If you are going to drag all sorts of frequency components into this discussion, then you have different frequencies all the time. I am, if you didn't get it, referring to harmonics.
You have to decide, Are you discussing frequency as measured with a frequency meter? Or are you discussing transients? Or frequency defined as d(alpha)/dt? Or are you trying to define some sort of Unidentified Frequency Object aka UFO?
We need to hear what the OP actually meant with his question. It may even be that he wasn't aware of the differences?
As to "local load shedding": There is no such thing. If grid frequency goes down, then all objects set to switch off at that frequency do switch off. No matter where in the grid they happen to be situated. There may exist load dependent extra conditions to help decide which object to shed. But then you have something else then frequency that decides.
Freaquency is stable for all the nework (in avarage) but this is not true during transients in the system, so you can find different frequencies in one system at one time (during transient, so if you are near the faul location the frequency at this piont will be effected more than the other points in the network locating so far from the piot of the fault.
I must warn against bringing current frequency into the discussion. While current has frequency, it is the voltage frequency that is usually the measured quantity or used for load shedding
Also, the harmonic frequencies are unchanging. While voltages or currents at harmonic frequencies might have changing magnitudes over the system, the harmonic frequencies themselves do not. The second harmonic on a 60 Hz fundamental system is 120 Hz where ever you are.
If anyone is still interested in this thread, there is a document on Baslers website which explains the frequency swings at remote busbars during faults.
Thanks for the link, Marmite.
Before we can get on the same page, we will have to agree on how to measure frequency. Do we measure peak to peak, full cycle zero crossings or half cycle zero crossings.
For example, we are familiar with the wave form of an asymetrical fault current with a DC offset. If we measure the peak to peak time of a complete cycle we will see our fundamental frequency. If we measure the zero crossing time of a complete cycle we will see a small drift in frequency as the phase shift introduced by the fault slowly decays. It can be seen that each succesive full cycle zero crossing is not at the same point on the wave form as the DC component decays.
Measuring the full cycle zero crossings will show a frequency different from the fundamental frequency but the frequency difference will follow the decay of the DC component. When the DC component has decayed to zero the zero crossing frequency difference will be zero and we will be back at fundamental frequency.
Determining the frequency by the half cycle zero crossings will give alternating high and low frequencies for every half cycle.
I still stick to my original thought that the frequency is basically the same everywhere on the system.
Different measurment techniques may show apparently different frequencies.
Even though a DC offset or a phase shift on a wave form may look like a frequency change to a relay using zero crossing techniques, the full cycle peak to peak may well still be the fundamental frequency.
I will accept the case of the "Hunting" generator as a differend frequency, because the speed of the generator is varying, but as the excursions are limited to about plus and minus 90 degrees, I am comtent to also call this a series of phase shifts.
Note that the Basler information is at some points discussing frequency drop when a section of the grid has become islanded on overloaded, and the frequency of the islanded portion is decaying, and at some points the information is discussing "apparent' frequency shifts as determined by a relay using zero crossing determination of frequency.
What's the point?
How do we measure frequency? Half cycle, full cycle, peak to peak, zero crossing, or average.
Can we accept the small timing variation when a phase shift is introduced or must we consider this a short duration frequency change.
When I want to pick up some more load on a paralleled generator, I open the governor. The shaft assumes a new relative position a few degrees ahead of its previous position. Technically, yes, to affect the relative position change, the frequency has to have increased a very small amount for a very small time.
I'd rather call a phase shift a phase shift, and not a very limited frequency increase for a very short time.
"I also tend to call a spade a spade, rather than speaking about the dog's hysterectomy."
This is an attempt at a little humor and I hope it is taken that way. I respect all the points of view that have been presented here. I agree with most of them.
Bottom line, how do we determine frequency.
Respectfully
