Frequency bias/response

[AnotherEskomite]

I know there is difference between frequency bias and frequency response and both are typically express in MW/0.1Hz.

What I need to know is how does one calculate or predict what the frequency will be be if a certain amount of MW of generation is lost. In other words can I predict what the frequency will be given that the load connected does not change following the tripping of a generating unit.

I know in the states it works very complicated with Are Control Error and balancing authorities. Lets just keep it simple for now where the system is quite a tight system and then you loose generation. Does the calculation make a difference if it is 50 or 60 Hz or is it based on per unit system and then you convert it to your base.

Any reply will be appreciated.

 

[mgtrp]

Immediately upon loss of generation, the frequency will begin to decay at a rate that is dependent on the combined inertia of the rotating equipment within the system (all the generators and motors) and the amount of mismatch between generation and load.

Loads may be reduced when the frequency and/or voltage are lower than normal, which can decrease the rate of decline.

It takes at least several seconds for the unit governors to begin to respond, at which point they should act to increase generation and arrest the decline in frequency. If there is insufficient spare capacity, then under-frequency load-shedding schemes may operate, or the system may collapse.

Once governor action has completed, the system will settle out at a new steady state frequency. For systems with at least one unit in some form of frequency-keeping/isochronous control, this will be back to 60 Hz. For a system based completely on droop, this will be something slightly below 60 Hz (unless to much load was shed by the under-frequency load-shedding scheme, in which case it could be slightly over 60 Hz). Systems with automatic generation control (AGC) may also settle out 60 Hz as the AGC increases output on the droop-controlled governors.