Power systems

1. May 29, 2007

ineedmunchies

I've come across a problem on unit commitment to a power system.
i.e. which power plants to use to power a system, which is determined by their loading range, incremental cost of power generation, and the cost of the station running at all.

However I'm being asked to also determine the governor gain of each operational unit (power station).

Anybody able to help me out?
What is governor gain and governor droop?
I can't find anything about them online, I have a feeling they're perhaps terms my lecturer has coined himself.

2. May 29, 2007

Staff: Mentor

It sounds like the governor would help to control the speed of the turbines or something, to help keep the plant synchronized to the grid? What do your textbooks and other sources say about how to keep your plant synchronized to the grid? What factors are involved? How do you measure how much power you are putting into the grid (and therefore can charge for)?

3. May 29, 2007

brewnog

Gain controls how sensitive the governor is to speed/frequency changes seen by that individual unit. A higher gain will result in faster synchronisation and better transient response, but may result in very 'twitchy' governor responses.

Droop is the amount by which the speed setpoint is above the actual network frequency. This mode of running allows a generator to 'load share' with the network to which it is attached, so that power output of that unit can be controlled.

No idea how you'd calculate these based on the information you've provided in your post though.

Last edited: May 29, 2007
4. May 29, 2007

dlgoff

If this is an economic dispatch problem, then aren't they asking how much each unit on-line will need to increase (gain) for a given increase in demand? e.g. for a unit that is cold, you would have to look at start-up heating; for running units, some might be economical to ramp up more based on the system impedance than others.

5. May 29, 2007

ineedmunchies

Thanks for all of your answers, very helpful in understanding it. Its a simplified mix of what you've all been saying, its first a costing exercise, and then a problem about gain.

I found the relations i needed for my problem;
governor gain = -(delta P/ delta f)
Where delta f could be found using the governor droop value (delta f = governor droop(%) * f) where f is the original frequency.

I don't know if these will be any use to anybody.