Islanding detection on grids with distributed energy resources

[chao_david]

There are lots of existing methods for detecting islanding conditions when the circuit breaker at the side of the utility unintentionally trips. I've seen passive methods, active, and intelligent systems that detect the islanding and trips the DG breaker within 2 seconds as required by IEEE standards. I think I'm missing the basics here because from the diagram, why not just coordinate the tripping of DG CB2 and utility CB1? Why not just automatically trip both CB1 and CB2 when there is a fault at the CB1?

I've thought about this because active method equipment are expensive to install and would distort the power quality.

 

[bacon4life]

Stevenal-The WECC UFLS plan is not intended to have particularly precise amounts of load/gen since it is a broad safety net. The WECC UFLS plan specifies the minimum load shed amounts only on the date/time of the local system peak. Depending on the weather and date/time, the actual load on UFLS circuits can be quite a bit different than the load on UFLS feeders during the local system peak. This particularly true if UFLS relays are are preferentially assigned to certain kinds of loads (i.e. mostly residential). On my own system, the actual amount of load armed to trip varies from about 25% during summer 2 am hours, to about 30% during a peak hot summer afternoons, and to over 40% during a peak cold winter mornings. With the changes in generation resource mix, I am concerned that the most severe UFLS case may shift from being summer/winter peaks to occurring during should season off-peak hours.

For momentary interruptions on residential feeders, I typically see load return to normal very fast. For industrial feeders, often see significant delays for customer running production lines. It might be interesting in a future assessment of the plan to look at the kinds of load being armed for restoration.

If I had magic wand to address UFLS, I would start requiring all large battery chargers to include a frequency droop response just slightly higher than the first stage of UFLS relays.

 

[davidbeach]

What can be very nasty is when you trip/shed say something less than 1MW of load but then reclose/pickup 10MW of load moments later. I just don't think that's being accounted for.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[bacon4life]

As in solar or other distributed energy resources on the distribution circuits? Newer inverters resume pretty quickly, but I cannot remember exactly how fast. I had not thought about your point since in my region we still have less than 2% solar penetration. Hopefully folks with large amounts of DER consider that when picking which circuits to arm for UFLS.

It seems like the interconnection needs to do a better job modeling generation separately from load, rather than just netting all distribution level generation at the substation level for interconnection wide simulations.

 

[chao_david]

As I'm new to this field, just wanted to say I've learned a lot from your discussions. Currently doing studies on risk assessment of unintentional islanding in a particular area. Found out that parallel operation of inverters with respect to the point of common coupling degrades unintentional islanding detection performance of inverters, especially with combinations of passive and active methods. Although, I'm not really sure if it's practical to conduct this study given that we have low penetration level (unlike with HECO with 25% of its customers have solar PVs).

Do you have a rough idea on what inverter anti-islanding scheme majority of your net-metering customers use?

 

[cranky108]

In a short answer, NO. Information on inverters owned by customers is not available to utility engineers. This should come from the customer, but as most customers wouldn't know what the information is, they would likely not provide it. That said, most inverter manufacturers also will not provide what they consider propriety information. This is the misinformation cloud that many utility engineers have to live with, and why there is and will be more regulations on the inverter side of the utility interface.

 

[bacon4life]

As a practical matter, Cranky is right that utility modeling engineers have very little info on behind the meter solar generation. I think my area is a bit unusual in that customers are required submit equipment model numbers as part of distributed generation projects. However, solar installers regularly substitute equipment after the initial info is submitted to the utility, and customers regularly replace equipment without filing as-builts. Unfortunately, even if we had perfectly accurate equipment lists, it is still quite hard to get any technical info out of the invert manufacturers that is appropriate for grid level modeling. It is also unknown how to effectively aggregate detailed inverter models to be effective for interconnection wide studies. Within the WECC region, the distribution system is usually equivalized into chucks of at least 10 MW.

FYI, last fall FERC published an order that will probably lead to significantly more rules for modeling of inverter based resources. Currently NERC is working through the standards development process on this topic.

https://www.ferc.gov/news-events/ne...andards-registration-improve-grid-reliability

 

[wcaseyharman]

SAR 2020-06, which should become the new NERC-MOD-26-2, if eventually approved will likely require extensive modeling of transmission-connected inverters including Electromagnetic Transient models. This won’t cover distribution resources though, so it will be interesting to see how the utility industry will address the distribution performance when the IBR penetration gets high enough.

 

[cranky108]

Still I doubt that MOD-026 would resolve the issues of the protection engineers that are dealing with reduced negative sequence which is used for determining fault direction, and the over all reduction in fault currents. I believe at some point some synchro's machines will need to remain on the grid for stability, and inertia.
People may complain, but in addition to net metering, I believe there should be a service charge to pay for the connection, and billing (meter reading). No matter what people think, it is not a free ride.

 

[wcaseyharman]

I’ve spoken to the protection engineers at my company, and right now the hard part is to get equivalent sequence impedances from inverter manufacturers. There’s a number of major issues we’ve identified that distributed inverters cause:
- my company runs an effectively grounded system to limit the overvoltage due to ground faults. High negative sequence impedances can erode this and cause higher overvoltages, damaging customer equipment and possibly failing surge arrestors
- desensitization of the substation protection, since each Distributed generator acts as a remote infeed to line end faults.
- the aforementioned islanding concerns
-eventual backfeed onto transmission and the FERC implications.
- lack of requirements for grid modeling. DG will eventually affect system stability, but right now there’s no established models to represent them.

It’s going to be an interesting future in the utility business.