed control would increase the load only if the fuel is raised manually/Machine speed reference is changed or through some type of frequency control. Am I correct?
No; Selk. It's completely automatic because as the Machine Actual Speed changes (due to the frequency disturbance) the ERROR between the Machine Speed Reference and the Machine Actual Speed changes, which changes the load being produced by the machine. The same as if an operator changed the Machine Speed Reference while watching the MW meter while the Machine Actual Speed is stable (and hopefully at rated); the ERROR changes and the fuel automatically changes. Or, if some DCS sends a signal to RAISE or LOWER load, it's actually changing the Machine Speed Reference, and that change causes the ERROR to change which changes fuel flow-rate.
It's all about the ERROR (between the Machine Speed Reference and the Machine Actual Speed; in the Mark* case, between TNR and TNH). When the ERROR changes for any reason, it will cause the fuel flow-rate to the turbine to change (or the energy flow-rate into the prime mover).
And that's what the system/grid operators/regulators want to happen--automatic response to help support grid stability while they are working on restoring system/grid frequency to normal.
It's all about the ERROR (between the Machine Speed Reference and the Machine Actual Speed; in the Mark* case, between TNR and TNH). When the ERROR changes for any reason, it will cause the fuel flow-rate to the turbine to change (or the energy flow-rate into the prime mover).
And that's what the system/grid operators/regulators want to happen--automatic response to help support grid stability while they are working on restoring system/grid frequency to normal.
I think I've fixed all the typos and poor grammar in this post; apologies if I missed anything.
While the IEEE paper has some very good information and clarification I have a difficult time with some of the terminology--and using a No-Load Frequency to determine Droop Regulation, which is typical of many Woodward Governor generator prime mover control systems. GE's formula is very simple: FSRN=((TNR-TNH)*Gain)+Offset (y=m(x2-x1)+b--the typical expression for the formula for a straight line where the x-axis value is the difference between two variables instead of one, in this case two speed-related variables).
I'm going to violate something that previous posters (MARKVGUY and CSA, as well as others) never did: I'm going to show how GE incorporates Droop Regulation setpoint into the Mark* fuel control of GE-design heavy duty gas turbines without ever displaying the Droop Regulation value (so it can't be easily changed).* All of this can be very easily determined from the Expected Fuel Characteristic paragraphs of the Control Specification document which is (or at least it was) supplied with every Mark* turbine control system provided with GE-design heavy duty gas turbines. I'm not going to caution anyone about changing the Droop Regulation setpoint without VERY SERIOUS consideration of the possible effects of the change; it should be obvious to everyone. If one doesn't know and understand Droop Speed Control and how it impacts MANY aspects of GE-design heavy duty gas turbine control philosophies and schemes AND how the system/grid regulators might react to learning of the change to the Droop Regulation setpoint of a machine synchronized to their system grid, well, even if one knows how to change the Droop Regulation setpoint it should be left to others to decide IF it should be done or not, and who needs to know if it is changed.
FSRN is Droop Speed Control FSR (Fuel Stroke Reference--which is the calculated amount of fuel flow into the turbine combustors (for gas fuel it's fuel control valve position; for liquid fuel it's liquid fuel flow-rate)
TNR is the Turbine Speed Reference, in percent (Machine Speed Reference)
TNH is actual turbine speed, in percent (Machine Actual Speed)
Gain in percent-per-percent, (calculated by dividing the (calculated fuel flow-rate at Base Load FSR minus the calculated Full Speed-No Load FSR) by the Droop Regulation setpoint for the machine (which usually isn't expressed in any value in the Mark* control system--because too many people would be incorrectly changing it if it were expressly specified and easily visible without understanding what it does and what the knock-on effects would be))
Offset is calculated Full Speed-No Load FSR. Expected fuel flow-rates are calculated when the machine is being built by using the expected fuel characteristics of the fuel(s) available at the site where the turbines will be operating (heating values; methane percentage; etc.) supplied by the purchaser. Typical parameters/values would be as follows:
Calculated Full Speed-No Load FSR: 20%
Calculated Base Load FSR: 70% (This is the calculated fuel flow at Base Load)
Gain: This is the calculated amount of change of fuel flow per percent of Droop Speed Regulation, or the change in Base Load fuel flow minus 0 load fuel flow (FSNL FSR); remember, there is a certain amount of fuel flow required to maintain rated speed (frequency); it's specific to just about every machine; in this example the Droop Regulation is 4%, as it is for most heavy duty gas turbines of any manufacturer
Offset: 20% (FSNL FSR; the amount of fuel flow-rate required to maintain rated speed (100% actual speed (frequency))
Gain = (70-20)/4=50/4=12.5% FSR/%
We (should) know that for a machine with 4% Droop Regulation 50% of rated load will occur at a Machine Speed Reference of 102% (half of 4% plus rated speed, ((4/2)+100)=102%). Plugging this into the Mark* equation for Droop Speed Control FSR (fuel flow reference, which is the energy flow-rate into the turbine!) for a machine operating at rated speed/frequency while producing power we get:
FSRN =((102%-100)*12.5% FSR/% Droop Regulation)+20%=((2%)*12.5%/%)+20%=25% FSR = 20% FSR=45% FSR.
