Critical Fault Clearance Time - Stability Criteria

[drchaos]

Hi, I am looking at critical clearance times (CCTs) at a variety of different 220kV and 110kV buses for 3-phase and single-phase faults, using PSS/E. I've been looking at the different stability criteria people use. For instance, a previous study of this sort deemed that instability occurred if there was a 300 degree angle difference between smallest and largest rotor angle. Other criteria I've seen used are max rotor angle > 120 degree, or > 180 degrees. Does anyone have any thoughts on the above?
Also, for single-phase faults, I find that faults can be left on almost indefinitely without the system becoming unstable (I am not modelling any protection relays - just looking at power and angles) - but my boss isn't too happy with that. Are there any power unbalance criteria that people use to determine max length of a single-phase fault?

Thanks
DrChaos

 

[QBplanner]

Hi :
in our system unless any identified critical machine stability issues, 220kV we normally use 6 cycles to clear the faults 3-P or 1-P fault. 138kV system typically 8-10 cycles. transformers faults clearing are faster than the line faults I mentioned above.
We don't use largest or smallest machine to define machine stabilities. First of all, you have to understand which angle you are using for your stability study. relative angle or absolute ones. if relative angle compare to synchronous axis of the whole system or a remote lager capacity units.

when I run PSS/E simulation , power angles are the last state vectors I will check.

By the way don't use 180 120 or 300 degree I saw 540 degree angle differences between two machines. it does not mean machine will loss synchronism.

Again, I don't know your system but I don't believe you can leave single phase to ground bolted fault w/o causing any problems to the mahcine stability. BTW how do you do you SLG fault? if your basecase have sequence data may be you can do it. otherwise, you have to use positive sequence equivalent method to get the series impedance and shunt impedance before you run the PSS/E study.

 

[drchaos]

Hi QBPlanner, Our protection system also operate within the time-frame you mentioned: 6 - 10 cycles. These studies are intended to show whether certain machines are more at risk of becoming unstable in the event of a protection failure etc.
In my studies, I use relative angles, relative to a big machine I use as the swing generator. Yes I have negative and zero sequence data so I can do unbalanced faults in pss/e.
So, what are the first things you check, if not power angles? I had always thought that once you get pole-slipping, then the machine will trip.

 

[QBplanner]

Hi:
does your base case have build in generator out-of step relay based on phase angle or impedance?
If yes, then you may have to look at rotor angle first.
In our base case, we don't have build in out-of step relays. So I always check the Delta W to see if your machine is speed up or not. Check the terminal voltage to see if your machine can maintain the terminal voltage after distrubance. Check the field current to see if your excitation current belows up to what level assuming you don't have build in over-Ex limit models. Then the last thing is your power angles make sure your reference machine is far remote from your study machines. Having said that, it also depends on the accuracy of your base case modeling. Don;t always trust PSS/E results. sometiems you have to use your own judgements.

 

[drchaos]

In my case, there is a PSSE option to scan for out-of-step conditions. But I also look at rotor deviation. I didn't realise that terminal voltage was something to look at - are you saying that if the terminal voltage does not recover post-fault to pre-fault levels, then this is a bad sign?

Thanks for your replies!

 

[QBplanner]

As far as I know that out-of-step in PSS/E get nothing to do with your machine synchronism. You better check it out since I know what you are referring to
If machine terminal voltage cna not recover to pre fault condition or beyond certain range check out with your area codes. Then check out your field current /field voltage if it can not come back to a pre fault level (Aassuming you don't have over excitation limiter models) then you machine is in the risk of lost of synchronism no matter your rotor swing to what degree. the 180 degree is a pure one machine to infinite bus case. all the text book case will show you 180- rotor angle stuff. In reality, it never works out in that way. Depending on the accuracy of your models. machine, exciter,limiter, governor, relay models if you really wants to know exactly what will happen. EMTP is a better software but need more efforts.
PSS/E is a system software, and will only tell approximate critical clearance time. Based on my experience, whenever, you get a fault clearing time, add 1-2 cycles to cover your study errors and leave some room.I think I talked too much your bosses may have different ideas how how the works has to be done.
Good luck.