5 ms·

Estimates vary. Others say that 12 hours of storage would be needed to reach 80%: https://pv-magazine-usa.com/2018/03/01/12-hours-energy-storage-80-percent-wind

Estimates vary. Others say that 12 hours of storage would be needed to reach 80%: https://pv-magazine-usa.com/2018/03/01/12-hours-energy-storage-80-percent-wind-solar/ https://pv-magazine-usa.com/2018/03/01/12-hours-energy-stora... And for what it's worth, even just 5 hours of electricity storage is 12,500 GWh, compared to just 400 GWh of battery production (most of which goes to EVs and electronics, not grid storage). Pumped storage does indeed require rivers or lakes to operate effectively. Transporting vast amounts of water to a storage site is prohibitively expensive. It also requires a hill nearby on which to build a reservoir. And this site needs to be near a transportation hub, otherwise construction costs will increase as transportation of material gets more expensive. The number of potential storage sites is indeed large, but the number of economically viable storage sites is much narrower. Most studies regarding hydroelectric storage potential work by running an algorithm over a height map. Very few of the sites flagged by such an algorithm can be developed economically.

pydry 4y ago
>Estimates vary. Others say that 12 hours of storage would be needed They do. The study I cited above used real time production data rather than synthesizing it from weather data like yours. >And for what it's worth, even just 5 hours of electricity storage is 12,500 GWh, My calculations say 2280 GWh in the US. This says 35,000 GWh of potential storage is available: https://www.nrel.gov/docs/fy22osti/81277.pdf https://www.nrel.gov/docs/fy22osti/81277.pdf. 6% of that would be enough. >compared to just 400 GWh of battery production (most of which goes to EVs and electronics, not grid storage). Skeptics said the same thing about solar production in 2012. It'll probably similarly scale up exponentially (as will lithium production) just coz the unit costs are so attractive. Yes, we will probably want to use it mostly for electric cars and use pumped storage as grid storage coz you can store a ridiculous amount. Snowy 2 is a "battery" with 350GWh of storage all by itself. China has one that isnt much smaller. The US can probably handle building the equivalent of 6 and a half of one of those. For what pumped storage can't handle we can synthesize and store gas and we'd still save ~10% compared to nuclear power's abysmal economics. >Very few of the sites flagged by such an algorithm can be developed economically. 6% is apparently enough to get us to 97%. I'll bet we could manage plenty more though. Plus there are disused mines.
- Manuel_D 4y ago
  > This says 35,000 GWh of potential storage is available: https://www.nrel.gov/docs/fy22osti/81277.pdf https://www.nrel.gov/docs/fy22osti/81277.pdf. 6% of that would be enough. Again, this study just ran an algorithm over a heightmap. Most of the hydroelectric potential exists in remote mountain ranges that would be impossible to develop in practice. > For what pumped storage can't handle we can synthesize and store gas and we'd still save ~10% compared to nuclear power's abysmal economics. Unfortunately, no we can't. The Sabatier process (aka "synthesized gas") requires sources of hydrogen and carbon dioxide. The former is almost entirely produced through steam reformation, as electrolysis is not efficient or cheap enough. The latter is present in the atmosphere but in concentrations too small to be useful. The Sabatier process is opportunistically used with concentrated CO2 byproducts from industrial processes. But you can't just build a synthetic gas plant and use it as a battery.
