I guess most energy in UAE is used for cooling which should be highly correlated to solar output. They could probably meet most of their energy demand with solar power without significant amount of storage.
Furthermore, desalination will probably also play a major role in electricity consumption and I can imagine that plants can be adjusted to solar power production instead of running them at the same level 24/7 (pure speculation, no idea how the technology behind this works).
Edit: All water in UAE comes from desalination[0], so using clean water storage with larger plants instead of energy storage could be cheaper.
> They could probably meet most of their energy demand with solar power without significant amount of storage.
Not really, I was living in Dubai last year, and during the summer it was around 48C/120F in the day but it's still 35C/95F+ at night. As you can imagine literally everywhere inside is air conditioned, even some bus stops are!
Due to the lower latitude the days are also quite short (at least compared to Europe), so solar less effective without some sort of backup or storage.
The desalination plants are usually combined with power plants. The main source of fuel for power generation is natural gas, which is a by-product of oil drilling and often just burned off as it's harder to transport or store. Clean water is needed to run the turbines, so sea water is desalinated and then afterwards used for the water supply. It leads to an overall efficiency of ~86%, but if you assume the gas would have just be burnt off it's effectively free energy.
Some cooling systems make ice when energy prices are lower to use as a heat sink when energy prices are higher. The same principle can be applied here - during day the energy is used to cool and to make enough ice for cooling to last through the night.
Ok, agree with the AC, but you could use an ice storage solution as proposed in the other comment (at least for commercial buildings).
For desalination, you could still use solar power for desalination plants and sell the natural gas instead of burning it? At 2.5c/mwh I can imagine that it's more lucrative to switch power and sell the gas to a country like US or UK where NatGas is needed for power plants as solar power is much more expensive.
It is a thoughtful suggestion but Im guessing that if it was possible to ship off that gas and make a profit, it probably would have already been done. Perhaps its not that cost effective right now.
In any case, I think now is a good time for them to invest in developing these alternatives, perhaps on a smaller scale, so that when they run out of all the plentiful fossil fuels, they would then have another system that meets their needs.
Still unclear whether future developments (like mass desalinization mentioned by dx034) will change this. Or storage just could become unexpectedly cheaper, like solar itself did.
You can get to approximately 50% penetration safely without storage[1], then you have to start adding factors that can deal with intermittent sources like wind and solar[2]. It doesn't have to be storage, either. You can used demand response signals to outsource your "storage" by paying people to turn off their electric vehicle charging or ice manufacturing for a period of time. In energy, reducing demand has the same result as increasing supply.
Additionally, in hotter places like the UAE, solar irradiation tends to correlate much more with grid load, since cooling is the primary energy use. So, you could safely push well beyond the 50% penetration point for solar/wind.
> solar irradiation tends to correlate much more with grid load.
This is a very interesting remark. Last week, we had relatively high temperatures in Germany and France (above 30°C). In Germany, because the weather was perfect for solar, this reduced the middle of the day spot price where in France, this increased the spot price[0].
> You can get to approximately 50% penetration safely without storage[1], then you have to start adding factors that can deal with intermittent sources like wind and solar[2].
Keep in mind that this number will vary depending on generation mix. Specifically, it will depend on the response time of your dispatch-able generators. If your primary source of non-renewable generation is big steam turbines with a painfully slow ramp-rate (e.g. coal generation) then you're going to have a bad time.
> If your primary source of non-renewable generation is big steam turbines with a painfully slow ramp-rate (e.g. coal generation) then you're going to have a bad time.
Which is why most coal in the US is being displaced by combined cycle natural gas turbines; ~10-30 minute ramp, depending on load demand.
> There are many instances where solar electricity price is negative.
That's not how it works. When someone wants to buy power they get bids, the bids are sorted by price, and capacity.
Then they work their way through the list, cheapest first, till they meet the capacity they need.
And then pay the (same) TOP price to everyone! NOT the price they bid.
So "negative" is just a tactic to get a good position on the list. Risky if you are a normal fuel-based plant, since you might actually have to make good on it. Not risky for solar power (but not something they actually want).
