Rolling blackouts and power grid design

[Mark R]

A major 400 kV substation in my home town was damaged by severe flooding earlier in the week. As a result, a significant part of the city is on rolling blackouts, as it isn't possible to reroute all lost power through local grids.

The local electricity supplier has been good enough to give details of the blackout rota, stating which addresses will be disconnected at which times - however, I can't understand the organisation of the disconnect groups.

Does anyone here work in the industry? What criteria are used in determining which areas get disconnected first? Why are the individual switch groups so scattered and irregular?

Surely, if the failure was at a single high voltage substation, I would expect whole city blocks to be isolated at a time. However, looking at the map (X marks the spot), the distribution seems to be random (each color is a switch group which is dropped simultaneously).

The areas closest to the dead substation seem to have preserved power, but it's the areas further away that are rationed. All the areas are residential with some light commercial, so it's not even that that seems to be a major criterion.

Any explanations?

 

[JohnCU]

The grids are set up in such a manner as to ensure that a large percentage of customers in the same neighborhood would not be blacked out concurrently, which could invite looting and other related problems.

From Wikipedia...

and from howstuffworks

Hospitals, police stations, fire departments and some residents located near these emergency agencies are unaffected by the rolling blackouts.

I'll try to ask someone tomorrow who works in transmission (i'm in systems engineering for a nuke plant on the east coast). It's about that time of year for the "system reliability" messages to come out saying that we are at peak demand and don't do anything to the plant that could cause a shutdown. It's about $1 million a day per unit when we are shut down in the summer.

 

[PowerEngineer]

Sorry to hear about your problem!

I think it's fair to say that most utilities will try to spread the inconvenience caused by a rolling blackout across as many customers as possible. The number of customers that it makes sense to involve depends on the nature and location of the problem. It sounds like your outage reduced capability to deliver power (from the high voltage transmission system) down to the lower distribution levels; they're probably shedding load to keep the strain on the remaining equipment within ratings. The locations of the loads may be closer to the equipment being strained by the outage rather than to the location of the outage itself. I'd expect that you'll see rolling blackouts during the higher-load weekday days and probably fewer (or none) at night when customer demand is less.

So the map shows you the locations of the loads that help reduce the flow on the weak links in the damaged electrical system. Load shedding is usually accomplished by opening distribution feeder breakers in substations. Utility practice on this varies from place to place, but a feeder commonly serves 500-1500 customers with combined nonsimultaneous demand of 2-6 Mw. Therefore, each color may represent the areas affected by opening many individual feeder breakers (that together form one block, which in my utility's case is roughly 50 Mw).

As John's suggested, there will be some loads that are judged to be too important to include in rotating blackouts. Hospitals, police and fire are often exempted (although as more hospitals install back-up generators, the need to exempt them is lessened). 911 and other call centers are usually exempt too, as are water pumping and treatment plants. If one of these loads is on a feeder, that whole feeder is taken off the list (i.e. not included in any of the blocks for load shedding) -- a good deal for the other customers sharing that feeder!

Out of professional curiosity, I'd like to know how long each block is left out of service before being restored.

Thanks and good luck! :light:

 

[JohnCU]

transmission guy was on vacation, he should be back next week.

on a side note, i remember we had a power flow simulator/software in power class in school and we had all these scenarios we could run where you take out a line and see what happens or a generator goes down, etc. and this was just using 6 nodes or so, imagine the thousands and thousands of nodes on the real grid in an area.

 

[heyheybooboo]

I'm not an engineer but formally a town manager for an 'electric city' who had to pay close attention to the system engineers in situations such as this. I would think as noted the anticipated 'Peak Demand' loads drive all decisions regarding their policy.

The areas closest to the dead substation seem to have preserved power, but it's the areas further away that are rationed.

In effect I believe they are protecting those areas from substantial voltage drop in the case of a spike in demand.

 

[Mark R]

Power back on everywhere now.

Rolling blackouts were for 3 hours at a time. 6 groups - 1 group would be blacked out at any one time, 1 for blackout incase of power shortage, and 1 for blackout in the event of severe power shortage. The other 3 would then have 3 hours with reliable power.

Interestingly, the bulk of the rolling blackouts only went on for about 36 hours. Although the flooding damaged 2 co-located substations (a 400->275 kV step down maintained by 'National Grid' which was a major infeed to the city, and distributed 275 kV to the rest of the city; and a 275 kV to medium voltage substation used by the local power distribution company), the damage to the medium voltage substation was relatively minor, but repairs to the high voltage equipment took several days.

