Solar + 4 hours of storage clocking in at $40/MWh in ideal conditions. Substitute for windpower and some worse conditions and nuclear energy still look awful from a cost perspective.
Yeah, people really underestimate the scale of the storage problem for solar. Energy usage peaks around 6pm[1], while the amount of energy solar can produce is trending towards zero. Without significant storage capability (i.e. enough to supply ~a days electricity), solar energy has a pretty limited use case in a real-time electric grid.
"But what led to wind power's sudden fall? Statistics officials said the weather was partly to blame.
A lack of wind from January to March this year sharply reduced the amount of electricity produced by Germany's wind turbines. In contrast, stormy weather in the first quarters of 2019 and 2020 sharply boosted the electricity produced."
The wind might go at night, or it might not, or it might not for months.
Wind speeds near the ground peak in the early afternoon. Higher altitude winds peak at night, but you need quite a large turbine to catch those. Sometimes there's no wind for days at a time.
Quoting from your link: "It is found that daily extreme wind speeds at 10 m are most likely in the early afternoon, whereas those at 200 m are most likely in between midnight and sunrise"
Ground is irrelevant; there's a reason why wind turbines are being put on top of tall towers.
I think you’re being glib, but the issue is that you can’t guarantee the alignment of sun and wind to match demand. Sometimes the wind blows at night, yes, but sometimes it does not. On a pure renewable grid, no wind at night means rolling brownouts or blackouts. This is particularly problematic during the summer when people want to run AC in order to sleep, and will also be an issue when people move to electric heat in the winter.
Long term we’ll use a mix of big grids, over provisioned renewables, and grid level storage to smooth this out, but we’re not there yet. Right now the choice is between a nuclear power plant during these moments, or a natural gas one.
Germany is closer to 50% coal if you look at actual demand being fulfilled instead of just production. In the linked image, the grey area is conventionals (mostly coal), the green area is exports of surplus.
I would expect that in most places this will shift to natural gas, given that natural gas is more economically viable than coal for new plants. But still, both are fossil fuels, so neither option is great.
It's all about the geographical decoupling with local storage and smart consumers to smooth the loads. For example tying the charging of your electric car while it's parked to the current price. Spatial and temporal arbitrage of energy, which nuclear is completely awful at.
Or using Swedish, Norwegian hydro and wind together with German, British and Danish wind. These links are on the same scale as nuclear reactors.
No, gas is not more economical everywhere. It is in places which have gas, but most of Europe doesn’t have it — it has coal, though. For most of Europe, betting on gas means putting themselves under Kremlin control for its basic needs.
The UK power grid has pumped hydro storage for decades to handle a huge tick in power demand every time the BBC goes to commercial break during a very popular program, as everyone simultaneously goes to make a cup of tea with their several-kilowatt electric kettle.
> Right now the choice is between a nuclear power plant during these moments, or a natural gas one.
False dilemma. You left out energy storage, of which there are multiple proven technologies.
Iron chemistry batteries are looking like the biggest win. Non-toxic, cheap, simple, easily scaled, easy to build and maintain, and the materials needed are bountiful...and there are working production systems right now.
> The UK power grid has pumped hydro storage for decades
Pumped hydro is a pretty good choice for grid level storage today, but it has some pretty severe terrain limitations. There’s lots of places that will never use it because the terrain won’t allow it.
Of course there’s the issue of the actual numbers. Pumped hydro can store a ton of power in terms of watt hours, but the peak output in watts is very low. UK pumped hydro can produce 2.8GW of power combined. Pretty impressive until you realize that that’s less than some single nuclear plants. Blayais produces 3.6GW, Cattenom produces 5.2GW, etc. etc. A typical modern reactor has a nameplate capacity of 1.3GW or higher; literally adding a single reactor to an existing power plant is equivalent to half of the UK’s pumped hydro capacity.
The big advantage of pumped hydro is that you can combine power production with water storage, which is good! But it is not capable of producing enough power to enable a 100% renewable grid, and probably never will.
> False dilemma. You left out energy storage, of which there are multiple proven technologies.
Huh? I “left out” energy storage? I literally said “grid level storage” in the full sentence, you just cut it out when you quoted me.
To reiterate: I didn’t “leave out” energy storage. I do not believe that we have enough energy storage yet to make fossil fuel plants unnecessary, especially as we push to electrify everything (transit, industry, etc.). One day we will be there, but currently we’re woefully short. It’s my assertion that nuclear power is a good way to bridge the gap while we keep building storage, because as of today we’re literally burning fossil fuels when renewables fall short of demand for whatever reason.
> Iron chemistry batteries are looking like the biggest win. Non-toxic, cheap, simple, easily scaled, easy to build and maintain, and the materials needed are bountiful...and there are working production systems right now.
Sounds great, how many gigawatt hours are installed, or are being installed now, and how much can/will they be able to produce?