  - Schroedingersat 4y ago
    What are you proposing instead?
    - Manuel_D 4y ago
      Solar and wind are good to supplement a dispatchable source like fossil fuels or hydroelectricity. But they are not viable as a primary producer without storage. And we don't have the capacity to build storage at anywhere near the required scale. It's often said that storage hasn't been developed because until renewables became significant there wasn't any demand for energy storage. This is untrue, energy storage has been attempted in order to smooth out fossil fuel use for the better part of a century. Sites at Turlough Hill [1], Ludington [2], and Bath County [3] were built to smooth out energy load on coal fired power plants. But these sites were expensive to develop and couldn't scale because of geographic limitations. They also were used for temporary daily demand, not the seasonal fluctuations of wind and solar that last for months. Energy storage simply isn't within the capacity to build, unless some breakthrough in technology happens. Until that breakthrough happens, it's not viable. And if you're willing to accept plans contingent on an engineering breakthrough, then just use fusion. So how do we decarbonize the grid? We can look at how countries that have successfully driven their fossil fuse to near zero and examine how they did it. Countries like Norway, Brazil, and Tajikistan use hydroelectricity for almost all their generation. This is great, but is geographically dependent. We should definitely built it where possible, but it's not a universal option. Same deal with Iceland and geothermal power. France was also able drive fossil fuel use below 5%. Only 10% of its electricity comes from hydroelectric power, since most of country is on the North European plain (flat land = no dams). They made up the rest through serial production of nuclear power. They built multiple copies of the same nuclear plant designs [4]. Within 15 years the country had over 80% of its electricity coming from nuclear power. The United states was actually on a similar track during the 1960s and 70s. Nuclear plants were being built at scale, and being delivered for costs between $1-2 billion per GW of capacity. These plants were actually some of the cheapest sources of energy amortized over the entirety of their lifetime. Overreaction from 3 mile island led to reduced nuclear construction. This diminished benefits from economies of scale, and led to higher nuclear costs. Vogtle costs over $20 billion because it's one of only (the only?) reactor under construction, and no economies of scale exists. Plenty of parts like steam generators [5] have no demand, save for nuclear power plants, so instead of serial production of a run of 30 or 40 it's only an order of 3 or 4. This is less efficient, and leads to larger costs. 1. https://en.wikipedia.org/wiki/Turlough_Hill_Power_Station https://en.wikipedia.org/wiki/Turlough_Hill_Power_Station 2. https://en.wikipedia.org/wiki/Ludington_Pumped_Storage_Power_Plant https://en.wikipedia.org/wiki/Ludington_Pumped_Storage_Power... 3. https://en.wikipedia.org/wiki/Bath_County_Pumped_Storage_Station https://en.wikipedia.org/wiki/Bath_County_Pumped_Storage_Sta... 4. https://en.wikipedia.org/wiki/Nuclear_power_in_France#Messmer_Plan https://en.wikipedia.org/wiki/Nuclear_power_in_France#Messme... 5. https://en.wikipedia.org/wiki/Steam_generator_(nuclear_power) https://en.wikipedia.org/wiki/Steam_generator_(nuclear_power...

