Turbine Generator Control 4

[121202]

How is the output power of a turbine generator controlled?

Please explain the entire system including reference to

- Control of the gas generator and power turbine

- The control of the fuel\heat rate

- How the electrical output of the generator is controlled

My understanding is pretty basic so please make any explanations explicit.

 

[CESSNA1]

121202:

The output power is dependent on the load. The power turbine is set to a given speed, say 3600 RPM. If additional load is added to the generator then the power turbine speed will drop. This speed drop will be sensed by the power turbine governor. This governor then sends a signal to the gas generator fuel control to increase the fuel feed to the gas generator. The oppposite happens in a speed increase of the power turbine. In essence the gas generator is slaved to the power turbine. The turbine/compressor speed is allowed to vary (within limits) to provide the power turbine with the energy to meet the load.

The generator load depends on what it is connected to. If it is conencted to lights then as more lights are turned on more load is put on the generator. The generator speed drops and the governor calls for more fuel to the gas generator.

Regards
Dave

 

[zeusfaber]

I like CESSNA1's description of life on the machanical side.

On the electrical side (Following all assumes AC output):

You have two variables you can control yourself
1. Torque input to the alternator (seen further back into the system, position of the steam valve, or fuel input if a gas turbine).

2. Excitation (an electrical variable - the amount of DC fed through the alternator rotor to magnetise it).

What happens depends a bit on the sort of load you're connected to.

If you're the only show in town, or are the dominant generator in a network, winding up the torque changes your output power and frequency. Winding back the excitation reduces output voltage (and hence power). The usual control strategy is to regulate voltage by changing the excitation, and using a governor to regulate output frequency (by altering the input torque in response to changing power demand).

Things change if you're a small part of a synchronised network of generating units. In this scenario, frequency and voltage are driven by the network.

Power output still depends on input torque, but simple speed governing is no longer enough to control it. Once synchronised, the alternator is going to rattle round at synchronous speed however hard, soft, or not at all it's driven. To push power out, you have to advance the phase of your machine relative to the waveform on the network - increasing the Load Angle (achieved by appling more torque at the shaft). This is a bit of an artform, since beyond a certain point, increasing Load Angle reduces torque/power and you enter a statically unstable regime (you get pole slipping, massive power, torque and adrenaline transients, and associated grief from the boss). The system frequency depends on the balance between power being put onto the network and power being drawn out. Network stability is a big issue when you start synchronising generating units.

Changing the excitation has a rather counterintuitive effect. Instead of altering the generated voltage, the main thing it changes is the phase of the generated current waveform - the power factor of the machine. If you over-excite (generate Reactive Power), the machine acts like a capacitor. Under excite it (absorbing Reactive Power) and (perhaps logically)it behaves like a large inductor. The system voltage depends on the Reactive Power balance between generators and loads.

Hope this helps with the other end of the story.

A.

 

[DRWeig]

Zeusfaber,

Nicely narrated!

I'd like to add that in the case of driving power into a large network (of which you're a small part), an induction generator can be a lot more economical and easier to control. Simply by driving the machine past synchronous speed (increasing slip), power flows from what would otherwise be an induction motor into the grid.

Small power producers around this area (typically burning waste sawdust to make steam) use this technique often. Synchronization is easy, the machine is started like a motor. Once it reaches idle speed, the steam valve is controlled to maintain a constant backpressure. Then you have a constant-pressure supply of steam for your process and the power output varies only with the flow rate.

It's a good energy recovery tool.

Best to ya's,

Old Dave

 

[abeltio]

121202...
although i may be repeating to some extent the concepts already posted i will try, as usual, to put it in layman's terms...

1. from your posting i understand that you are dealing with a aeroderivative unit... which has a GG and a PT... most of the power generation by Gas Turbines is performed with single shaft units (industrial type, also called heavy duty)

one basic concept...
when the generator IS NOT synchronized to the grid:
the governor adjusts turbine speed
the excitation adjusts generator voltage

when the generator IS synchronized to the grid:
the governor adjusts turbine (generator) load
the excitation adjusts generator vars

2. the load control scheme depicted by CESSNA1 is typical of units connected in DROOP to a big grid... this droop control only very seldom happens, because DROOP is a PARTIAL LOAD CONDITION... in this case the unit is said to be in "SPEED CONTROL"... most operators select nominal/max/base load whatever the denomination is. because otherwise the unit has excess power that is not being utilized...
depending on the contract with the grid, if the operator is compelled to run the unit at partial loads by the dispatch it will not be below the "technical minimum" declared by the operator... and usually at a PRESELECTED LOAD.
Otherwise the dispatch will ask the operator to shut the unit down.

in the case of aeroderivatives the GT control is looking a the speed of the PT. in case of the industrial type units the control is looking at the Unit speed... there is a "called for reference speed" and the actual speed of the unit...

the difference between those two speeds is proportional to the excess fuel required to maintain the nominal speed of the unit as the generator is loaded...