We can also use real time price mechanism to match demand with supply intra day, energy intensive loads like water/home heating and ev charging can cheaply (vs cost of batteries connected to grid) buffer many hours without issue.
An observation of Griddy during the Texas disaster shows that we should be very careful in designing a real time price mechanism that’s applied to consumers. In the worst case this can take the form of passing all the risk and cost onto consumers who cannot handle this. This is particularly pernicious for HVAC, because during outages this puts consumers in the un-enviable position of deciding between their savings and not freezing/boiling to death.
Some loads can be moved, such as ev charging, but others cannot. The issue is that a lot of consumer load is less shiftable than you suppose. House heating (presumably we’d electrify this, because global warming) cannot be deferred for too long, and most households go through a hot water tank a day. These loads can be deferred for short periods of time, but proposing that people go without hot water or temperature control for even moderate periods of time is a political non starter. From an infrastructure standpoint it’s also worth mentioning how many people die and how many buildings get ruined by even a few days without power during severe weather events, which is often exactly when prices would rise.
A holistic view of the situation shows that while there is some smart grid stuff we can do, we still need to provide a guaranteed minimum worth of power generation no matter what. Ideally this would come from grid level storage so that we can run everything off renewables, but we’re just not there yet. The reality is that until we get there, the choice is between nuclear power and some other form of fossil fuel, typically natural gas. I think on the whole we’d be well served to commission one last set of nuclear power plants to get us through this crisis, and plan on decommissioning them in 2050 or so once grid level storage makes them obsolete.
The disaster scenario "no matter what" power can come from regulated backup turbines and emergency (fossil) fuel reserves I think. It's a separate/saparable subproblem. Just make the normal power companies pay a lot for tapping it, so there is disincentive to purpousefully use it.
That plan only works if we still have a functioning oil extraction and refinement system, which is something we’re obviously trying to get away from. Given how hard petroleum products are to store, these “just in case” power generation systems would require an entirely redundant but ready to operate fuel production process in the case renewables failed. It’s doable, but the cost would be absurd.
Probably easier to leave a bunch of nuclear power plants in standby, actually.
You can use fuels made from renewables for this too, was just thinking about the near future.
I'm somewhat pessimistic about large scale fossil fuel production ending in the world anytime soon. We'll have a long tail petering out period probably where the products get progressively more expensive, and if we avoid the mad max scenario, at some point get displaced by synthetic compatible fuels.
A well insulated modern construction can hold a temperature for a long time. Heating and cooling during energy availability and using your insulated building as a thermal battery of sorts is a real thing, and should be I think considered part of the solution here.
Yes, it should. But this drives up the cost of new housing when we’re already struggling to produce enough housing of any quality. As always, these things come with trade offs.
(For the record my house is very well insulated because I agree with you emphatically, but I also recognize that I have the material means to afford that insulation/construction cost that not everyone else is able to)
You can if you have decent insulation (which at least cold climates do). In fact people historically have taken advantage of cheaper night time electricity prices to shift heating to night time. Better insulation will become a more attractive investment than in the past.
> Night time is usually when heating is most needed.
If we are still talking about cold climates where houses are well insulated, it's a more convenient timing but not really needed. If you cut heating for half the night or even more, indoor temp won't drop too much even if outdoor temp is notably lower than day time. It might drop a few degrees as you are sleeping warmly tucked in, but that's not dangerous, and you can still heat your bedroom if you want to avoid even that slight adjustment.
From 2020 to 2021, Coal is up 8%, Wind is down 7%, Solar is down 1%.
Non-renewable in total is up 8% and renewable is down 8%.
If they can find a battery solution that actually works, then that would be amazing, but existing solutions clearly aren't cutting it.
While it might cost more, nuclear power is extremely reliable and offers energy independence, a form of national security that is vital with Russian pipelines strongly influencing the geopolitics of the region. I hope better batteries can offer them the same, because it's sad watching coal expand dramatically.
When it comes to consistent power generation, Germans can't just wait until next year. Wind power has it's ups and downs, which is why you need a lot more of it than coal or nuclear to meet the same demand.
There is no storage solution. The required amount of batteries would be enormous, well in excess of production capacity. The only feasible solution to even out throths is (natural) gas plants. Neither coal nor nuclear are quickly adjustable in output, when combined with wind/solar they lead to peaks of overproduction. This is the situation in Germany.
Nuclear is massive-grid-scale base load only in a world that does not need more base load nor massive grid-scale generation. Everything about nuclear is slow, difficult, expensive, dangerous. They are increasingly not handling climate change well. They are decade-scale projects that take a long, long time to become carbon neutral; we cannot wait that long.
* Solar and wind are here to stay and they mean we need fast-reacting spare capacity and storage. Nuclear power plants are very slow to react and because they are so monumentally expensive they have to run at as high capacity as possible, as much as possible. They are strictly base load.
