In Grid-Scale Storage of Renewable Energy: The Impossible Dream, Energy Matters (November 20, 2017), Euan Mearns used a full year of data from England and Scotland, with one hour resolution, to calculate that to have firm power, it would be necessary to have 390 watt hours of storage per watt of average demand.
In Is 100 Percent Renewable Energy Possible? (May 25, 2018), Norman Rogers performed a similar calculation for Texas and concluded that 400 watt hours would be necessary.
I obtained data from the California Independent System Operator with one hour resolution from 1 January 2011 until 30 November 2020, and five minute resolution thereafter. The situation for California is much worse than Mearns calculated for Britain, or Rogers calculated for Texas. The details of the calculation method are explained in Storage Situation is Worse than Mearns and Rogers Calculated.
The graph shows the accumulated difference between generation by renewable sources, and demand, that is, how much would be accumulated into or discharged from storage, assuming average generating capacity is equal to average demand. The notional capacity of renewable generators is much larger than average capacity; the capacity factor for solar generators is around 25%, and around 30% for wind.
The “Equal Increase” line assumes all renewable generation sources are magnified by the same amount, so that all demand is satisfied by renewable sources. It is unlikely that biomass and geothermal can increase as much as solar and wind. Activists want to remove dams, so hydro might well decrease. The “Weighted Increase” line increases each renewable source by its average growth in output during the period of analysis, which depends upon weather conditions. The California Energy Department predicts total demand, but does not predict growth of supply, either in toto or separately by generation technology.
The storage capacity necessary to avoid blackouts is the difference between the maximum surplus and the deepest deficit. Between 1 April 2015 and 13 January 2023, the greatest surplus is 873 watt hours per watt of average demand on 27 July 2022, and the deepest deficit is 2003 watt hours per watt of average demand on 28 February 2017. To avoid blackouts, a capacity of 873 + 2003 = 2876 watt hours per watt of average demand would be necessary, precharged to 2003 watt hours per watt of average demand on 1 April 2015.
Activists insist that an all-electric United States energy economy would have average demand of about 1.7 TWe. Assume California average generating conditions from 2015 through 2022 apply to the entire nation, and therefore 2876 watt hours of storage per watt of average demand is adequate (this is optimistic). The total cost for Tesla PowerWall 2 storage units, not including installation, with 2876 × 1.7 terawatts = 4.89 quadrillion watt hours' capacity would be 4.89 quadrillion × $0.543 = $2.66 quadrillion, or about 133 times total US 2018 GDP (about $20 trillion). Assuming batteries last ten years (the Tesla warranty period), the cost per year would be 13.3 times total US 2018 GDP. The cost for each of America's 128 million households would be about $2,075,000 per month. This analysis assumes 100% battery charge and discharge efficiency. They're closer to 90% (81% round-trip), so the necessary capacity and cost would be about 25% more.
Update 14 March 2023: The same calculations were performed using nationwide hourly generation data provided by the Energy Information Agency since 2018. The situation is more optimistic but still impossible. 848 watt hours of storage capacity per watt of average demand would be required to provide firm power. The cost would be “only” about 3.8 times total GDP per year for batteries alone, or “only” about $49,776 per household per month. This is, of course, an optimistic estimate that assumes no calamity such as the 1815 eruption of Mount Tambora on Sumbawa Island in the Dutch East Indies, that gave us the “year without a summer” in 1816. Something similar will happen again. The only question is “when?”
Either way, Elon Musk would have more money than God.
If average renewables' generating capacity were to have been increased to three times average demand, and 12 hours' storage were provided, as is claimed to be sufficient by many environmentalists, power would have been available 97.5% of the time, i.e., blackouts 2.5% of the time. The industry definition for “firm power” is 99.97% availability, or about two hours and forty minutes per year without power. 2.32 gigawatt hours of output — 134% of total demand or 42% of average capacity — would have been dumped.
The cost for only twelve hours' storage, for an all-electric 1.7 TWe American energy economy, would be $11 trillion, or about $1.1 trillion per year. The cost per American household would be about $721 per month (for batteries alone). And electricity would still be available only 97.5% of the time.
Renewables provided 36.6% of California electricity between 2020 April 1 and 2022 December 15. Electricity satisfies about one third of total California energy demand. To provide all California energy from renewable electricity sources whose average generating capacity is three times average demand would require a capacity increase of 2460% above the capacity to satisfy all current California electricity demand. Increasing hydro at all, or increasing biogas, biomass and geothermal by 2460%, is unlikely. With solar and wind alone, blackouts are frequent.
Other analyses might reach different conclusions in detail, but probably not overall.
There is a significant problem with increasing generating capacity above average demand, or indeed building enough capacity. This was explained in The Great Green Energy Transition Is Impossible: The amounts of materials necessary to build the “technology units” demanded in the Great Green Reset dreamt of by the International Energy Agency simply do not exist. Six times more copper is needed than humans have so far extracted from the Earth, or about five times more than is known to be recoverable. The situation for other metals, such as nickel and cobalt, is worse.
The bottom line is that an all-renewable energy system is simply not possible. If it really is necessary to eliminate carbon dioxide emissions arising from energy production (and there are real scientists who have doubts about this necessity), the only way forward is to switch to natural gas in the short term, and nuclear power in the long term.
Fortunately, nuclear power is the safest thing humanity has ever done. In the entire civilized world, it is safer than Teddy Kennedy’s car. If the inherently unsafe, unlicensed, and incompetently operated Chernobyl reactor, built in a country that had neither safety culture nor licensing criteria is included, the death toll from nuclear power increases to 28. Nothing like the Chernobyl reactor will ever be built again.
Is anything safer than nuclear power?
Lovely analysis of the challenges of scale.