Well said, but as the article goes on, it seems even with no technical hurdles this innovation seems dead in the water with policy. At least they're not doing what the Germans have done, shutting down coal power plants and buying Polish electrical generation... made with coal power plants.
Oh this is the oldest trick in the book. If you need to greenwash, then just burn the coal somewhere else. Easy.
Los Angeles has been doing this for decades - for years the largest single energy source for LAPW has been an 1800MW coal burning plant that they operated in Utah, which has very loose environmental regulations.
Interesting. It is as bad as you say (21% coal), but after a decade of planning, coal is about to be phased out next year[0].
> The Agency planned to build the third unit of 900 MW capacity. This unit was expected to go online in 2012; however, the project was cancelled after its major purchaser, the city of Los Angeles, decided to become coal-free by 2020. [0]
> The plant includes a HVDC converter. It is scheduled in 2025 for replacement with an 840 MW natural gas plant, designed to also burn "green hydrogen."[0] (released by the electrolysis of water, using renewably generated electricity)
Yeah, I don't understand the use-case for hydrogen here. Why convert from electricity -> hydrogen -> electricity if everything is stationary? I suppose it could be useful for storage or long-distance transmission, but it seems like it would be much less efficient than other, simpler options.
The ACES project aims to use electrolysis to produce up to 100 metric tons of hydrogen per day, which will be stored in naturally occurring salt caverns at the site. The caverns have a potential storage capacity of 300 GWh of energy, according to Mitsubishi Power, which is developing ACES jointly with now Chevron-owned Magnum Development.
For comparison, last year the largest battery system in the world was the Moss Landing project in California with 3 GWh of capacity:
Do we need that much storage, though? Presumably we won't have the hydrogen-powered generation capacity to use that much energy quickly, so the comparison to utility-scale batteries isn't quite apples-to-apples.
From the article about the Chevron project:
> The project will initially provide fuel to the Intermountain Power Project, an 840-MW blended gas power plant also under construction in Delta, but Chevron believes there will be opportunities to supply hydrogen to the transportation and industrial sectors as well.
So even if the hydrogen storage facility was full, we're still limited to 840 MW of generation capacity. Sure, we get ~350 hours of runtime, but that's not really needed.
The Bath County Pumped Storage Station has 3003 MW of generation potential, with 11 hours of runtime from full.
Looks like the Moss Landing project is rated to be able to discharge 1/4 of its capacity per hour, so that 3 GWh facility can provide 750 MW. Batteries also have the advantage of being able to be sited much closer to the end user.
It's a demonstration project to show how a renewable powered system can cope with weeks of bad weather. If deep decarbonization doesn't actually require weeks of storage, not many systems like this will get built in the future. But if they are required at least we'll know how to build them.
Yes, but not for power generation. The majority of Australia's coal exports are metallurgical coal, for manufacturing. Australia owns about 58% of the global trade for metallurgical coal, which means everyone imports it from us.
This is a common talking point, but it really doesn't mean much in terms of the reliability of electric systems.
If you import a lot and export a lot, that doesn't mean both countries could be autonomous, it may mean your country strongly depends on imports sometimes, and on exports at other times, which typically is the case for Germany.
Because electricity is difficult to store, but because of how the grid works you always have to generate exactly as much as is used. Not electricity generation fluctuates (obvious with wind and solar; all plants need maintenance; nuclear often has to shut down in the summer either because rivers don't carry enough water or are so hot that feeding warm cooling water back into them endangers fish). Demand also heavily fluctuates both over the course of a day and with the seasons.
Trading electricity over geographically large areas smoothes out some of these fluctuations and gives you more options to deal with planned outages
> nuclear often has to shut down in the summer either because rivers don't carry enough water or are so hot that feeding warm cooling water back into them endangers fish).
I don't believe this has much impact on Germany's export patterns: in fact, summer is usually the time scheduled maintenance is planned in France, because electricity usage is much lower during summer than during winter. So there's still a lot of wiggle room before France ends up being forced to import in the summer because of that phenomenon.
The main driver for import/export patterns are different consumer patterns (not all countries have the same daily load curve - for instance, France and Spain both benefit a lot from trading because France's peak use time is an hour ahead of Spain's) and renewables availability.