* Siting a nuclear power plant is very difficult just in terms of geology. Siting a nuclear power plant is very difficult grid-wise as well, because they are only cost-effective at massive scale. Super-high-voltage DC transmission systems can help, but they only tack on more to the project cost. You can't just inject gigawatts of power anywhere you want. And it isn't just injection that is the problem. Nuclear power plants that are not producing power need massive amounts of electricity to get things like cooling pumps running or keep them running until everything is up to temperature and you can get the turbines up to temperature and speed. Then there's the matter of needing sufficient cooling - usually done via river, ocean, or lake. Except climate change and other factors are making those sources of cooling increasingly unreliable (for example: invasive species like zebra mussels have made life hell for a lot of power plants)
* Nuclear power plants require lots of highly trained people to design, operate, and maintain. More power plants means more of them. Training them up isn't a short term affair. Solar and wind require far less of all of this. And frankly, I have serious doubts about societal stability in 20-30 years, and nuclear power plants are not even remotely friendly to any sort of societal instability. Not just in terms of security, but upkeep. They have very complex, deep supply chain needs.
* Building nuclear power plants from the start of planning to grid synchronization takes a decade or two, and it then takes another decade or so for the plant to become carbon-neutral in part due to the massive amount of concrete they require. Right now, we need to be reducing carbon footprint as much as possible, as fast as possible. Not causing huge increases in carbon footprint that will only balance out well past catastrophic climate conditions.
* Nuclear reactor containment vessels can only be constructed by a small handful of facilities and their capacity is very limited, and by and large already spoken for. We can't just wave a wand and start building more reactors tomorrow. Or even in the next several years.
* Nuclear waste may be a "solved" problem tech-wise as nuclear power proponents are fond of saying, but reality is that nuclear waste is a huge problem. Even short-term storage is a problem, as demonstrated, again, by Fukashima where fuel cooling ponds caught on fire.
Time and time again we demonstrate that we are not responsible enough to handle nuclear power; we've had numerous military nuclear power disasters; the commercial ones haven't stopped, either. A "1st world" country, arguably one of the most technologically advanced ones around, repeatedly bumbled every aspect of Fukashima, starting with the plant's design, its maintenance and procedures, and the response to the incident. What was Japan's excuse?
How many Mulligans does nuclear power need?
You know what happens when a solar or wind power plant is incompetently designed or run? A bunch of people lose lots of money. You don't end up with thousands of square miles of land uninhabitable. You don't need people with years of training supervising a bunch of solar panels. Maintenance on a wind turbine is a standard-industrial-equipment sort of job, no bunny suits required.
You know what happens when a country with solar or wind power has a government that is full of incompetent suit-stuffing chair-warming morons, or gets taken over by a despot dictator, or has an economic collapse? Nothing.
If you want to look to the future in power grids, look at the iron chemistry liquid batteries that are non-toxic and almost trivial to deploy at electrical substations. They can provide spare capacity at the neighborhood/regional level while helping balance distribution loads and allow those neighborhoods to continue to function in isolation in the event of transmission grid problems.
Nuclear is dangerous and needs to be treated as such.
Flying is also dangerous and is treated as such.
Just like the aviation industry has had an excellent track record in managing the danger, so has the nuclear industry. Just look at the number of victims of nuclear in the past 30 years. One (1) dead at Fukushima.
Chernobyl killed thousands, but it's as relevant to the safety of the industry in 2021 as a 1950s Antonov is to an Airbus.
We're not talking about building _new_ plants here. We're talking about maintaining the current fleet. Is this expensive? Yes. But what's the alternative? Burning coal and importing gas? This should simply be the cost of transitioning off fossil fuels. Once renewables are scalable and reliable, there will be no reason to build nuclear plants. Progress over perfection.
> But what's the alternative? Burning coal and importing gas?
Energy storage. See my last paragraph.
It's where a considerable amount of research and capital investment has been going and there are already numerous systems in use. Tesla's grid-scale batteries, pumped hydro around the world in numerous locations including the US, UK, Australia, and Europe.
Iron chemistry liquid batteries are in private commercial and grid use right now, likely to see widespread adoption in the coming years. It's cheap to manufacture, nontoxic, easily serviced, easily scaled.
Also, you can phone up your local solar installer today and get a battery storage system for your house or business. Some require a grid, others have transfer switches and will happily run off-grid. This has been the case for a number of years now.
Thank you for taking the time to go into the details on this! I‘m still sitting on the fence on this issue, because it‘s obviously massively complex and one rarely hears a decently argued take on it (and I don‘t have time to study nuclear engineering and electricity economics). But I think your comment helped me to understand just a little bit more about what‘s involved…
> Everything about nuclear is slow, difficult, expensive, dangerous.
While this might not be true for the latest generation of reactors in development (of which ofcourse no failures are know because they don't exist) it applies well to the current nuclear infrastructure Germany has.
