[joeshort (OP)]

You can now see the state of the UK power grid from your web browser.

newbielink:http://www.dynamicDemand.co.uk/grid.htm [nonactive]

I'm from a new not-for-profit called Dynamic Demand. We're promoting the idea of using ordinary electrical appliances as a way of removing dangerous spikes in electricity demand. The ideas is to design air conditioners or refrigerators etc which can respond to the current state of the grid so that they are less likely to be demanding power at times when power is scarce. The idea is to make the grid more efficient and prepare the grid for renewables. We're trying to get support for this idea and would love to get people's views.

Joe

See the current state of the UK electricity grid:
newbielink:http://www.dynamicDemand.co.uk/grid.htm [nonactive]

[chris]

Interesting idea, Joe. How do you plan to inform power hungry gadgets such as those you mention (air conditioners etc) of the on-going status of the grid ? Presumably some kind of load signal transmitted via the power line itself ?

Chris

"I never forget a face, but in your case I'll make an exception"
 - Groucho Marx

[daveshorts]

The frequency of the grid indicates whether it is over or underloaded - if you are drawing more power from the system than you are putting in the turbines will loose kinteic energy and slow down essentially, so you can use this frequency information to decide when to turn on the fridge compressor.

Nice idea, but you would have to do some very careful system modelling as if you got into a vicious feedback loop it could be very messy!!!

I guess that you would also have to introduce new electricity meters that charge different prices for electricity depending on what the frequency is doing to make it worthwhile installing a fridge with this feature....

[chris]

Are you sure about that Dave? Are you telling me that the grid runs at less than 50Hz when the load is high ? I thought there were stringent controls to preserve the integrity of the grid frequency ?

Chris

"I never forget a face, but in your case I'll make an exception"
 - Groucho Marx

[chimera]

quote:Originally posted by daveshorts

Nice idea, but you would have to do some very careful system modelling as if you got into a vicious feedback loop it could be very messy!!!

Dave's right there... I remember that such large systems have quite a lot in common with turbulent/chaotic systems, and with some of those you want to be quite careful, feedback is a real danger. It's what causes some major problems at times anyway, and they sometimes have no real clue as to what was the cause of it. There's factors like load balance, but also the new windmill parks give trouble, sunspots and all, you name the culprits. It's not a simple matter.

[daveshorts]

It is closely regulated but not constant according to:
http://www.nationalgrid.com/uk/activities/mn_quality.html
the frequency is leagally requred to be held to within +/- .5Hz of 50Hz and it normally runs within .2Hz

Measuring grid frequency to within .1Hz would be quite easy.

I think if I were designing the system I would want some sort of radio control of the feedback in case it all went horribly wrong and unstable...

[gsmollin]

I already have it, it's called off-peak power. It is connected to my hot water heater, and the pool pump, although you can connect to any 240 V load that cannot be transferred back to the residential power. For 6 hours a day in the summer, and 4 hours a day in the winter, a radio signal disconnects those appliances. There is no need for feedback loops. Most power loads are predictable, and with enough load margin, the power grid remains stable.

The price is right. I pay $0.14 USD a kW-hr for rate R power, and $0.05 for rate OP.

[daveshorts]

I think this is to deal with the unpredictable power loads such as when everyone turns on their kettles in the middle of a sports game. I think that at the moment there are lots of flywheels (or something similar they are not clear) kept spinning to act as short term energy storage. This wastes a load of energy, and they want to spread this load balancing to fridges rather than centralising it...

[joeshort (OP)]

quote:Originally posted by daveshorts

It is closely regulated but not constant according to:
newbielink:http://www.nationalgrid.com/uk/activities/mn_quality.html [nonactive]
the frequency is leagally requred to be held to within +/- .5Hz of 50Hz and it normally runs within .2Hz

Measuring grid frequency to within .1Hz would be quite easy.

I think if I were designing the system I would want some sort of radio control of the feedback in case it all went horribly wrong and unstable...

The frequency is a direct result of the speed of rotation of all the synchronous generators on the system. When there's too much load, they all slow down slightly because there is no longer enough torque being generated by the steam. So frequency is like an indicator of power imbalance.

Frequency is controlled by automatic governor action. Certain part-loaded power stations monitor frequency all the time and alter their output accordingly to keep the frequency within limits. In other words, they open up the steam valves whenever the frequency drops.

When I kicked off this topic, the link I posted actually goes to an on-line frequency meter (which is very cheap and measures frequency to within 0.001 Hz or more, though it is uncalibrated as yet), so you can see this frequency signal for the UK grid in real time: newbielink:http://www.dynamicDemand.co.uk/grid.htm [nonactive]

Dave made an interesting point about feedback. At Dynamic Demand we've done some detailed simulations and there's no disastrous feedback detected yet, but you'd certainly want to be sure! I think the system operator (i.e. National Grid) would want to do some simulations on their detailed network simulators. However, my opinion is that considering there is already feedback on the system (in the form of governor action), you'd just need to ensure that the two control loops were similar and compatible.

Some people reckon you may well want some radio-switching ability in order to be able to participate in a market. The National Grid pays quite a lot for frequency control (over £80m a year) so you could offer a competative service as long as the regulations allowed you to. I'd like to think it won't be necessary but at this stage there are many possibilities.

Glad this has stimulated some interest!

See the current state of the UK electricity grid:
newbielink:http://www.dynamicDemand.co.uk/grid.htm [nonactive]

[chris]

I love your real time frequency meter - that's fantastic !

When the frequency drops to the "Areas start blacking out" zone, what sort of voltage drop does this represent ? Presumably there is one if the load rises sufficiently to cause a frequency drop ?

Chris

"I never forget a face, but in your case I'll make an exception"
 - Groucho Marx