Say it's a super sunny day, Germany generates more solar power than they use themselves (Germans don't usually have aircon at home anyway), Poland is happy to import and use it since it's basically free since it has no input costs of fuel, so they can idle their coal plants and use less fuel. Now it turns to night and the sun goes down and Germany generates 0 solar power. Then they can import coal power from Poland.
Expand this over a whole content and you get flows of electricity between companies depending on how the grid interconnect loads look and the prices of the various power sources. If Polish coal costs more then French nuclear, then Germany can sell their cheap wind power to Poland for something in between polish coal costs and French nuclear costs, and then buy nuclear from France and pocket the difference. Assuming the grid interconnects can't handle Poland importing direct from France.
Again none of these examples are real, just an explanation of why you would want to trade electricity and why you have a big market like the EU power market.
If you look at the electricity map at various times of day or over a week, you'll see the import arrows flip direction depending on the weather and time of day.
The UK exports wind power at night, but imports power during the day. Scandinavia often exports hydro power.
Taxes on carbon emissions mean Poland will import green power if it's available, as it will be cheaper than burning their own coal.
It's strange to see Germany so against nuclear, but happily buying all that clean nuclear energy from it's neighbour. Nuclear accidents don't care about borders, you'd think if they really meant what they preached they'd boycott their closest nuclear plants aswell
In bavaria, chernobyl was an event my parents generation fully experienced.
It was very clear (especially when you watch the old reports, i did a presentation about this in school) people didn't know a lot about nuclear and how to act upon an accident. Kids were not allowed to play on playgrounds, food had to be washed etc.
Even today if you shoot a dear, you have to check it for radiation!
its not strange.
And another reason why politics are stupid: Bavaria is reigned by the CSU. They have the majority for a very long time and the partner CDU was in power for over 16 years. None of them made any long term nuclear power strategy ever.
No one cared to plan longterm enough at all. Building nuclear is not easy and its a lot harder in a country like germany were we want to be extra save.
Even the newest europeon nuclear power plants take very long time. The last one took i think 18 years instead of 10?
> At least they're not doing what the Germans have done, shutting down coal power plants and buying Polish electrical generation... made with coal power plants.
And shutting down nuclear power plants and buying French electricity.. ..made with nuclear power plants. Strong the hypocrisy is with this one.
But hey they're doing their part, right? This is the flaw in European and international agreements, they're still done on a per-country basis instead of EU or a continental basis.
It’s short term measures while it transitions to a heavily renewable based energy system. Aren’t they currently at 60%+ renewable and are aiming to be 80% by 2030? Claims of greenwashing with those stats is disingenuous.
A part of the increasing of the renewable share in Germany electricity production is also the result of a decrease in total electricity production. In 2017 Germany was producing 560 TWh (with 138TWh solar + wind) vs 430TWh (with 192TWh solar + wind) in 2023 and probably even less this year. That's why Germany is importing electricity.
These are very informative numbers one reads nowhere in the news.
From destatis[0], we go from 608 in 2019 to 514TWh in 2023. So not the same absolute values, but same trend. Thank you.
We often read equivalent figures in the British news, as it shows the energy efficiency push of the last 30 years has reduced power use, leaving space for charging electric vehicles.
Suppose it's the same as online dating vs "meeting the old fashioned way".
I'm also pretty partial to "just work on your hobby and meet someone else". But probably a lot of opportunity left on the table that way. Especially if you know you want to build something serious.
Exactly. "All SAAS for b2b is just an SQL wrapper". That's until they do something distinguishing, make implementation easier/faster/cheaper or come up with a new technology. Yes the playing field is level now, but what can they do in 1,3,10 years? How will new models play into this?
I'd agree with much of the comments that there's a nebulous amount of moat, but certainly one exists. Pointing this out though, from the outside when AWS and Google's infrastructure was being built, was it obvious they were building significant moats?
Also looking at the counter-factual, would it be better for OpenAI just give up their momentum? Really they're choosing the most rational pathway they have, considering all of the partnership deals they've made. The real question is if you can tell they're picking up pennies for the Microsoft steamroller behind them or if they're going to keep revolutionizing. They've got the trackrecord, but it does seem like OpenAI is entering its late Summer-Fall season of growth.