If everyone was solar, they would never use negative prices - they might actually have to make good on that. It only works because their capacity is so low they are never the top bid.
I'm not sure there are cases where solar is negative, but there is with wind. That's because in the IS wind producers get paid the wind production tax credit when they produce electricity regardless of demand. So you can have negative prices over night when you get a lot of wind and low electricity demand.
Efficient renewable energy is great, but isn't the problem storage of that energy? Without storage, the energy generated by renewables needs to be used up by the grid immediately. Won't most of it be wasted?
I've been wondering why AC units for buildings don't come in solar cell combi packages, where if it's hot, it's sunny, so the ac can be powered by the sun, ideally making it a zero sum calculation.
Because it is not cost effective. Fossil fuel is still cheaper. Unless solar prices ( installation- panel, batteries) comes down or subsidised to match those existing prices, it will still take some time .
Because the amount of solar required to power an air conditioner is significantly larger than the air conditioner itself and often you need more solar than you have roof space for buildings 2 stories and greater. (i.e. the total amount of solar you can fit on the roof of a 4 story building will likely not be sufficient to run the air conditioner required to cool that building.)
That's an overemphasized problem. There's a lot more space to cheaply time-shift electricity demand (i.e. not with batteries) in response to variable pricing than most people realize.
I don't think it's overemphasized at all. Time shifting is only effective right now due to oddities of our current system that are in no way guaranteed to be permanent.
Once we get enough solar installed that it is powering most of the air conditioning when it's most used (which, contrary to what some are saying, is close enough to grid parity to be inevitable in many places), time shifting will be much less advantageous.
At some point, storage is going to play a huge role (there are places where it's already starting to, like Hawaii). It's frustrating that we're still having so much trouble bringing down cost per kWh, but with all the disparate technologies being researched, I think it will happen eventually. Just not as soon as some of us would have liked.
Well, that of course depends on whether you're talking about monetary efficiency (for which time shifting of large loads to lower night time rates makes monetary sense currently) or actual electrical efficiency (clearly needing a ton of work to properly improve upon).
Ultimately the latter is better, but there is so much nuance involved that I don't expect it any time soon. I personally believe that anyone interested in pushing the true efficiency envelope should be looking at getting the most out of off-grid systems right now, at least until more promising innovations show up within the grid itself.
Doesn't that really depend on if you can't use it all? If you get 50% of your electricity from solar during the day, can't you use it all? At night you simply have to get the other 50% from fossil fuels.
Isn't that still a big win in the short term? If half the world's daytime electricity came from solar, we'd be in much better shape.
First you can transport electricity east or west to shift supply. This is only good for around +/- 2 hours before local storage is generally a better idea, but just that works out to ~1/3 of your storage needs. (1040 miles = 1 hours at the equator, less as you move north.)
Next, there are many ways to store energy. Desalination is a great example where they need high pressure water so just pump it up a hill in the day and all you need to 'add' is slightly more pumps and a large pond. Heating and cooling are another area where storing the output tends to be very cheap.
Finally, people just need less energy when they sleep. Your work PC is probably also asleep for example.
So wind and solar are currently the cheapest, but you can only get to about 50% penetration with those without storage[2]. The rest has to be "dispatchable".
>So wind and solar are currently the cheapest, but you can only get to about 50% penetration with those without storage[2]. The rest has to be "dispatchable".
That % will probably go up as the demand side of the market reacts to variable pricing. We've barely scratched the surface of what's possible in terms of demand shifting.
One issue with these numbers that they are making and apples-to-oranges comparison with non-dispatchable sources (wind and solar) and dispatchable sources. There is value to being able to turn on and off a generation source that isn't accounted for here.
Also there is great value not having pollutant particles in the air and in the water that will kill your father from cancer when he is 48.
There is value not releasing radioactivity in the air from the natural isotopes you find in coal for example too.
There is value not having the fish or the cow you eat contaminated by waste.
There is value in not having your crops destroyed by acidic water.
There is value in having a living coral reef and not being dissolved.
