[ewood]

Anyone here have deep insight on this? I thought a large part of the problem is the way that the Australian Energy Market Operator (AEMO) is operating the market? Their mandate seems to be to prioritise price over reliability causing artificial shortages when there are not enough bids to form a market for excess load.

It appears that there is sufficient capacity in SA but providers are holding out for higher prices or not wanting to meet demand because they are providing higher priced capacity from their operations in other states.

[roel_v]

Not a 'deep' insight I'm afraid, but the particular problem they're referencing is that SA isn't generating any power on their own (after shutting down coals plants a few years ago). So they import from Victoria, but all of that import goes through a single line; so then there's a storm that knocks that line out, and the whole state goes black. That particular line was fixed quite quickly (I was in downtown Adelaide at the time, I think we were out of power for 4 or 5 hours or so); but the more remote areas didn't have power until 3 or 4 days later.

I'm not sure why it took that long, but I guess it's that you can't just flick a switch and everything is powered on again. You have to phase it to avoid peaks and some areas/power stations might be hard to reach or a bunch of coordination/equipment checks are needed before they can be started up; and the more remote, the lower the priority.

So, more local buffering = less problems when the main feed drops out. If local areas can be self-reliant for some time, and help to smooth out peak demands on startup, it's much less of a problem when there is a failure high up the line. It doesn't solve the total production problem, obviously.

[will_hughes]

Reading the outage report it seemed more like piss poor planning on everyone's part. [1]

From the looks of things, even a 100MW system being instantly available wouldn't have helped prevent the outage (given the 445MW shortfall from windfarms disconnecting due to voltage drops), but it may have helped reduce the time taken to restart the network.

[1] https://www.aemo.com.au/Media-Centre/Update-to-report-into-S...

[londons_explore]

In a properly working market, at the point the power grid fails due to lack of supply, the spot price should be infinitely high.

Given that, no sane supplier would refuse to sell given they would be making unlimited profit. Hey, I'd be connecting up my windup flashlight if I got paid $999,999,999 per kilowatt hour.

The issue must be a bad market. Either people aren't paying for what they use, there isn't sufficient transmission capacity to have a sufficiently large market to avoid local monopolies, there are price caps or floors, the market isn't fast enough (no automated trading), or new entrants are limited from participating.

In a correctly working market, electricity never fails because all people using electricity at the time of the failure would be paying an infinitely high price, and therefore be bankrupt.

[alextheparrot]

Maybe the utility derived from having electricity doesn't make the good actually worth the prices you are saying? A seller makes no money if they price their good so far above market that there is no demand for it, not infinite.

[Callmenorm]

You're missing his point. Prices steadily go up as demand increases so suppliers bring on more (expensive) capacity to meet. You never get to the failure point unless demand exceeds all possible supply capacity. He just took this argument to the extreme.

[alextheparrot]

I believe I completely understood his point and my comment responded to that the price does not approach infinity, because no one will logically pay a price approaching infinity. The root cause of this is that I will only purchase something if I think the utility obtained from having it is equal to or greater than the utility my currency could purchase elsewhere. I don't have a perfectly inelastic demand for electricity.

Even if a good is quite scarce (Such as electricity in a blackout), the price is dependent on demand. Low supply does not inherently make a good expensive.

[euyyn]

The demand isn't going to increase in a blackout, it's the supply that diminishes, bringing price up. But the curve isn't asymptotic: the price doesn't go to infinity as the supply goes to zero.

[pjc50]

This is the "free market cannot fail, it can only be failed" argument of the day.