Wind farms behaved as they should in S.A. storm, page-4

The exact reason for the first few wind farms reducing output is not defined in the AEMO Preliminary report, I read it over lunch yesterday.

There was a 123 MW reduction in output from 4 wind farms at 16:18:09, which is 6.4% of system load pre-event and 14% of total wind generation pre-event.

There was a further 192 MW reduction in output from 2 wind farms at 16:18:15, which is 10% of system load pre-event 22% of total wind generation pre-event.

The line faults added instability and complexity to the derivation of simple solutions, but the above highlights that in the absence of rapid load shedding, the Heywood interconnector needed to provide an extra 315 MW of power on top of the 550 MW that it was providing pre-event.

We cannot derive any concrete conclusions prior to the release of the final report. However, if wind farms are likely to drop output during storms (as is expected for their own protection, remember, solar is also not immune), then the system needs to adopt a few changes.

The federal government needs to stop talking about slowing the growth of renewable generation and needs to start talking about adapting the grid into a 'smart grid' to cope with rapid changes in load and generation

The following options are possible in tandem:
- Faster response for the Under Frequency Load Shedding (UFLS) system (black out low priority loads faster)
- Rapid power dump into grid (battery storage with inverters can do this, it need only have enough capacity to slow the frequency droop enough for the UFLS system to operate); This may be as simple as mandating a battery system into renewable projects with inverter capacity rated to say 80% of the system output (read inverter capacity, not battery capacity, it only needs to run long enough for the UFLS to operate suitably)

The biggest flaw with my argument is that I am proposing intentional power outages. Unfortunately, without mass scale storage technologies already available, the interim answer (possibly for the next 2 or 3 decades?) is to maintain a controlled load shedding system. This allows the grid to remain reasonably stable whilst thermal and other generation comes online or ramps up output.