25 ms·
With 100% grid-supplied power, changes in load are driven by slow-moving, predictable systems like sunrise/sunset, weather, and seasons. If you are running you
by gngeal 14y ago
With 100% grid-supplied power, changes in load are driven by slow-moving, predictable systems like sunrise/sunset, weather, and seasons.
If you are running your own solar array, though, you will vary your grid load on much shorter time scales
I'm not so sure that these latter variations are significantly less predictable than what you mention in the former paragraph. Isn't the output of a solar array dependent on the insolation? Isn't the insolation dependent on the cloud cover? Can't you predict the cloud cover in any single place simply by taking advantage of real-time meteo satellite data? A similar feedback could be established for wind power. Given enough data, I'm reasonably certain that models could be established that would allow you to predict how the solar and wind power generation distribution is going to change in the next hour(s) so that you could prepare for it.
- akiselev 14y agoThe problem isn't the prediction, full solar is probably only slightly more volatile than full grid on a large scale. Maybe not hourly but the companies would still be able to work out the supply side. The issue is that the supply side and supporting infrastructure is built for an entirely different system where power leaves the power plants and goes through the grid to consumers. Now people are adding solar panels which generate a ton of electricity during the day (when everyone is largely at work/school) that must then be fed back into the grid, opposite the direction of normal flow.