So, for a machine in a new and clean condition (clean IGVs; fairly clean axial compressor; good internal clearances and parts condition), with fairly clean turbine inlet air filters and a normal exhaust duct back-pressure operating at near turbine nameplate ambient conditions AND burning the fuel which was used to calculate the fuel control parameters and size fuel control valves and fuel nozzles the Droop Speed Control FSR (FSRN) would be equal to approximately 50% of rated. For a turbine rated at 25 MW, that would be approximately 12.5 MW; for a machine rated at 38 MW that be approximately 19 MW. (It should be noted that fuel make-up can vary over time as new sources are brought on-line and old sources are taken off-line.)
What the formula and the Gain value say is: For every 1% change in the error between the Turbine Speed Reference (Machine Speed Reference) and the Turbine Speed (Machine Actual Speed) the calculated fuel flow-rate will change by 12.5%. This means that as long as the machine actual speed is at or near rated and stable that the actual fuel flow-rate will also be stable at any Machine Speed Reference (TNR in this case). It ALSO means that if the machine speed reference is stable (which it should be if the load is stable) and the machine actual speed changes (because the system/grid frequency changes) the fuel flow-rate into the machine will change in an effort to support grid stability. System/grid operators/regulators need to be able to count on machines behaving in a certain manner in the event of frequency disruptions--and Droop Regulation setpoints provide that method by knowing how much a machine's load will change for a given change in system/grid frequency.
The way that the IEEE paper explains it works, too; though, again, I have a very difficult time with some of the terminology. (And I'm sure many people have a difficult time with GE's terminology, which I am familiar with and adopted; it's all really the same thing.) AND, representing Droop Speed Control by implying that machine speed changes as load changes is, well, not correct. For G1, 62.4 Hz "No Load Frequency" is 104% of 60 Hz--which is 4% Droop Regulation. And, for G2 61.2 Hz divided by 58.8 Hz is also 104%--which is also 4% Droop Regulation. But this statement sums it up for me: "Operating with a no-load frequency of 62.4 Hz may be objectionable to operations because at no-load traditional turbine over-frequency alarm or trip limits (105% x rated frequency) may be marginal." NO machine synchronized with other machines on a system/grid (even just one machine) EVER runs at 62.4 Hz (104% of rated speed) unloaded and prior to synchronization or even immediately after synchronization with no load. Not to produce 60 Hz at No Load it doesn't. That might be the machine's speed (frequency) reference but if it were the control system would have to have a negative Gain value (for the slope to be negative as shown) and to arrive at rated output and a machine reference of 100% (a difference of 104%-100%).
The authors DID actually state the operating conditions for this graph better than any other reference I've ever read--but it's still misleading to a lot of people who think that because machine speed changes with changes in fuel flow-rate during start-up and shutdown that machine speed will change when the machine is being loaded and unloaded while producing electrical power. Synchronous generators synchronized to a well-regulated system/grid with other synchronous generators (and their prime movers) simply do not change actual speed as the load they are supplying changes. It just doesn't happen. And expressing Droop Speed Control as if machine actual speed does change just adds to the mystery and mythology and wives' tales about Droop Speed Control. It just doesn't happen. It's wrong. Full stop. Period. F=(P*N)120. Full Stop. Period. You want 50- or 60 Hz coming out of the wall socket, EVERY machine (synchronous machines, that is) MUST operate at the same (synchronous) frequency, which is a function of generator rotor speed AND the number of poles of the generator rotor (synchronous speed).
It might have been easier to explain and understand Droop Speed Control if Droop Regulation was expressed as a value from 0% to 100% of expected energy flow-rate into the prime mover from No Load to Full Load? It could have been done in several ways, but using No Load Frequency versus rated frequency (without explaining no machine ever actually runs "off-frequency" except during grid frequency disturbances) isn't very good, either.
