Hacker News Reader

Comment by lol_catz | Hacker News Reader

My concern is what happens to grid stability as we replace synchronous generators, with their enormous rotational inertia, with thousands of grid-following inverters. They can all phase-lock to 60 Hz, but without enough inertia, what keeps that frequency stable during sudden changes in load or generation?

View on HN

Synthetic inertia. Essentially, loads that you can quickly either shed or bring on in response to fluctuations. There can be a market for that too and that's what many grids are slowly discovering.

There are still a lot of synchronous motors out there directly wired that provide real inertia. Though these two are moving to VFDs (which are cheap and variable speed). Still I expect battery backup systems will come online more than fast enough to replaces the losses.

> but without enough inertia, what keeps that frequency stable during sudden changes in load or generation?

Decentral battery units in substations, similar to what Tesla did in Australia many years ago. They can absorb or provide reactive power as well.

grid forming inverters : )

I don’t see many (any?) plants using them, is the technology there yet? Genuinely curious.

And on an unrelated note, it seems that hybrid power plants are becoming a talking point now and they seem to make a lot of sense. Like covering dams with solar panels saves on water, make PVs cooler/work more efficiently and you don’t need to worry about permits/building more grid connections because you can simply slow down hydro during daytime (and get along with grid following inverters just fine).

California requires all new solar installations to contribute to stabilizing the grid by helping out with regulating the voltage and frequency https://www.openintl.com/california-rule-21-interconnection/

The technology to act as inertia isn't actually hard, is it? You deliberately offset your voltage by a fraction of a millisecond and you have to handle some percent extra current.

I believe that frequency is also variable so you'd need to sample that too.

ERCOT frequencies have sagged when the grid was stressed.

Reply on news.ycombinator.com