Risk of Islanding Evaluation

[paddymurphy]

I have a photovoltaic site that I am designing to be interconnected on a particular utility feeder. The site inverter is equipped with an active anti-islanding algorithm, however there are a lot of other photovoltaic sites on the feeder, so there is a concern about how all the sites will interact. I am trying to put together a method of formally evaluating whether or not a risk of islanding is actually a concern on that particular feeder. I have created a model of the feeder (which includes conductor impedances and information of other loads & DG on the feeder). I also have a list of all the capacitors, regulators, and reclosers on the feeder. I'm told that looking at the "reactive power (VAR) match" on the feeder is the thing to do, but I'm not clear on how to do it. Any advice, explanation, or reference to a white paper would be much appreciated.

Thanks in advance!

 

[Parchie]

IMO, whenever the utility power cuts out, you don't want utility personnel to get fried when they will work on downed lines just because you'r still feeding back to the lines.
Your units should cut-out from your interconnection with the utility if it sees the utility is out.

 

[crshears]

Question, paddymurphy: do you work for the utility, or the PV plant engineering firm, or what? In my experience such analysis as you are describing is performed by the utility.

Also, depending on connection voltage, my utility often stipulates the incorporation of remote or transfer tripping from the transformer station to the individual DG site to not only trip the DG off if the feeder breaker opens, but to prevent the DG from closing in their point-of-common-coupling device until the breaker at the TS is once again closed.

This may or may not be applicabel in your case...

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[snerts50]

OP I don't know if you are still actively monitoring this thread, but check out this Sandia Paper.

http://energy.sandia.gov/wp-content/gallery/uploads/SAND2012-1365-v2.pdf

It's something we typically follow to determine if there is a risk of islanding, and if we need more detailed modeling. This next level of detail includes the actual real controls of the inverters, typically using Matlab. Then various island scenarios are modeled to determine if a risk is there, and if so, recommend mitigation, usually DTT.