I can certify there is value in all of that because I was raised in a coal extracting place. Probably you only know about the part of switching a lever on and off.
To get those numbers, nuclear is also non-dispatchable, because it must be run continuously except for the small maintenance windows every 18 months.
Additionally, the combined cycle power plants that allow for super efficient and cheap fossil fuel plants do not respond quickly to turning on and off; it takes time.
With solar and wind, you can choose to not generate but you can't choose to generate if the sun isn't shining. For cheap fossil fuel electricity, you need to plan out a day in advance. It's not so easy to spin a nuclear plant up and down.
Peaker plants, that are truly dispatchable, are about the cost of solar+storage on that chart these days.
and to add more context, the bid in this article is $24/MWh, which would imply a $100/MWh Solar + Storage cost given the numbers above from Lazard. For comparison, the new Hinkley nuclear plant in England has a PPA guarantee at ~$120/MWh.
Obviously there are geographic differences, but the trendline for solar is impressive.
I didn't realize that solar+storage was already so cheap.
If there's a mix of say, 50% Wind + Solar, and 50% from stored solar (or wind), then at worst that's an average electricity cost of $83/MWh. That's already cheaper than nuclear, which I didn't think would happen this decade.
Natural gas produces .61 tons of CO2 per MWh [1]. If you take Exxon's planned tax amount of $80/ton of CO2, that would put natural gas at $101/MHw. This is Exxon's estimate, so I think we can safely say that $80/ton is a lower bound on the true cost of fossil fuels to the rest of the economy.
So in reality, renewables+storage are already cheaper than fossil fuels. We're just subsidizing fossil fuels by ignoring their negative externality.
Natural gas can presumably be easily switched on and off to take up the intermittent 50% of the demand. So the realistic combined price is somewhere around $47.
Why then build nuclear at $97 ??
I am certain that nuclear power takes days to "spin up," so even if gas plants are not responsive (it is my impression that they are), it is still sensible to compare them.
The real reason to consider nuclear is that there is more than construction cost, there is also fuel cost. The fuel over the life of the nuclear plant is considerably cheaper than for a gas plant. Of course, even that isn't the whole story: I don't know how total operating cost and negative externalities balance out.
EDIT: I misread; it appears that those numbers are total cost per MWh produced, not construction cost. In that case, it appears that gas definitely beats nuclear hands down (ignoring the CO2 problem).
> I am certain that nuclear power takes days to "spin up,"
Yes, which is why you would want it for base load rather than responsive. If we invested in nuclear technology rather than condemn it all as a terrible venture we could produce far cheaper and cleaner nuclear plants.
Interestingly the need for base load is going away as we increase the mix of renewables: plants that can't load follow are becoming less useful and more expensive to operate.
Diablo Canyon nuclear plant in California is being shut down in a few years mostly due to environmental concerns but also in part because it can't be spun up and down easily to adjust for the output of the state's increasing mix of fluctuating solar and wind production.
We're headed towards a mix that looks like almost entirely renewables and peakers, or maybe eventually renewables and storage, if things like the tesla battery installation work out.
Storage is the key. There are a lot of places where renewables aren't stable enough and storage just isn't achievable yet on the grid scale. Lots of exciting work in this area and it's interesting to see where things are heading however I think nuclear is underfunded and could have a much more positive role in our energy needs. Maybe it's the BNFL ad's as a kid imprinted in my mind.
There's a bit of misdirection employed in distilling every consideration in building a power plant down into one single dollar per whatever ratio. Other factors may include volume and risk of waste byproduct, amortization and duration of a particular type of power plant, geographic factors coupled with insurance costs, potential economic factors of switching everything over to a particular technology, number of suppliers in the market, etc. If the entire USA were to switch over or away from a particular technology, that's going to have market effects on all of the other numbers...e.g. dollar figure $X is not static, but rather each dollar figure is dependent upon the others in ways that are not covered in this one paper.
I'm confused why you're confused. You pay for energy, not for power. You wouldn't pay the same for a kW delivered for 1 hour as you would for a kW delivered for a whole year, right?
