Vast solar power plant at the Moroccan desert | Hacker News Reader

Vast solar power plant at the Moroccan desert

bbc.com

45 pointsoori30 comments

160MW peak. 42MW average over a year. That ranks about 25th in the world for solar power installations. Typical nuke or coal plant is about 1000MW per unit peak, and average is slightly less. Three Gorges Dam, 22,000MW peak. This thing is not going to "power Europe".

[1] https://en.wikipedia.org/wiki/Ouarzazate_Solar_Power_Station

To be fair, something like this will upgrade as the tech improves, something that alternatives cannot do. But yeah, it's certainly not the definitive answer.

Sure you'll be able to upgrade this with more efficient turbines or something, but you'll also be able to put those more efficient turbines in a hydro-electric dam or wind turbine, or coal burning plant. I'm not seeing the magical potential for improvements in concentrated solar that you wouldn't get from any other type of generator.

How? This isn't photovoltaic. It's just a solar-powered heat engine. That's a mature technology.

Hyperbole aside, the Sahara does have such humongous solar potential that supplying Europe is a possibility... But I never hear anyone mention the mightily expensive necessary upgrades to the electricity grid - and more specifically the extremely long HVDC power transmission lines to the consumer regions...

Why would this necessitate upgrades to the electricity grid and HVDC in particular? This plant's electricity is turbine-generated, just like in regular coal, oil, gas, and nuclear power plants.

The issue is that currently the power is generated locally in Europe. In order for Morocco to export it's power to Europe, it would need the HVDC (high voltage direct current) power lines installed to cross the Mediterranean Sea.

I don't follow the logic here:

1. HVDC is not a strict requirement for submarine power lines. AC works just fine, even if less efficiently. Efficiency losses are generally offset by lower capital costs. Remember that Gibraltar strait is only 9 miles wide. In fact, there already exists a 16 mile-long 400kV AC interconnect between Morocco and Spain now.

2. Even if HVDC makes more economic sense for transmission, it won't impose any sort of "electric grid upgrade" requirement on Europe or Morocco. The HVDC line would interface to the existing grid via a HVDC converter; these are built at GW scale. There's already plenty of HVDC submarine power lines within Europe now. (Britain-Netherlands, Norway-Denmark, etc).

You may argue that new infrastructure will have to be built to distribute the added capacity, but that's a different argument.

It's very unlikely that importing sun energy from Africa is cheaper than producing it domestically

There is no need, Europe is getting on average the same amount of daylight hours as the Sahara.

Temperature isn't a factor, even overcast isn't that much of a factor, especially for residential installations.

Thermal is more affected by overcast due to the light being more diffused but still doable, southern Europe will be good for thermal solar, all of Europe will be good for PV, and central and northern Europe is very good for hydro and wind (anything on the golf stream is excellent for wind)

There is very little need to import electricity from across the med, the loss in transmission alone not to mention the instability in the region makes this entire deal pretty silly.

If nothing else there is no reason to replace one type of mono-economy that keeps oppressive regimes in power and forces the west to balance the region on the head of a pin more often than not through coercion with another one.

There are a few things that aren't quite right here.. Primarily that Europe and the Sahara have equal sunlight availability.

The Moroccan plant is in Ouarzazate, which has an average annual irradiance of ~2,200 kWh/sq. meter.[1] Southern Spain is the sunniest part of Europe and the best locations there are closer to 1,800 kWh/sq. meter.[2] That's over 20% more solar energy in Morocco.

The other bit that's not really accurate is about Transmission losses. It's about 400 miles from Ouarzazate to Gilbatrar. Modern HVDC lines have losses of ~3.5% over 1,000km, so call it 3% for this project.[3] That's extremely reasonable for an energy project.

[1] - https://upload.wikimedia.org/wikipedia/commons/6/6d/SolarGIS...

[2] - http://www.mappery.com/maps/Solar-Radiation-Map-of-Spain.png

[3] - https://en.wikipedia.org/wiki/High-voltage_direct_current#Ad...

You are off by about a factor of 1000 unless you are on mercury and even there I'm not sure it gets to 2 megawatt per square meter.

20% isn't really an issue especially for thermal.

Maybe, just maybe, if we filled the entirety of Northern Africa with solar farms and somehow found a way to prevent them all getting covered in sand, damaged by animals, crapped on by animals, stolen, shot at, sabotaged, and used as both a shelter and cooking utensil - we might be able to provide Sicily with a few hours of power each day. Great success!

This will most likely have unintended negative effects.

I think you are severely underestimating just how big the deserts in Northern Africa are. You could power the world with a fraction of it if we had power storage and transmission tech. We could power Europe with todays tech. But yes, it involves a lot of engineering and logistics work, like any power generation of that magnitude. But the real problems are economics and maybe political. Does it make economic sense, and are countries comfortable relying on power from foreign nations?

I was being a little facetious there, there's probably enough vacant space in Northern Africa to power a bit more than Sicily. But I think you've hit the nail on the head there - why would anyone want to spend a disproportionate amount of money to get power from another country?

I don't know much about this, but doesn't the second sentence on that wiki page say they are aiming for 580 MW at peak once the plant is done?

Of course, on the day I visit the sky is covered in clouds. “No electricity will be produced today,“ says Rachid Bayed at the Moroccan Agency for Solar Energy (Masen), which is responsible for implementing the flagship project.

