3 ms·
Base load isn’t an inherent advantage it’s strictly worse than dispatchable power unless it’s cheap, and Nuclear isn’t that cheap. Which is a problem because
by Retric 5y ago
Base load isn’t an inherent advantage it’s strictly worse than dispatchable power unless it’s cheap, and Nuclear isn’t that cheap. Which is a problem because excess wind and solar makes Nuclear power economically worthless for much of the day. https://en.wikipedia.org/wiki/Dispatchable_generation#/media/File:Renewables_need_flexible_backup_not_baseload.png https://en.wikipedia.org/wiki/Dispatchable_generation#/media...
Subsides could make up the difference for nuclear as could vastly increased nighttime prices.
PS: It’s only when your looking at islands like say Hawaii where you want that kind of massive overproduction. Across larger areas you never see 100% production from everywhere at the same time, but you also don’t see minimum production from everywhere at the same time.
- criley2 5y agoThe LCOE for utility scale solar is around $68/MWh while new nuclear is $88/MWh. The LCOE for grid storage of power (necessary addition to solar) is $160/MWh. Double solar with grid storage is as much as $300/MWh at the utility scale using globalized averages reported to NEA, BNEF, Lazard, IRENA 2018-2020. Obviously some countries will be a lot more or less on costs depending on their circumstances and location. Interconnect isn't enough, all of the US is in darkness for several hours a day. (And Texas will never connect). You require storage, and I still question if 2X is enough to produce enough power during a 10 year minimum without firing up extensive hydrocarbon peaker plants.
- Retric 5y agoNuclear is nowhere near that cheap, LCOE even with massive subsides in the US sees 2020 nuclear well past 160$/MWH and rising. Wind and solar are both under 50$/MWH LCOE. https://en.wikipedia.org/wiki/Levelized_cost_of_energy#/media/File:20201019_Levelized_Cost_of_Energy_(LCOE,_Lazard)_-_renewable_energy.svg https://en.wikipedia.org/wiki/Levelized_cost_of_energy#/medi...
- criley2 5y agoThe number I quoted is from the same Wikipedia article on LCOE that you used, and I used a globalized average as I said in my post. Your $50MWh also does not include storage LCOE, and not including secondary costs for solar when Nuclear has no secondary costs is considered "a trick" to lie about solar costs. Include the costs of gigawatts of Elon's lithium ion batteries, and include the cost of having to 2X-4X solar compared to Nuclear to account for variable solar output due to days and seasons, as I said, and your $50MWh becomes around $300MWh
- Retric 5y agoFirst demand isn’t flat, so 100% nuclear needs massive amounts of storage to reduce the number of nuclear power plants that would sit idle 90+% of the time. In the end something needs to take up the slack and nuclear doesn’t work by it’s self. If natural gas isn’t an option the only thing left is hydro which is limited or storage. Your also trying to cherry pick numbers the Wikipedia article also lists “$164/MWh” for nuclear. I found the table you got those numbers from which included multiple estimates from 2018 that used data from 2015 and double counting very optimistic nuclear estimates. Frankly utility solar is already being installed at sub 2c/kWh in the US. 68$/MWh would be 6.8c/kWh which shows how far off your umbers are. Anyway, I would be happy to see the power plant actually run that cheaply but your looking at very optimistic estimates for something that isn’t even online yet and is getting some serious subsides. Having said that if you’re data is from “Source: IEA/NEA, 2015.” https://www.oecd-nea.org/ndd/pubs/2018/7441-full-costs-2018-es.pdf https://www.oecd-nea.org/ndd/pubs/2018/7441-full-costs-2018-.... I can understand why you picked those numbers.
- pfdietz 5y agoYeah. Si monocrystalline PV modules fell in price by a factor of 4 from 2015 to 2020. So many people talking about energy seem to be stuck in an obsolete past.
- megaman821 5y agoYou are being unfair in the other direction. LCOE for wind/solar needs to be blended by figuring out the optimal mix of wind and solar for your area and adding on 4hrs of battery storage. There is not one set number for this but $200Wh and falling is a fair estimate.
- DennisP 5y agoThose interconnects take a long time to build, mostly due to political opposition, and aren't that cheap either. For example, the 2GW 400-mile Southern Cross has turned out to be a 17-year project, costing $2 billion. https://www.eenews.net/articles/what-a-2b-texas-project-says-about-u-s-quest-for-co2-free-grid/ https://www.eenews.net/articles/what-a-2b-texas-project-says... They're not the only ones: > the 732-mile TransWest Express high voltage transmission line filed its first permit application in 2007, but did not receive all of the approvals until 2020. The Environmental Impact Statement alone took eight years to complete. Local landowners tend to object to these projects. On average, they take over a decade to complete and that includes much shorter lines, which are quicker. Building underground can help with political resistance but multiplies cost by 5X to 10X. As of 2020, over 750 GW of proposed generation, most of it wind and solar, were waiting for transmission to be built. https://www.belfercenter.org/publication/challenges-decarbonizing-us-electric-grid-2035 https://www.belfercenter.org/publication/challenges-decarbon... China builds transmission a lot faster, and we might be able to speed things up with the SITE Act, which gives the federal government authority to override local and state resistance: https://www.theatlantic.com/science/archive/2021/07/america-is-bad-at-building-power-lines-lets-fix-that-transmission-climate/619591/ https://www.theatlantic.com/science/archive/2021/07/america-... But if our solution set can include unpopular new laws to override NIMBYism, then we can make nuclear a lot faster and cheaper, too.
- Retric 5y agoThat’s a solid point, but it only needs 3-4 years to build. “$2 billion project, proposed by Pattern Energy Group LP, likely won’t be ready until 2026 at the earliest, if construction starts in 2023.” Nuclear also takes 4-6 years to build but can take decades to get permission to start.
- DennisP 5y agoOf course. I used "complete" in a later sentence, which is really what I meant, and why I talked about political opposition being the main issue.
- ncmncm 5y ago