Using Symmetrical Component’s to calculate an unbalanced load - For: jghrist & cuky2000

[Elec81]

This thread is regarding closed thread238-106630 which discusses the use of symmetrical components to study load imbalance.

Figure Fig. 21(l) as given by jghrist (from the T&D book) appears to indicate that two of the loads (Zb) have identical impedances, is this correct? Any idea how this would be done with three different phase impedances?

I’d also like to get a copy of the working given by cuky2000, this no longer opens in the original thread.

Many thanks in advance.

 

[HamburgerHelper]

I don't think sequence components are going to help you with studying load imbalance. It is not that they can't be used but the sequence networks become very hairy once you start considering more than one type of a network. Let's say that you have different load impedance on on all three phases. That is a 2 contingency network. The networks start becoming very convoluted as you add contingencies. At best, you'll have a conceptional understanding of why you have certain components but the actual solution will be a mess to solve. Just throw it into ATP-EMTP. There will be no clean calculation.

There is a port method that is in a book by Anderson. (

https://books.google.com/books/about/Analysis_of_faulted_power_systems.html?id=0ugeAQAAIAAJ)

that makes it easier but I have never used it. David Beach used this method I believe to analyse a cross country fault.

------------------------------------------------------------------------------------------
If you can't explain it to a six year old, you don't understand it yourself.f.

 

[jghrist]

I agree that symmetrical components is not the best way to analyze unbalanced load conditions. For some particular situations, like two phases having identical impedances, solutions have been developed as shown in the Westinghouse T&D Book. There was an interesting paper presented at this year's Georgia Tech Relay Conference about phase component analysis as an alternative to symmetrical components. See

https://static.selinc.com/assets/Li...aulted_SC_20180213_Web2.pdf?v=20180502-155546

Phase component analysis wasn't practical without a means of solving sizeable matrices with complex values. These matrices can now be easily solved with general math programs like Mathcad.

 

[davidbeach]

What I did was way back when I was an Application Engineer and my phone didn't ring a whole lot; I was also doing a lot of related analysis for my Master's Degree course work. I certainly wouldn't bother today, I'd just drop it into OneLiner and put in enough simultaneous faults to cover the different load conditions. Faults don't have to be bolted, I run a lot of faults with a fault impedance of 99999 + j99999 ohms; that forces a prefault load calculation. Well you can do the same for your different phase loads; make three simultaneous faults, one SLG fault per phase with that phase's load on it.

Build up to it with what ever modeling software you're using so you have confidence in the results, but use that computational horse power.