The local distributor worked around the lack of 275 kV, by running several temporary 'extra high voltage' cables (their term - I don't know what they mean by that) between several of their local substations on the edge of the most severely affected region.

 

[Modelworks]

I live in an area that experienced similar blackouts due to flooding.
At the time though they brought in the national guard with the huge military generators and restored the power.
It went down for several hours a day, but it was better than nothing.

They rebuilt the substations, they placed them on platforms with steel poles beneath, about 15 feet off the ground.

 

[Mark R]

Well, you never know what you can find with google. Amazingly, I managed to pull up maps and schematics for the major grids (33 - 400 kV) interconnects, transformers and busbars.

This also explains the workaround by the local power distribution company. I couldn't work out why they were running 33 kV cables between 2 substations barely 1/4 mile apart. Turns out that they are seperately fed from 2 major infeeds at different corners of the city. Only one, however, has an interconnect with the flooded area. By installing a temporary interconnect, they were able to increase the capacity to route in extra power.

As an aside, I knew I lived in an industrial town, but I must admit I was pretty surprised how many industrial plants took their power at 275 kV (i.e. they had 275 kV o/h lines running direct from a 'supergrid' substation to their premises). I don't think power entered the building at 275 kV, because google earth shows the lines terminating at private substations with up to 4 transformers.

 

[JohnCU]

Originally posted by: Mark R
Well, you never know what you can find with google. Amazingly, I managed to pull up maps and schematics for the major grids (33 - 400 kV) interconnects, transformers and busbars.

This also explains the workaround by the local power distribution company. I couldn't work out why they were running 33 kV cables between 2 substations barely 1/4 mile apart. Turns out that they are seperately fed from 2 major infeeds at different corners of the city. Only one, however, has an interconnect with the flooded area. By installing a temporary interconnect, they were able to increase the capacity to route in extra power.

As an aside, I knew I lived in an industrial town, but I must admit I was pretty surprised how many industrial plants took their power at 275 kV (i.e. they had 275 kV o/h lines running direct from a 'supergrid' substation to their premises). I don't think power entered the building at 275 kV, because google earth shows the lines terminating at private substations with up to 4 transformers.

you can get all that from google?

 

[PowerEngineer]

Originally posted by: Mark R
Rolling blackouts were for 3 hours at a time. 6 groups - 1 group would be blacked out at any one time, 1 for blackout incase of power shortage, and 1 for blackout in the event of severe power shortage. The other 3 would then have 3 hours with reliable power.

Thanks for the response. 3 hours is a pretty long time to be without power. Glad to hear everythings back to normal now.

 

[dkozloski]

In our area we have the largest electric storage battery in the world that supplies sufficient power to keep the lights lit until standby turbine and diesel generators can be started to pick up the load. The grid can go completely down for extended periods and the lights never flicker. There can still be local outages from trees across the highline and such but nothing major. The design criteria is a little different when dealing with the possibility of an outage at -60F.

http://www.gvea.com/about/bess/

 

[Analog]

Originally posted by: dkozloski
In our area we have the largest electric storage battery in the world that supplies sufficient power to keep the lights lit until standby turbine and diesel generators can be started to pick up the load. The grid can go completely down for extended periods and the lights never flicker. There can still be local outages from trees across the highline and such but nothing major. The design criteria is a little different when dealing with the possibility of an outage at -60F.

http://www.gvea.com/about/bess/

it can provide 27 megawatts of power for 15 minutes.

* 13,760 liquid electrolyte-filled Ni-Cad cells
* Each battery is roughly the size of a large PC and weighs 165 pounds
* Total BESS weight - 1,500 tons
* Batteries have an anticipated life of 20-30 years


Think about all the toxic waste when they have to dispose of those things! (recycle maybe?)

:shocked:

 

[dkozloski]

The original contractor, SAFT, is responsible for disposal of all waste and recycled batteries. It is all in the contract. The whole deal is as clean as a hounds tooth. No environmental hazard here.

 

[Mark R]

Yeah, I'm sure the manufacturer would want the batteries back anyway. Nickel and cadmium are both very expensive metals - Nickel especially.

Interesting that they chose NiCd - I suppose, it's probably because NiCds have longer service life, better low-temp performance and better reliability than lead-acid.

 

[dkozloski]

Ni-cads have a very low internal resistance and there is much less heat at high discharge rates compared to lead acid.