This is a big problem with concentrating thermal solar power: it requires intense direct normal irradiance to work and performance falls off catastrophically when skies aren't clear. A conventional non-concentrating PV module can deliver about 10% of its rated output if it is illuminated at 10% of standard test conditions, e.g. sun through light cloud cover. Performance falls off almost linearly with decreasing illumination until very low levels. Concentrating solar thermal performance falls off a cliff with insufficient and/or diffused light.

You can see this effect in the EIA's data about US solar farms here: https://www.eia.gov/electricity/monthly/epm_table_grapher.cf...

In the summer months, PV and concentrating solar thermal plants have comparable capacity factors. In the winter months, PV plant capacity factor falls off by less than half while thermal plant performance tumbles by ~80% (June 2016: 33.6% and 33.5%, January 2016: 17.9% and 6.8%).

A decade or so ago, solar thermal was justifiable because PV was so much more expensive. Even after PV matched and then beat solar thermal costs for instantaneous generation, thermal still had a value proposition for after-sunset operation via thermal storage (like this Moroccan plant has with molten salts). But as costs of PV and alternate forms of storage continue to fall, while solar thermal costs barely budge, I expect that solar thermal electricity is going to become obsolete even for the very sunniest regions.

Concentrated Solar Power (CSP) is a pretty interesting, but I chuckled at this phrasing, "The hot oil is used to produce high-pressure water vapour that drives a turbine-powered generator." Seriously? Why not just call them steam turbines?

Because I was going to college in LA and my parents lived in Las Vegas, I drove that route a lot and got to watch them build Solar One[1], and toured it after it was operational.

These days with better thermoelectric generation capability[2] you can extract even more of the energy out of the heat difference than just steam turbines (which have a Carnot efficiency of about 50%) mostly because they can work down to lower differentials in heat.

[1] https://en.wikipedia.org/wiki/Solar_power_plants_in_the_Moja...

[2] https://www.technologyreview.com/s/531526/an-industrial-size...

"Hundreds of curved mirrors, each as big as a bus, are ranked in rows covering 1,400,000 sq m (15m sq ft) of desert, an area the size of 200 football fields."

"The plant keeps generating energy after sunset, when electricity demands peak. Some of the day’s energy is stored in reservoirs of superhot molten salts made of sodium and potassium nitrates, which keeps production going for up to three hours. In the next phase of the plant, production will continue for up to eight hours after sunset."

They could modulate the use of the molten salts according to the weather forecast or electricity grid prices/demand. I'm assuming Morocco doesn't have grid-level dynamic pricing but if they're forecasting clouds the day after sunlight they could maybe hang on to the hot salts over night and push the heat to the steam-generator the next day when demand spikes again.

Tangentially related side note, as a plasma physicist, every time I see advances like this, I worry a little bit for my future career[0]. It turns out there's a fusion reactor at the center of this solar system that's been dumping energy on us for billions of years. Of course, eventually all energy sources on earth were solar originally.

Not that doing fusion won't be interesting from a scientific point of view, but if we develop solar and batteries further, we could really require no other energy revolution than this.

[0] tongue-in-cheek, there's always stuff to research.

We would still need fusion for interstellar travel.

Or even for ships and planes. Whats more likely in the long term, that cargo ships will be powered by batteries or that they could be powered by a compact fusion reactor?

With the right batteries, why not? You can pack them in 40ft containers and swap them at the port to charge them on land when energy is cheapest. That way it could probably be even possible to retrofit old container ships.

>"Of course, eventually all energy sources on earth were solar originally"

I originally came to reply "Not geothermal".

But just read up a bit and found out only a fraction of geothermal heat came from Earth's gravitational collapse. The bulk of the heat is from radioactive decay. Which of course was sourced from stellar fusion.

It looks pretty impressive in satellite imagery [1]. Does anyone know what the black and green rectangles at the southern end are? Heat storage?

Also, there is what appears to be more land immediately to the north that has been leveled - presumably for more arrays.

[1] https://www.google.co.uk/maps/place/Centrale+solaire+Noor+1+...

It looks like they're water storage ponds. Cut from the Wiki[1] article "Water consumption for the Ouarzazate Noor complex is estimated at 2.5 to 3 million m3 per year for one wet- cooling project (Noor I) and two dry – cooling projects (Noor II and III) and due to the need to regularly clean the reflectors."

[1] https://en.wikipedia.org/wiki/Ouarzazate_Solar_Power_Station

The following NREL page provides information on Ouarzazate (Phase I), a concentrating solar power (CSP) project, with data organized by background, participants, and power plant configuration:

http://www.nrel.gov/csp/solarpaces/project_detail.cfm/projec...

  Solar Resource:	2,635 kWh/m2/yr
  Break Ground:	        May 10, 2013
  Start Production:	December 1, 2015
  Cost (approx):	1,042 € million
  PPA/Tariff Date:	November 19, 2012
  PPA/Tariff Rate:	1.62 Dirhams per kWh
  PPA/Tariff Period:	25 years
  PPA/Tariff Information: US$ cents 18.9 per kWh
  Project Type:	        Commercial

nice to see a solar thermal plant that doesn't set fire to birds. Unfortunately they are considering the eye of sauron design for the third phase of the project. Maybe there are no birds there.

While it may only average 42 MW year round, 35 days of the year it produces nothing and it probably takes some time in the morning for things to heat up before any energy can be produced, so say 12 hours of darkness a day it produces nothing and the three hours of heat storage and three hours of morning start up time cancel out. Then it averages 92 MW when it is actually generating.

I wonder what types of adjustments have to be made to the design of the plant to endure being thermally cycled every day.

Reply on news.ycombinator.com