* So, the formula above is GE's tried and true (and proven!) Droop Speed Control formula used for decades, until the advent of DLN (Dry Low NOx) combustion systems. With the advent of DLN they modified their Droop Speed Control formula to be called Constant Settable Droop (which is, as others have said here on Control.com) is a very poor name for the function it performs). VERY poor. Anyway, even though the formula has been modified to used derivative control and incorporate load into the formula, this Constant Settable Droop Speed Control formula still functions in a very basic manner just like their older Droop Speed Control formula. Some older machines with conventional combustion systems and newer Mark* control systems still use the original Droop Speed Control.
While the IEEE paper has some very good information and clarification I have a difficult time with some of the terminology--and using a No-Load Frequency to determine Droop Regulation, which is typical of many Woodward Governor generator prime mover control systems. GE's formula is very simple: FSRN=((TNR-TNH)*Gain)+Offset (y=m(x2-x1)+b--the typical expression for the formula for a straight line where the x-axis value is the difference between two variables instead of one, in this case two speed-related variables).
I'm going to violate something that previous posters (MARKVGUY and CSA, as well as others) never did: I'm going to show how GE incorporates Droop Regulation setpoint into the Mark* fuel control of GE-design heavy duty gas turbines without ever displaying the Droop Regulation value (so it can't be easily changed).* All of this can be very easily determined from the Expected Fuel Characteristic paragraphs of the Control Specification document which is (or at least it was) supplied with every Mark* turbine control system provided with GE-design heavy duty gas turbines. I'm not going to caution anyone about changing the Droop Regulation setpoint without VERY SERIOUS consideration of the possible effects of the change; it should be obvious to everyone. If one doesn't know and understand Droop Speed Control and how it impacts MANY aspects of GE-design heavy duty gas turbine control philosophies and schemes AND how the system/grid regulators might react to learning of the change to the Droop Regulation setpoint of a machine synchronized to their system grid, well, even if one knows how to change the Droop Regulation setpoint it should be left to others to decide IF it should be done or not, and who needs to know if it is changed.
FSRN is Droop Speed Control FSR (Fuel Stroke Reference--which is the calculated amount of fuel flow into the turbine combustors (for gas fuel it's fuel control valve position; for liquid fuel it's liquid fuel flow-rate)
TNR is the Turbine Speed Reference, in percent (Machine Speed Reference)
TNH is actual turbine speed, in percent (Machine Actual Speed)
Gain in percent-per-percent, (calculated by dividing the (calculated fuel flow-rate at Base Load FSR minus the calculated Full Speed-No Load FSR) by the Droop Regulation setpoint for the machine (which usually isn't expressed in any value in the Mark* control system--because too many people would be incorrectly changing it if it were expressly specified and easily visible without understanding what it does and what the knock-on effects would be))
Offset is calculated Full Speed-No Load FSR. Expected fuel flow-rates are calculated when the machine is being built by using the expected fuel characteristics of the fuel(s) available at the site where the turbines will be operating (heating values; methane percentage; etc.) supplied by the purchaser. Typical parameters/values would be as follows:
Calculated Full Speed-No Load FSR: 20%
Calculated Base Load FSR: 70% (This is the calculated fuel flow at Base Load)
Gain: This is the calculated amount of change of fuel flow per percent of Droop Speed Regulation, or the change in Base Load fuel flow minus 0 load fuel flow (FSNL FSR); remember, there is a certain amount of fuel flow required to maintain rated speed (frequency); it's specific to just about every machine; in this example the Droop Regulation is 4%, as it is for most heavy duty gas turbines of any manufacturer
Offset: 20% (FSNL FSR; the amount of fuel flow-rate required to maintain rated speed (100% actual speed (frequency))
Gain = (70-20)/4=50/4=12.5% FSR/%
We (should) know that for a machine with 4% Droop Regulation 50% of rated load will occur at a Machine Speed Reference of 102% (half of 4% plus rated speed, ((4/2)+100)=102%). Plugging this into the Mark* equation for Droop Speed Control FSR (fuel flow reference, which is the energy flow-rate into the turbine!) for a machine operating at rated speed/frequency while producing power we get:
FSRN =((102%-100)*12.5% FSR/% Droop Regulation)+20%=((2%)*12.5%/%)+20%=25% FSR = 20% FSR=45% FSR.