example:
the synchronous speed of the unit is 3600 rpm = 100%, the called for speed or reference speed is 102%
the difference 2% is proportional to the load of the unit.

what is this proportion? well that depends on what DROOP was programmed into the control system... if the droop is 4% then at 102% reference speed the load of the unit would be 50% (102 - 100 = 2, 2/4 = 50%)
this also means that when the called for speed is 104% the load of the unit will be 100%...
when does the unit reach 100% load? that is usually defined as TEMPERATURE CONTROL. this means that the fuel is not limited by the speed control, but by the exhaust temperature of the unit.

in other words...
in speed control... the fuel system says: something is stopping the generator (the load) give more fuel to maintain the speed!!!
what does the temperature control do now? it says: hey, my exhaust temperature is quite low... give me more fuel (more than requested for by the speed control)...
so the control system decides:
though luck, temp control... speed control wins... (always go for the lowest bidder).

the fuel system keeps giving more fuel... at some point (some load) the temperature control system says... hey, hey, hey! hold it bud! if you put more fuel into the unit... we will destroy the combustion system! that's enough!

the control system then says: ok, temp control... you win now.

IF THE OPERATOR SELECTED MAX LOAD, THE UNIT WILL LOAD GOING THRU SPEED CONTROL UNTIL IT REACHES TEMPERATURE CONTROL

IF THE OPERATOR SELECTED PARTIAL LOAD, THE UNIT WILL STAY AT WHATEVER LOAD WAS WHEN THE OPERATOR STOPPED PLAYING WITH THE GOVERNOR... AND WILL INVERSELY FOLLOW THE FREQUENCY CHANGES AS EXPLAINED BY CESSNA1

MOST UNITS ARE OPERATED WITH MAX/NOMINAL/BASE LOAD SELECTED.
when the fuel called for by the temperature control is LOWER than the fuel called for by the speed control...
the unit reached temperature control... the load cannot go any further, that day with that ambient temperature... and the unit stopped reacting actively to the changes in frequency... if the frequency of the grid drops the load will drop a bit... if the frequency of the grid goes up the load will increase a bit... but basically it will not react.

if the frequency changes are big enough it will activate protection relays that will either trip loads (underfrequency) or trip generator breakers (overfrequency)

another operating mode is PRE-SELECTED LOAD. the unit is programmed to achieve a certain load (lower than the nominal load) IN THIS MODE... THE UNIT OPPOSES the droop control... why?
preselected load has a deadband where the speed control controls the unit (remember, load < nominal load means speed control fuel is lower than temp control fuel) and therefore the unit is drooping to the grid frequency...
as soon as a deadband limit is hit... the unit OPPOSES THE DROOP CONTROL... if the grid frequency is decreasing... it will load the unit as explained before... as soon as the load hits the upper limit of load control the control system will say: hey! i've been told not to go above this load! so it will UNLOAD THE UNIT until the load is again within the deadband... JUST THE OPPOSITE OF WHAT THE GRID NEEDS!!! BECAUSE IF THE frequency is dropping... the last thing the grid needs is a unit unloading!

3. in case the unit is the only one in the grid (island mode operation) or is the "isochronous driver"... usually a big steamer or hydro, not a GT) the operating mode selected is usually "isochronous" which means that the unit will maintain the nominal frequency (50Hz or 60Hz) VERY TIGHTLY, NO MATTER WHAT LOAD is required...

this has an inherent danger... that the unit will try to load BEYOND the temperature control...
if you have a couple of units, one in isochronous and the other drooping to the first one, and the droop unit trips...
all the load in the system will go to the isochronous unit... the control system should then reset from isochronous control to droop control, so the system frequency will drop and the unit will only reach temperature control (hopefully) and not trip

the protection switchgear will then detect underfrequency and trip non critical loads until the system is stabilized again... this is to avoid tripping the unit because of external causes.

there are many more interlocks and control schemes... but if i keep going this will turn into small novel...

4. generator control... at the operator's level there is not much to control on the generator... the excitation system will help maintain the voltage of the grid by means of VARS... if the generator puts out positive vars (excitation increase) this will try to increase the line voltage...
if the generator "absorbs" vars (negative vars) it will try to lower the grid voltage...
who decides whether to push or pull vars? the grid dispatcher... they would call the operator and ask to put so many + or - vars to help control the voltage of the grid...

operators are usually reluctant to absorb vars because negative vars may hit the UNDER EXCITATION LIMIT which is dangerous because it may cause the generator to skip a pole (out of step) - not a pretty sight...
most generators are designed to operate between certain power factors and within the capability curves... operating for long periods of time at low power factors may affect the life of the generator.
this by no means pretends to be the ultimate guide to GT-generator control and operation... there are so many other considerations and details involved that it is impossible to list them all here...


if you made it this far... thanks for your patience.
HTH

saludos.
a.