4 ms·

GWh/day unit analysis— Power * time / time Equivalent to just GW scaled by a constant, no?

s1artibartfast 2y ago
Correct, that would be intuitive. However, it is not what is being done here. When reported in press releases, battery capacity is not divided by day. Peak possible GW is reported independent of duration. This obfuscates the fact that this can only be provided for 1 or two hours
parineum 2y ago
No, it's not. The point of batteries is, typically, as a storage mechanism for renewables that doesn't have 100% uptime so, battery storage in itself doesn't have a defined GW used per day. The reason why I say it should be scaled to actual long term usage (preferably a calendar year) is that the capability of batteries to replace power plants is dependent on how much excess energy exists to charge them. For example, a TW of battery capability isn't going to do you any good in a Canadian winter if it's paired with solar that gets a few hours of generation a day. The comparison required to know how much fossil/nuclear generation these batteries are capable to replace is comparing the worst case scenario for the whole system, batteries and generation included. If the worst day for these batteries they discharge 2Gwh of power, they can only replace 2GW/24 of traditional power generation. The nuclear comparison is a bit ironic to me as well because one of the things that the batteries are actually very capable of is replacing the "peaker" generation that's required by nuclear/hydro/geothermal generation since it's difficult to scale those responsively. A Solar/Wind/(Nuclear|Hydro|Geo) generation backed by significant battery capacity could completely replace the need for fossil fuels in the generation side and, as renewable production and battery capacity increase, you can drop the rods a bit lower and scale back your nuclear production until it's zero. I see this article as clearly meant to refute the nuclear energy argument when, in practice, building battery capacity makes nuclear make more sense, not less.

Correct, that would be intuitive. However, it is not what is being done here. When reported in press releases, battery capacity is not divided by day. Peak possible GW is reported independent of duration. This obfuscates the fact that this can only be provided for 1 or two hours

parineum 2y ago

No, it's not. The point of batteries is, typically, as a storage mechanism for renewables that doesn't have 100% uptime so, battery storage in itself doesn't have a defined GW used per day. The reason why I say it should be scaled to actual long term usage (preferably a calendar year) is that the capability of batteries to replace power plants is dependent on how much excess energy exists to charge them. For example, a TW of battery capability isn't going to do you any good in a Canadian winter if it's paired with solar that gets a few hours of generation a day. The comparison required to know how much fossil/nuclear generation these batteries are capable to replace is comparing the worst case scenario for the whole system, batteries and generation included. If the worst day for these batteries they discharge 2Gwh of power, they can only replace 2GW/24 of traditional power generation. The nuclear comparison is a bit ironic to me as well because one of the things that the batteries are actually very capable of is replacing the "peaker" generation that's required by nuclear/hydro/geothermal generation since it's difficult to scale those responsively. A Solar/Wind/(Nuclear|Hydro|Geo) generation backed by significant battery capacity could completely replace the need for fossil fuels in the generation side and, as renewable production and battery capacity increase, you can drop the rods a bit lower and scale back your nuclear production until it's zero. I see this article as clearly meant to refute the nuclear energy argument when, in practice, building battery capacity makes nuclear make more sense, not less.