So, for a machine in a new and clean condition (clean IGVs; fairly clean axial compressor; good internal clearances and parts condition), with fairly clean turbine inlet air filters and a normal exhaust duct back-pressure operating at near turbine nameplate ambient conditions AND burning the fuel which was used to calculate the fuel control parameters and size fuel control valves and fuel nozzles the Droop Speed Control FSR (FSRN) would be equal to approximately 50% of rated. For a turbine rated at 25 MW, that would be approximately 12.5 MW; for a machine rated at 38 MW that be approximately 19 MW. (It should be noted that fuel make-up can vary over time as new sources are brought on-line and old sources are taken off-line.)
What the formula and the Gain value say is: For every 1% change in the error between the Turbine Speed Reference (Machine Speed Reference) and the Turbine Speed (Machine Actual Speed) the calculated fuel flow-rate will change by 12.5%. This means that as long as the machine actual speed is at or near rated and stable that the actual fuel flow-rate will also be stable at any Machine Speed Reference (TNR in this case). It ALSO means that if the machine speed reference is stable (which it should be if the load is stable) and the machine actual speed changes (because the system/grid frequency changes) the fuel flow-rate into the machine will change in an effort to support grid stability. System/grid operators/regulators need to be able to count on machines behaving in a certain manner in the event of frequency disruptions--and Droop Regulation setpoints provide that method by knowing how much a machine's load will change for a given change in system/grid frequency.
The way that the IEEE paper explains it works, too; though, again, I have a very difficult time with some of the terminology. (And I'm sure many people have a difficult time with GE's terminology, which I am familiar with and adopted; it's all really the same thing.) AND, representing Droop Speed Control by implying that machine speed changes as load changes is, well, not correct. For G1, 62.4 Hz "No Load Frequency" is 104% of 60 Hz--which is 4% Droop Regulation. And, for G2 61.2 Hz divided by 58.8 Hz is also 104%--which is also 4% Droop Regulation. But this statement sums it up for me: "Operating with a no-load frequency of 62.4 Hz may be objectionable to operations because at no-load traditional turbine over-frequency alarm or trip limits (105% x rated frequency) may be marginal." NO machine synchronized with other machines on a system/grid (even just one machine) EVER runs at 62.4 Hz (104% of rated speed) unloaded and prior to synchronization or even immediately after synchronization with no load. Not to produce 60 Hz at No Load it doesn't. That might be the machine's speed (frequency) reference but if it were the control system would have to have a negative Gain value (for the slope to be negative as shown) and to arrive at rated output and a machine reference of 100% (a difference of 104%-100%).
The authors DID actually state the operating conditions for this graph better than any other reference I've ever read--but it's still misleading to a lot of people who think that because machine speed changes with changes in fuel flow-rate during start-up and shutdown that machine speed will change when the machine is being loaded and unloaded while producing electrical power. Synchronous generators synchronized to a well-regulated system/grid with other synchronous generators (and their prime movers) simply do not change actual speed as the load they are supplying changes. It just doesn't happen. And expressing Droop Speed Control as if machine actual speed does change just adds to the mystery and mythology and wives' tales about Droop Speed Control. It just doesn't happen. It's wrong. Full stop. Period. F=(P*N)120. Full Stop. Period. You want 50- or 60 Hz coming out of the wall socket, EVERY machine (synchronous machines, that is) MUST operate at the same (synchronous) frequency, which is a function of generator rotor speed AND the number of poles of the generator rotor (synchronous speed).
It might have been easier to explain and understand Droop Speed Control if Droop Regulation was expressed as a value from 0% to 100% of expected energy flow-rate into the prime mover from No Load to Full Load? It could have been done in several ways, but using No Load Frequency versus rated frequency (without explaining no machine ever actually runs "off-frequency" except during grid frequency disturbances) isn't very good, either.
* So, the formula above is GE's tried and true (and proven!) Droop Speed Control formula used for decades, until the advent of DLN (Dry Low NOx) combustion systems. With the advent of DLN they modified their Droop Speed Control formula to be called Constant Settable Droop (which is, as others have said here on Control.com) is a very poor name for the function it performs). VERY poor. Anyway, even though the formula has been modified to used derivative control and incorporate load into the formula, this Constant Settable Droop Speed Control formula still functions in a very basic manner just like their older Droop Speed Control formula. Some older machines with conventional combustion systems and newer Mark* control systems still use the original Droop Speed Control.
My confusion stems from the general view that isochronous machine acts as swing bus and the droop machines would supply only the fixed loads.
All the droop machines would supply constant (fixed) load only if the bus frequency/grid frequency is constant.