pydry 4y ago

>Estimates vary. Others say that 12 hours of storage would be needed They do. The study I cited above used real time production data rather than synthesizing it from weather data like yours. >And for what it's worth, even just 5 hours of electricity storage is 12,500 GWh, My calculations say 2280 GWh in the US. This says 35,000 GWh of potential storage is available: https://www.nrel.gov/docs/fy22osti/81277.pdf https://www.nrel.gov/docs/fy22osti/81277.pdf. 6% of that would be enough. >compared to just 400 GWh of battery production (most of which goes to EVs and electronics, not grid storage). Skeptics said the same thing about solar production in 2012. It'll probably similarly scale up exponentially (as will lithium production) just coz the unit costs are so attractive. Yes, we will probably want to use it mostly for electric cars and use pumped storage as grid storage coz you can store a ridiculous amount. Snowy 2 is a "battery" with 350GWh of storage all by itself. China has one that isnt much smaller. The US can probably handle building the equivalent of 6 and a half of one of those. For what pumped storage can't handle we can synthesize and store gas and we'd still save ~10% compared to nuclear power's abysmal economics. >Very few of the sites flagged by such an algorithm can be developed economically. 6% is apparently enough to get us to 97%. I'll bet we could manage plenty more though. Plus there are disused mines.

Manuel_D 4y ago
> This says 35,000 GWh of potential storage is available: https://www.nrel.gov/docs/fy22osti/81277.pdf https://www.nrel.gov/docs/fy22osti/81277.pdf. 6% of that would be enough. Again, this study just ran an algorithm over a heightmap. Most of the hydroelectric potential exists in remote mountain ranges that would be impossible to develop in practice. > For what pumped storage can't handle we can synthesize and store gas and we'd still save ~10% compared to nuclear power's abysmal economics. Unfortunately, no we can't. The Sabatier process (aka "synthesized gas") requires sources of hydrogen and carbon dioxide. The former is almost entirely produced through steam reformation, as electrolysis is not efficient or cheap enough. The latter is present in the atmosphere but in concentrations too small to be useful. The Sabatier process is opportunistically used with concentrated CO2 byproducts from industrial processes. But you can't just build a synthetic gas plant and use it as a battery.
- Schroedingersat 4y ago
  What are you proposing instead?
  - Manuel_D 4y ago
    Solar and wind are good to supplement a dispatchable source like fossil fuels or hydroelectricity. But they are not viable as a primary producer without storage. And we don't have the capacity to build storage at anywhere near the required scale. It's often said that storage hasn't been developed because until renewables became significant there wasn't any demand for energy storage. This is untrue, energy storage has been attempted in order to smooth out fossil fuel use for the better part of a century. Sites at Turlough Hill [1], Ludington [2], and Bath County [3] were built to smooth out energy load on coal fired power plants. But these sites were expensive to develop and couldn't scale because of geographic limitations. They also were used for temporary daily demand, not the seasonal fluctuations of wind and solar that last for months. Energy storage simply isn't within the capacity to build, unless some breakthrough in technology happens. Until that breakthrough happens, it's not viable. And if you're willing to accept plans contingent on an engineering breakthrough, then just use fusion. So how do we decarbonize the grid? We can look at how countries that have successfully driven their fossil fuse to near zero and examine how they did it. Countries like Norway, Brazil, and Tajikistan use hydroelectricity for almost all their generation. This is great, but is geographically dependent. We should definitely built it where possible, but it's not a universal option. Same deal with Iceland and geothermal power. France was also able drive fossil fuel use below 5%. Only 10% of its electricity comes from hydroelectric power, since most of country is on the North European plain (flat land = no dams). They made up the rest through serial production of nuclear power. They built multiple copies of the same nuclear plant designs [4]. Within 15 years the country had over 80% of its electricity coming from nuclear power. The United states was actually on a similar track during the 1960s and 70s. Nuclear plants were being built at scale, and being delivered for costs between $1-2 billion per GW of capacity. These plants were actually some of the cheapest sources of energy amortized over the entirety of their lifetime. Overreaction from 3 mile island led to reduced nuclear construction. This diminished benefits from economies of scale, and led to higher nuclear costs. Vogtle costs over $20 billion because it's one of only (the only?) reactor under construction, and no economies of scale exists. Plenty of parts like steam generators [5] have no demand, save for nuclear power plants, so instead of serial production of a run of 30 or 40 it's only an order of 3 or 4. This is less efficient, and leads to larger costs. 1. https://en.wikipedia.org/wiki/Turlough_Hill_Power_Station https://en.wikipedia.org/wiki/Turlough_Hill_Power_Station 2. https://en.wikipedia.org/wiki/Ludington_Pumped_Storage_Power_Plant https://en.wikipedia.org/wiki/Ludington_Pumped_Storage_Power... 3. https://en.wikipedia.org/wiki/Bath_County_Pumped_Storage_Station https://en.wikipedia.org/wiki/Bath_County_Pumped_Storage_Sta... 4. https://en.wikipedia.org/wiki/Nuclear_power_in_France#Messmer_Plan https://en.wikipedia.org/wiki/Nuclear_power_in_France#Messme... 5. https://en.wikipedia.org/wiki/Steam_generator_(nuclear_power) https://en.wikipedia.org/wiki/Steam_generator_(nuclear_power...