When my previous post was almost finished, I came across a video excerpt of a Wall Street Journal interview of Elon Musk. It seems to me that this short video presented his answer to some question on nuclear fusion:

[Elon Musk]
I think we already have a giant fusion reaction in the sky that is called the sun that shows up every day. It converts four and a half million tons of mass to energy every second and requires no maintenance. It is amazing. You don’t have to refuel it, you don’t have to maintain it, it’s just there. So my recommendation for fusion is solar power and batteries. And we can easily power all of Earth just with photovoltaics and batteries. I mean, not easily, but there’s just a very clear path to do so. And no miracles require, just work.
[Thorold Barker, WSJ]
Interesting.
[Elon Musk]
I am also an advocate of wind and of nuclear fission, geothermal, hydro and what not.

Basically, solar and batteries could power the entire world with no miracles required. I can surely understand the reasoning. The sun radiates out a massive amount of energy and I think there are some places where solar supported by batteries would make sense. That is however not the case everywhere. For example, solar is out of phase with demand at our latitude (and above) and these regions would therefore need seasonal storage. Solving that would not be straightforward. That is why I didn’t even do the effort to include solar electricity production in the comparison that I made in previous post.

I don’t know whether Musk actually entertained this idea or just said it proverbially, but it got me curious about how solar would fare when added to the comparison that I made in previous post.

The technique that I used in previous post to compare wind and solar plus wind comes down to tweaking the storage capacity until it would not draw empty over an entire year, meaning just enough storage to prevent blackouts. The idea was to find the smallest storage capacity that is needed in order to have something in storage at all times.
I will use the same set of assumptions that I used in previous post in this comparison:

• Belgian production and demand data from 2023
• A energy production of 1.5 times demand and 3.4 times demand
• The energy production would preferably be used directly
• Surplus production will be put into storage with a charging efficiency of 90%
• When there is a shortage, storage will fill in the difference with a discharging efficiency of 90%
• Surplus production will be curtailed when storage is full.

Let’s just jump in. This is how an uninterrupted production of 1.5 times demand would look like (from left to right: solar, wind, solar plus wind):

That doesn’t seem to be right at first glance. All three graphs show the exact same amount of production, but there seems to be a lot more solar production in that first graph. That is however not the case and this has to do with the resolution of these graphs. Let’s zoom in strongly and look at the details of the first week of the year:

Solar energy is exactly zero at night and, on average, there is no sun during roughly half of the year. The only way to still have the same production is going upwards between the gaps, especially at end of spring/summer/beginning autumn. That is why it seems that there is a lot more solar. One needs to enlarge that solar graph quite a lot to see what is happening, but that means not having an overview anymore, let alone comparing what is happening in those three graphs.

Everything above the blue line (demand) will or be put in storage or will be curtailed (when storage is full). Let’s have a peek at what is in storage over the year:

The storage requirement for solar is a whopping 21,250 GWh, which is waaaaaay larger than wind alone (5,000 GWh) or solar plus wind (1,900 GWh).

Another difference is that wind as well as solar plus wind are able to start from empty storage, but this is not possible for solar. There is not much solar electricity produced at the start of the year and this is also the period with the highest demand, so it would not be possible to get enough electricity in storage until the time of the year when solar production is ramping up. A 9,250 GWh head start ensures that storage doesn’t draw empty before that time (beginning of February).

In order to have the same production as wind and solar plus wind, I will also end with the same amount in storage. This will also ensure that there is enough in storage at the end of the year to get over the start of the following year.

This is what will be curtailed in the three scenarios:

Looking at those graphs, it seems that solar has much more curtailment than wind, but just as seen in the first figure of this post, that is again due to the resolution. The curtailment happens only during the sunniest hours of the day and for the rest of the day there isn’t curtailment. It would be necessary to blow up the graph in order to see that, but then there wouldn’t be an overview anymore.

Wind is more spread over the year, but more concentrated at the beginning and the end of the year. Solar plus wind show the most curtailment of the three scenarios.

Here are the numbers:

Let’s do the same for a production of 3.4 times demand. This is how uninterrupted production would look like:

Which has the same pattern as at 1.5 times demand, but has a much wider dynamic range (solar electricity production is between 0 and 220 at 3.4x demand compared to between 0 and 100 at 1.5x demand)

Everything above the blue line will be or stored or curtailed. This is what will be stored in the three scenarios:

Solar storage capacity (9,250 GWh) is now much lower than at 1.5 times demand (21,250 GWh), but again it is much higher than wind alone (1,550 GWh) or solar plus wind (500 GWh). There is much more direct use of the production, but this means much more curtailment and over a longer time span:

Here are the numbers:

Solar has slightly less curtailment than wind or solar plus wind at a production of 3.4 times demand, but that difference doesn’t mean much when about 70% of production needs to be curtailed.

Concluding. Solar has some serious drawbacks compared to wind and to solar plus wind. It has zero production during on average half of the year (at our latitude there is roughly 8 hours of sunshine per day in winter, roughly 8 hours darkness per day in summer and everything in between for the rest of the year), therefore it has higher peaks for the same production. It is also out of phase with demand at our latitude (more production when demand is low and vice versa). This will lead to a much larger storage capacity requirement in order to bridge that seasonal gap.

I can however see solar supported by batteries somehow working in regions with not much seasonal differences and where solar is more in sync with demand. However, looking at the scenarios above, it seems that quite some issues need to be resolved first to get solar working in countries with large seasonal differences. For example (in no particular order):

• Having a shitload of batteries to back up solar. But then, can enough materials be mined in order to produce the required amount of batteries?
• Be fine with (lots of) curtailment. But then, can a different incentive structure be made to make this profitable?
• Develop and incentivize systems that can handle intermittency, that just work when electricity is available and aren’t bothered when it is not available.
• Building interconnections over huge distances.
• …

These are considerable challenges, so I am curious what he exactly meant by the statement that there is a “very clear path” to power all of Earth just with photovoltaics and batteries.

The last sentence in which he also advocates for other power sources, even some conventional ones like nuclear energy, makes much more sense to m. I also think the energy landscape needs to be diverse and nuclear should not be excluded. Electricity production should certainly not be limited to intermittent power sources.

Solar (and/or wind) supported by batteries could surely make sense in some locations, but that the energy of the sun could potentially provide all of world’s energy needs doesn’t necessarily mean that photovoltaics supported by batteries is most efficient way to do it.

via Trust, yet verify

https://ift.tt/qJf0tSl

November 29, 2024 at 11:13AM