In case of one machine in Isochronous and the other machines in Droop, the isochronous machine automatically takes any variations in the load as it has to maintain the set speed and hence the droop machines
are forced to maintain the constant load. If the isochronous machine could not maintain the speed, the droop generator takes the load, according to the droop setting.
Droop function acts --> when TNH is varied
To raiser/lower manually --> on varying this TNR is varied
It seems I have got it.
When is it preferred to have one in Isochronous and others Generators in droop?
When is it preferred to have all the generators in droop mode?
All the droop machines would supply constant (fixed) load only if the bus frequency/grid frequency is constant.
In case of one machine in Isochronous and the other machines in Droop, the isochronous machine automatically takes any variations in the load as it has to maintain the set speed and hence the droop machines
are forced to maintain the constant load. If the isochronous machine could not maintain the speed, the droop generator takes the load, according to the droop setting.
Droop function acts --> when TNH is varied
To raiser/lower manually --> on varying this TNR is varied
It seems I have got it.
When is it preferred to have one in Isochronous and others Generators in droop?
When is it preferred to have all the generators in droop mode?
You're getting the concept now.
I'm a stickler for correct terminology. The Droop Machines are--when the system/grid frequency is stable--going to produce however much power they are set to produce (and I FERVENTLY HOPE that's not by using Pre-Selected Load Control!!!) regardless of what the Isoch machine does, or what's happening on the utility system/grid if the plant is connected to (synchronized with) the utility system/grid. I wouldn't call that "fixed load(s)", but that's just me. The Droop machines are going to produce whatever power they are set to produce regardless of what the Isoch machine is doing as long as the system/grid frequency is at rated (TNH is at or very near 100%). That's what "LOAD SHARING" means--the Droop machines aren't going to "fight" for load to try to control frequency--they are going to "play nicely" with the other Droop Machines--and even with the Isoch machine, because if the Droop machines are constantly changing their power output then the Isoch machine is going to not be working so well as it tries to respond to system/grid load changes (frequency changes) as well as load changes which might be caused by "unruly" Droop machines. That's what "load sharing" really means--stable operation when system/grid frequency is stable, not uncoordinated and random load changes. Operating machines in Droop Speed Control provides stable operation as long as system/grid frequency is stable--which is immensely important to a well-regulated and stable system/grid of any size. And, Droop Speed Control makes this possible.
Just remember: The Isochronous machine can't provide power above its prime mover rating/machine/ambient conditions, nor below 0 MW (it will trip on reverse power). EITHER a trained human operator OR a well-programmed power/frequency management system has to be constantly watching the load on the Isoch machine and trying to anticipate any large sysytem/grid load changes and take pre-emptive action. One CANNOT change the load on the Isoch machine's prime mover governor (one only changes the frequency setpoint--which you want to stay at 50 Hz or 60 Hz). To change load on an Isoch machine when synchronized with other machines (all operating in Droop Speed Control mode!) one changes the load on one or more Droop machines which will cause the Isoch machine to compensate in the opposite direction to maintain system/grid frequency. This is where a LOT of plants and power/frequency management systems fail, often failing miserably by expecting the Isoch machine to (magically) handle any and all load changes without any intervention (human or automation), and that's just not a realistic idea or expectation. They complain, "The Isoch machine isn't working correctly!"--even though it works perfectly for most all load changes between 0 MW and maximum power output (they (conveniently) forget that part in this circumstance, too!).
"When is it preferred to have one in Isochronous and others Generators in droop?"
ANY time the plant is not connected to the utility system/grid; so, when it's running in island mode. If the plant is connected to the utility system/grid and one machine is in Isoch mode it will try to respond to any utility system/grid changes and that can be VERY freaky and scary (again, people will say, "The Isoch machine isn't working correctly!''--again (conveniently) forgetting it works fine when the plant is separated from the utility system/grid AND the load is kept away from 0 MW and not at or near rated power output).
"When is it preferred to have all the generators in droop mode?"