When you buy solar panels, or build what TFA called "a solar plant capable of generating at least 350 megawatts" (note the units) you certainly are paying for power. Quoting an energy price hides all sorts of assumptions about finance, equipment longevity, and maintenance costs, so it's clear that doing so is just an excuse to say "cheapest solar on record".
The people who build the power plant pay for MW of solar panels. The people who buy the energy from the power plant pay for kWh. Notice that this an interesting article because the plant is going to have a power purchase agreement price of under 2.5 cents per kWh. We don't know what the equipment costs per Watt.
It's IAAS (infrastructure as a service (a term I just made up, but I think it is appropriate))
Governments nowadays don't buy a solar plant, they put out a tender "we're willing to buy X GWh of solar power a year for Y years. Solar panels can be placed in area A, and must be removed afterwards. Please quote us a price per kWh."
Advantage of that for the buyer is that, if they get a offer that is too good to be true, they haven't paid the full sum when the seller goes bankrupt.
Also, because the seller has to operate and maintain what he sells, he will think harder about maintenance costs, ease of removal, etc.
Risk for the buyer is that he commits to buy a fixed amount of power at a fixed price over the lifetime of the plant. Worst-case for the buyer, we develop cheap fusion next year, so that electricity prices drop to zero.
1 kWh is 3,600,000 Joules but it is more convenient to use some unit of watt because the power rating of the things you're attaching it to are often rated in Watts also.
Besides it is just the way things evolved. Like batteries being rated in mAh.
I'm gung-ho on solar but the article clearly says there is a low penalty for going over budget. They may just be guessing where prices will be by the time they have to pay for the panels.
71 comments
[ 2.8 ms ] story [ 116 ms ] threadAnyways will be interesting to see for the future if solar actually becomes a more dominant source of energy worldwide.
Furthermore, desalination will probably also play a major role in electricity consumption and I can imagine that plants can be adjusted to solar power production instead of running them at the same level 24/7 (pure speculation, no idea how the technology behind this works).
Edit: All water in UAE comes from desalination[0], so using clean water storage with larger plants instead of energy storage could be cheaper.
[0] https://en.wikipedia.org/wiki/Water_supply_and_sanitation_in...
Not really, I was living in Dubai last year, and during the summer it was around 48C/120F in the day but it's still 35C/95F+ at night. As you can imagine literally everywhere inside is air conditioned, even some bus stops are!
Due to the lower latitude the days are also quite short (at least compared to Europe), so solar less effective without some sort of backup or storage.
The desalination plants are usually combined with power plants. The main source of fuel for power generation is natural gas, which is a by-product of oil drilling and often just burned off as it's harder to transport or store. Clean water is needed to run the turbines, so sea water is desalinated and then afterwards used for the water supply. It leads to an overall efficiency of ~86%, but if you assume the gas would have just be burnt off it's effectively free energy.
For desalination, you could still use solar power for desalination plants and sell the natural gas instead of burning it? At 2.5c/mwh I can imagine that it's more lucrative to switch power and sell the gas to a country like US or UK where NatGas is needed for power plants as solar power is much more expensive.
In any case, I think now is a good time for them to invest in developing these alternatives, perhaps on a smaller scale, so that when they run out of all the plentiful fossil fuels, they would then have another system that meets their needs.
http://marginalrevolution.com/marginalrevolution/2016/08/fur... http://marginalrevolution.com/marginalrevolution/2016/07/wha...
Still unclear whether future developments (like mass desalinization mentioned by dx034) will change this. Or storage just could become unexpectedly cheaper, like solar itself did.
Additionally, in hotter places like the UAE, solar irradiation tends to correlate much more with grid load, since cooling is the primary energy use. So, you could safely push well beyond the 50% penetration point for solar/wind.
[1]: https://people.eecs.berkeley.edu/~vsmith/docs/renewables_sgc...
[2]: https://www1.eere.energy.gov/solar/pdfs/2010ulw_ellis.pdf
This is a very interesting remark. Last week, we had relatively high temperatures in Germany and France (above 30°C). In Germany, because the weather was perfect for solar, this reduced the middle of the day spot price where in France, this increased the spot price[0].
