2 ms·
Building a line to run @ only 25% capacity isn't exactly super smart. We have these conversations all the time in grid operations and planning when we hear abou
by 4thaccount 8y ago
Building a line to run @ only 25% capacity isn't exactly super smart. We have these conversations all the time in grid operations and planning when we hear about these big projects. In power transmission you always have to carry power reserves for your single greatest contingency (actually more). Let's say your largest contingency today is the loss of a 900 MW nuke plant. So you have to always have an extra 900 MW laying around. Now go build a 4 GW line. What happens if it trips? Big changes around the regions on each side. Chances are you now have to carry considerable more reserves unless you operate it at a much lower percent of capacity. It's the same with large amounts of wind. If you suddenly see your wind forecast fall off a cliff and need to make up multiple GW of lost generation, you're in a really risky spot. You have to carry extra reserves and try to best anticipate the event and that also costs money. In a future with lots of big batteries, this isn't as problematic, but we're very far from having that much storage on the grid.
- fspeech 8y agoWhich is why they need a network of the grids like the local ones in existence today. If you have 10 lines feeding into a mega city from different directions the reserve for a single failure can be lowered to 10%. You can also build pumped water hydro stations, which I don't see how batteries could ever match in capacity or efficiency. Batteries are good for smoothing out spikes but how can they deal with a failure that could take some time to resolve?