When the plant is connected to the utility system/grid, even if the plant is not importing or exporting power all machines running and producing power should have their prime mover governors operating in Droop Speed Control mode. If it's synchronized to the utility system/grid all the generator prime mover governors should be operating in Droop Speed Control mode. If the plant wants to run when it gets separated from the utility system/grid for any reason, there must be a status contact (usually) to indicate the utility tie breaker(s) are open to switch one machine to Isochronous Speed Control mode AND make any adjustmemts to the Droop machines to be sure the load on the Isoch machine is within the operating range of the Isoch machine (so not at or near 0 MW and not at or near rated power output.) (This is where a well-programmed power/frequency management system can be very helpful. But, it requires lots of signals to know what the total plant load is, how many machines are running, how much load will have to be adjusted on the Droop Machines, which machine to switch to Isoch, and what to do if the "primary" Isoch machine isn't available.... It's quite a matrix that requires lots of signals (circuits; wiring) to work correctly. AND, to transition back to utility system/grid operation, too!) Typically making this kind of adjustment on separation from the utility (if it was exporting power) is called "load shedding." If the machine was importing power at the time of separation from the utility then the Droop machines and the Isoch machines have to be adjusted properly to maintain the island load. And, again, this is another area where MANY power/frequency management systems fail, sometimes miserably. And, sometimes--depending on the loads and number of machines operating in the plant when it gets separated from the utility--it's just not possible to have a smooth transition to island mode; there's going to be some instability, hopefully it's just not very large and it doesn't last very long. (Another unrealistic expectation--that EVERY separation from the utility to island mode will be smooth and uneventful. Sometimes, it's just not going to be smooth, and the plant and its island system/grid is going to go black.)
Again, such a power/frequency management system for a plant with a large load to supply with multiple prime movers and generators is going to require an overall power system study and coordination of tripping schemes. It's a very large undertaking. Again, I highly recommend training the current plant operators and their supervisors, as well as the "customer" representatives who will be receiving power when islanding occurs, to know what's supposed to happen and when, and how to accomplish manually handling a utility system/grid separation. Only in this way will the testing of any power/management system be successful, AND the operators will also be able to identify and communicate any problems--during commissioning testing and afterwards. It's a very good investment, training operators. It brings everyone to a common level of understanding and they can even help each other with their understanding and operations. If trained operators leave for any reason new operators will benefit from the experience of those who have been through training, and if there are training manuals available for the new operators that will help, also. Having a total plant single-line diagram in the plant Control Room, and giving a copy to every operator and operations supervisor is also a very good thing. (My background from having work for a short time on ocean-going ships taught me that the more knowledge and familiarization EVERYONE ON BOARD has the safer everyone on board is. That doesn't seem to be the thought process in very many shoreside power plants, unfortunately. But, it should be. They are dangerous places to work, and only getting on-the-job (OJT) training sometimes isn't the best because many operators and their supervisors just aren't good teachers--even if they know their job and responsibilities very well; it's just a fact of life. If the answer to any question in a power plant is, "Because we've always done it that way!" that's not a good omen. (Of course, sometimes there's a right time and a wrong time to ask a question or questions, but answering 'because we've always done it that way' is totally inappropriate and can lead to lots of misunderstandings and errors.)
I'm a stickler for correct terminology. The Droop Machines are--when the system/grid frequency is stable--going to produce however much power they are set to produce (and I FERVENTLY HOPE that's not by using Pre-Selected Load Control!!!) regardless of what the Isoch machine does, or what's happening on the utility system/grid if the plant is connected to (synchronized with) the utility system/grid. I wouldn't call that "fixed load(s)", but that's just me. The Droop machines are going to produce whatever power they are set to produce regardless of what the Isoch machine is doing as long as the system/grid frequency is at rated (TNH is at or very near 100%). That's what "LOAD SHARING" means--the Droop machines aren't going to "fight" for load to try to control frequency--they are going to "play nicely" with the other Droop Machines--and even with the Isoch machine, because if the Droop machines are constantly changing their power output then the Isoch machine is going to not be working so well as it tries to respond to system/grid load changes (frequency changes) as well as load changes which might be caused by "unruly" Droop machines. That's what "load sharing" really means--stable operation when system/grid frequency is stable, not uncoordinated and random load changes. Operating machines in Droop Speed Control provides stable operation as long as system/grid frequency is stable--which is immensely important to a well-regulated and stable system/grid of any size. And, Droop Speed Control makes this possible.
Just remember: The Isochronous machine can't provide power above its prime mover rating/machine/ambient conditions, nor below 0 MW (it will trip on reverse power). EITHER a trained human operator OR a well-programmed power/frequency management system has to be constantly watching the load on the Isoch machine and trying to anticipate any large sysytem/grid load changes and take pre-emptive action. One CANNOT change the load on the Isoch machine's prime mover governor (one only changes the frequency setpoint--which you want to stay at 50 Hz or 60 Hz). To change load on an Isoch machine when synchronized with other machines (all operating in Droop Speed Control mode!) one changes the load on one or more Droop machines which will cause the Isoch machine to compensate in the opposite direction to maintain system/grid frequency. This is where a LOT of plants and power/frequency management systems fail, often failing miserably by expecting the Isoch machine to (magically) handle any and all load changes without any intervention (human or automation), and that's just not a realistic idea or expectation. They complain, "The Isoch machine isn't working correctly!"--even though it works perfectly for most all load changes between 0 MW and maximum power output (they (conveniently) forget that part in this circumstance, too!).
