Top Solar Power Plants | Hacker News Reader

Top Solar Power Plants

solarpower.guide

21 pointsInd00711 comments

A solar thermal power plant may also be referred to as a solar photovoltaic power plant.

Uh, no? According to my 6-year-old's copy of "Stuff You Should Know!" [1] those are, like, opposites? Very confusing.

[1] https://www.amazon.com/Stuff-Should-Know-Howard-Hughes/dp/17...

What I find noteworthy and maybe isn't appreciated enough is how many of these large solar installations are in countries like India or China.

You may implicitly think that these would be in the richest countries, but apparently they are not. (Granted China and India have good conditions for Solar, but so do parts of the US and Australia.)

As of 2020, India has about half as much solar generation capacity installed as the US.

The question is why fewer, very large facilities are preferrable in India. Economics (land or labor)? Population density/distribution? Or is it political reasons?

Good question. I do suspect it related to the value of land in proximity to major market, and command economy vs capitalism. Obtaining thousands of contiguous acres averaging 240 Wh per m^2 per day, close enough to demand to efficiently deliver it, may face tough competition with other real estate users. Couple that with actually having to compete, vs favorable eminent domain laws, and compelling interest in pollution reduction, and China & India very well may see larger installations.

Ballpark numbers with my own utility solar point to revenue of ~$50/000 per year per acre (paying 15¢/kWh, 50% above normal power rate). Equipment costs may be about $1,000,000 per acre ($250 per 1m^2 panels etc, 4000 square meter panels per acre). That’s 20 years just to break even on equipment. Now work in land cost (and making it viable) vs market rate per acre.

Lately I have been surprised at how much solar and wind play the main role in most reports building pathways to carbon neutrality.

I always believed that the intermittency of solar and wind would be too problematic and that's why people advocated for nuclear energy. However in most recent pathways stamped as "realistic", it seems that nuclear stays pretty much at its current level and solar and wind grows exponentially.

Is intermittency and dispatchability a problem we can really overcome with DC long transmission lines, synthetic gaz, smarter energy consumption and batteries?

I have a hard time understanding to which extent each of those solutions can help solve the dispatch problem. I would love to be pointed out to some good up-to-date resources that go in-depth into that topic.

(Extremely) long distance transmission lines are theoretically useful because nighttime for you is daytime 12 timezones away; and even in the same time zone, winter for you is summer on the other side of the equator.

I don’t know what the cost of the stuff is, nor the geopolitics, but the losses are low enough (3.5%/1000km [0]) to at least be worth asking that sort of question — if the answer is “that’s fine”, you only need to store energy for non-static use, like phones and cars.

Batteries may or may not be a viable solution. On the “it’s fine” side, if you can electrify every car, then you are close to the production level needed to get enough batteries even for a no-intercontinental-transmission-winter scenario (83 kWh is a large Model 3, that per American household is fine for summer but marginal for winter). On the “perhaps not” side, there is no guarantee that the impressive current growth of batteries will last long enough to get us there.

Synthetic gas and smarter energy consumption: I know far too little to comment.

[0] https://publications.jrc.ec.europa.eu/repository/bitstream/J... page 11

Nukes are not the solution because they are mature tech. They are very expensive, always have been, always will be. Nuke plants have to be big to be worth building, but like any big public works project, they attract corruption. Most of the price of a typical nuke plant in the US is (100% legal!) graft. That won't change either. It takes many years to build a nuke plant because nobody involved wants the gravy train to stop. Finishing means it stops.

(A fusion plant would need to be enormously bigger than any fission plant, so would attract enormously more corruption, and probably never be finished.)

Solar and storage prices are still in free fall. Battery tech optimized for utilities rather than phones and laptops is coming to market at 1/3 of lithum's price, and will fall from there.

There is no lower limit on useful size for solar. It can start producing useful power immediately while you add capacity. It is more valuable with storage, but works without. The more storage you add, the better it is, but even a little storage pays for itself immediately.

It don't really care about nameplate power let alone the area covered. Capacity factor is really not relevant either as it doesn't factor in the instantaneous market demand for power. What I'd really like to see is some kind of metric describing how much of power-demanded was actually covered by the facility at the right time.

This would make it much easier to compare intermittent energy sources with more stable ones. We always end up with MWh/$, but it feel that is hiding the true cost of solar and wind.

Solar/wind with plenty of storage is the future.

At the moment, generating-capacity without storage is equally as useful as with, so putting money into storage now would be wasteful, vs. getting correspondingly more absolute generating capacity online: each panel added displaces an exactly-known rate of carbon release.

Storage costs are plummeting even faster than solar generating capacity, so the longer you put off building out storage, the cheaper it will be when you do. By the time solar generating capacity is large enough that its variability could be a problem, storage will be mature and cheap, and storage buildout will ramp up. In the meantime, just a small amount of storage can take away all the profit from gas plants.

The largest Chinese plant is a lot smaller than the largest Indian plant, but produces nearly as much power - any insight why?

The data seems to be mixing a small portion of the Golmud Solar Park land mass with its latest power generation (2.2 GW). This article [1] says:

"With a capacity of 2.2 GW ac (3.1 GWp), the new plant in Qinghai’s Gonghe County is the world’s largest single solar installation. Covering over 52 sq km, it was built by Huanghe Hydropower in just 10 months and connected in September 2020."

So they're basically the same "size", though both have a lot of gaps (see pictures at [2]). Seems like the solar folks draw a convex hull around the area to use as a guide. The map views at [3] and [4] will let you decide for yourself :).

[1] https://renewablesnow.com/news/worlds-largest-solar-plant-ta...

[2] https://www.wiki-solar.org/library/papers/2SolarParks_v2.pdf

[3] https://wiki-solar.org/map/sites/index.html?Golmud_Solar_Par...

[4] https://wiki-solar.org/map/sites/index.html?Bhadla_Solar_Par...?

Reply on news.ycombinator.com