The United States gets almost 20% of its electricity from nuclear power. The facilities run greater than 90% of the time, with the less than 10% of downtime highly predictable. The almost 100 nuclear facilities have generated the lion’s share of CO2 free electricity in the United States, and – even with the massive growth of wind and solar – today generate nearly 50% of the CO2 free electricity in the country. Nuclear is definitely valuable and important, but this paper’s flaw is not in its fine tuned analysis – but the technology that it ignores.
Researchers as part of the MIT Joint Program on the Science and Policy of Global Change have published Deep Decarbonization of the US Electricity Sector: Is there a role for Nuclear Power? The document posits that in the various modeled scenarios below, that as the grid heads toward 90% CO2 reductions, nuclear power being available reduces the equivalent carbon tax from almost $120/ton to less than $40/ton by 2050 (in 2006 numbers). The basis of this is that as wind and solar break 40-45% of annual electricity generation, we’ll then see “curtailments, storage, and back-up capacity add to system costs” (in 2018 numbers).
The research puts the unsubsidized price of solar just under, and wind just above 5¢/kWh, with the subsidized number knocking about a penny per kWh off. Lazard’s LCOE analysis suggests these numbers a bit strong. The paper references the EIA in that advanced nuclear will costs between 7.6 and 8.4¢/kWh. The paper suggests pumped hydro as the energy storage choice available in the future, with no availability of flow/lithium ion type products (maybe they ought check out what Los Angeles just signed).
Note that in the Reference and Cap Scenario above nuclear heads towards almost nothing.
The paper’s crux is the intermittent nature of wind and solar power, and the need for dispatchable electricity sources that can balance out this nature. The logic is fundamentally sound as the power grid is a tool that must be looked at with safety and consistency being objective number one, as life and security is heavily dependent upon it. The chart below shows how the Cap scenario, in which wind and solar head over 60% of generation – before accounting for curtailment, leads toward the highest level of CO2 pricing.
The images shown in this article look at the aggregate model of the nation, whereas the paper shows that there is much greater complexity as they did also model the various mostly independent power grid regions of the nation (beyond the three main grids).
This author suggests that gas as the flexible source is the paper’s flaw. Researchers have the hourly electricity generation potential of wind and solar on the United State’s three unique power grids, and their research suggests that 80% solar+wind, would need about 12 hours – 5.4 TWh worth – of national electricity demand in energy storage. And very recently, NREL has looked at situations where solar power alone is greater than 50% of all electricity, and as you can see in the model below, the light blue of energy storage replaces the gas above quite nicely.
These prior models, and recent research seeing the price of energy storage going down – while volumes heavily expand – suggest that models need consider heavy and long term energy storage utilization. There are those in the business world who feel 24-7 solar might be coming much sooner than many expect, and our modelers need take some heed from those motivated business people.