"When is it preferred to have one in Isochronous and others Generators in droop?"
ANY time the plant is not connected to the utility system/grid; so, when it's running in island mode. If the plant is connected to the utility system/grid and one machine is in Isoch mode it will try to respond to any utility system/grid changes and that can be VERY freaky and scary (again, people will say, "The Isoch machine isn't working correctly!''--again (conveniently) forgetting it works fine when the plant is separated from the utility system/grid AND the load is kept away from 0 MW and not at or near rated power output).
"When is it preferred to have all the generators in droop mode?"
When the plant is connected to the utility system/grid, even if the plant is not importing or exporting power all machines running and producing power should have their prime mover governors operating in Droop Speed Control mode. If it's synchronized to the utility system/grid all the generator prime mover governors should be operating in Droop Speed Control mode. If the plant wants to run when it gets separated from the utility system/grid for any reason, there must be a status contact (usually) to indicate the utility tie breaker(s) are open to switch one machine to Isochronous Speed Control mode AND make any adjustmemts to the Droop machines to be sure the load on the Isoch machine is within the operating range of the Isoch machine (so not at or near 0 MW and not at or near rated power output.) (This is where a well-programmed power/frequency management system can be very helpful. But, it requires lots of signals to know what the total plant load is, how many machines are running, how much load will have to be adjusted on the Droop Machines, which machine to switch to Isoch, and what to do if the "primary" Isoch machine isn't available.... It's quite a matrix that requires lots of signals (circuits; wiring) to work correctly. AND, to transition back to utility system/grid operation, too!) Typically making this kind of adjustment on separation from the utility (if it was exporting power) is called "load shedding." If the machine was importing power at the time of separation from the utility then the Droop machines and the Isoch machines have to be adjusted properly to maintain the island load. And, again, this is another area where MANY power/frequency management systems fail, sometimes miserably. And, sometimes--depending on the loads and number of machines operating in the plant when it gets separated from the utility--it's just not possible to have a smooth transition to island mode; there's going to be some instability, hopefully it's just not very large and it doesn't last very long. (Another unrealistic expectation--that EVERY separation from the utility to island mode will be smooth and uneventful. Sometimes, it's just not going to be smooth, and the plant and its island system/grid is going to go black.)
Again, such a power/frequency management system for a plant with a large load to supply with multiple prime movers and generators is going to require an overall power system study and coordination of tripping schemes. It's a very large undertaking. Again, I highly recommend training the current plant operators and their supervisors, as well as the "customer" representatives who will be receiving power when islanding occurs, to know what's supposed to happen and when, and how to accomplish manually handling a utility system/grid separation. Only in this way will the testing of any power/management system be successful, AND the operators will also be able to identify and communicate any problems--during commissioning testing and afterwards. It's a very good investment, training operators. It brings everyone to a common level of understanding and they can even help each other with their understanding and operations. If trained operators leave for any reason new operators will benefit from the experience of those who have been through training, and if there are training manuals available for the new operators that will help, also. Having a total plant single-line diagram in the plant Control Room, and giving a copy to every operator and operations supervisor is also a very good thing. (My background from having work for a short time on ocean-going ships taught me that the more knowledge and familiarization EVERYONE ON BOARD has the safer everyone on board is. That doesn't seem to be the thought process in very many shoreside power plants, unfortunately. But, it should be. They are dangerous places to work, and only getting on-the-job (OJT) training sometimes isn't the best because many operators and their supervisors just aren't good teachers--even if they know their job and responsibilities very well; it's just a fact of life. If the answer to any question in a power plant is, "Because we've always done it that way!" that's not a good omen. (Of course, sometimes there's a right time and a wrong time to ask a question or questions, but answering 'because we've always done it that way' is totally inappropriate and can lead to lots of misunderstandings and errors.)
I really hope it's becoming, or will become, clear that Droop Speed Control is actually a very ingenious control method which has a very large number of positive benefits for many aspects of AC power generation and transmission. It is actually very simple once one understands what it's primary purpose is: To define how much the power output of a machine will change for a change in the ERROR between the machine speed reference (TNR) and the machine actual speed (TNH). It provides operating stability for systems/grids with many (sometimes hundreds) of generators and their prime movers synchronized together to power a load (loadS, plural, actually) larger than any single generator and its prime mover could ever hope to provide because of "load sharing" ability. Additionally, it can immediately (and automatically) respond to system/grid frequency disturbances and help support the entire system/grid until such time as the system/grid operators can take appropriate action to return the system/grid frequency to normal (usually by adjusting the machine speed references of one or more machines, or bringing more generators and their prime movers on line, or temporarily blacking out "blocks" of loads (homes and businesses) to prevent blacking out the whole system/grid until such time as sufficient generation can be brought on line, or by reducing the amount of generation as necessary to bring down the system frequency to return it to normal. Droop does this for the ENTIRE system/grid--because these machines are all synchronized together and all experience the same changes in system/grid frequency, without any other communication method (analog signals; Ethernet signals; SCADA systems; etc.); it's all done because the governors are monitoring speed which is directly related to frequency which is directly related to load. All with one little formula, using two variables, a gain and an offset. (Because no matter how Droop Speed Control is implemented and any prime mover governor has to respond in a similar fashion, as straight proportional control--y=m(x2-x1)+b (which is basically the formula for a straight line, which represents how Droop Speed Control works).)
If you haven't already, start with the primary concept: The amount of change in load (energy flow-rate into the prime mover, which is directly related to the amount of energy produced by the synchronous generator the prime mover is driving) for a given change in the ERROR between the machine's speed reference and the machine's actual speed (even if you use frequency instead of speed--because, after all, speed and frequency are directly related). A change in EITHER the machine speed reference OR the machine actual speed is going to produce the same change in energy flow-rate into the generator's prime mover. The normal method of loading and unloading a synchronous generator is to increase or decrease the machine's speed reference--even when watching the wattmeter, or the MW meter, because that's what's happening in the background. (Even Pre-Selected Load Control ultimately changes the machine speed reference, TNR!)
I still marvel at this creation--which happened before the advent of even analog electronics! Those were some extremely intelligent people back then, and the concept still works today (well, because it's still used today--but it also still works!). Simply amazing. If you ask me. Complicated? Not really, if it's explained correctly. (Sorry; I'm not good with graphics, so I'm confined to using words, and a little bit of simple mathematics.) There multiple ways Droop Speed Control works--and I think that's what confuses so many people, but it does it with the simple act of "defining" the change in energy flow-rate for a given change in an ERROR signal between two simple variables. Everything else is "icing on the cake" as it were. If the concept were developed today, it would be hopelessly complicated and obtuse. But, it was developed in a time when things were--and so had to be--simple.
Very.
Simply.
Amazing.
If you haven't already, start with the primary concept: The amount of change in load (energy flow-rate into the prime mover, which is directly related to the amount of energy produced by the synchronous generator the prime mover is driving) for a given change in the ERROR between the machine's speed reference and the machine's actual speed (even if you use frequency instead of speed--because, after all, speed and frequency are directly related). A change in EITHER the machine speed reference OR the machine actual speed is going to produce the same change in energy flow-rate into the generator's prime mover. The normal method of loading and unloading a synchronous generator is to increase or decrease the machine's speed reference--even when watching the wattmeter, or the MW meter, because that's what's happening in the background. (Even Pre-Selected Load Control ultimately changes the machine speed reference, TNR!)
I still marvel at this creation--which happened before the advent of even analog electronics! Those were some extremely intelligent people back then, and the concept still works today (well, because it's still used today--but it also still works!). Simply amazing. If you ask me. Complicated? Not really, if it's explained correctly. (Sorry; I'm not good with graphics, so I'm confined to using words, and a little bit of simple mathematics.) There multiple ways Droop Speed Control works--and I think that's what confuses so many people, but it does it with the simple act of "defining" the change in energy flow-rate for a given change in an ERROR signal between two simple variables. Everything else is "icing on the cake" as it were. If the concept were developed today, it would be hopelessly complicated and obtuse. But, it was developed in a time when things were--and so had to be--simple.
Very.
Simply.
Amazing.
| Thread starter | Similar threads | Forum | Replies | Date |
|---|---|---|---|---|
|
Ship Generator uneven load | Power Generation | 25 | |
| D | [MATLAB] load sharing not equivalent between hybrid generator | Power Generation | 0 | |
| R | Generator Load Sharing Issue | Power Generation | 2 | |
| R | Load sharing and generator ramps | Power Generation | 0 | |
| C | Generator load sharing methods | General Automation Chat | 5 |