[0]: https://www.eex.com/en/ "European Energy Exchange"
In the UK, meanwhile, peak load is driven by tea kettles and Eastenders.[1]
[0a] http://cleantechnica.com/2014/07/21/utilities-cry-fowl-over-...
[0b] http://businesstech.co.za/news/hardware/86252/new-battery-te...
[0c] https://www.cnet.com/news/a-technology-race-to-curb-peak-ene...
[0d] There's a great NPR viz on power sources around the US, though somewhat dated now: http://www.npr.org/2009/04/24/110997398/visualizing-the-u-s-...
[1] http://www.bbc.co.uk/britainfromabove/stories/people/teatime...
Keep in mind that this number will vary depending on generation mix. Specifically, it will depend on the response time of your dispatch-able generators. If your primary source of non-renewable generation is big steam turbines with a painfully slow ramp-rate (e.g. coal generation) then you're going to have a bad time.
Which is why most coal in the US is being displaced by combined cycle natural gas turbines; ~10-30 minute ramp, depending on load demand.
http://www.utilitydive.com/news/a-users-guide-to-natural-gas...
That's not how it works. When someone wants to buy power they get bids, the bids are sorted by price, and capacity.
Then they work their way through the list, cheapest first, till they meet the capacity they need.
And then pay the (same) TOP price to everyone! NOT the price they bid.
So "negative" is just a tactic to get a good position on the list. Risky if you are a normal fuel-based plant, since you might actually have to make good on it. Not risky for solar power (but not something they actually want).
If everyone was solar, they would never use negative prices - they might actually have to make good on that. It only works because their capacity is so low they are never the top bid.
That's an overemphasized problem. There's a lot more space to cheaply time-shift electricity demand (i.e. not with batteries) in response to variable pricing than most people realize.
Once we get enough solar installed that it is powering most of the air conditioning when it's most used (which, contrary to what some are saying, is close enough to grid parity to be inevitable in many places), time shifting will be much less advantageous.
At some point, storage is going to play a huge role (there are places where it's already starting to, like Hawaii). It's frustrating that we're still having so much trouble bringing down cost per kWh, but with all the disparate technologies being researched, I think it will happen eventually. Just not as soon as some of us would have liked.
It's not effective right now because we have a grid built upon the assumption that demand and supply are relatively static.
Ultimately the latter is better, but there is so much nuance involved that I don't expect it any time soon. I personally believe that anyone interested in pushing the true efficiency envelope should be looking at getting the most out of off-grid systems right now, at least until more promising innovations show up within the grid itself.
Isn't that still a big win in the short term? If half the world's daytime electricity came from solar, we'd be in much better shape.
Next, there are many ways to store energy. Desalination is a great example where they need high pressure water so just pump it up a hill in the day and all you need to 'add' is slightly more pumps and a large pond. Heating and cooling are another area where storing the output tends to be very cheap.
Finally, people just need less energy when they sleep. Your work PC is probably also asleep for example.
[1]: https://www.lazard.com/media/2390/lazards-levelized-cost-of-...
[2]: https://people.eecs.berkeley.edu/~vsmith/docs/renewables_sgc...
That % will probably go up as the demand side of the market reacts to variable pricing. We've barely scratched the surface of what's possible in terms of demand shifting.
There is value not releasing radioactivity in the air from the natural isotopes you find in coal for example too.
There is value not having the fish or the cow you eat contaminated by waste.
There is value in not having your crops destroyed by acidic water.
There is value in having a living coral reef and not being dissolved.
I can certify there is value in all of that because I was raised in a coal extracting place. Probably you only know about the part of switching a lever on and off.
Additionally, the combined cycle power plants that allow for super efficient and cheap fossil fuel plants do not respond quickly to turning on and off; it takes time.
With solar and wind, you can choose to not generate but you can't choose to generate if the sun isn't shining. For cheap fossil fuel electricity, you need to plan out a day in advance. It's not so easy to spin a nuclear plant up and down.
Peaker plants, that are truly dispatchable, are about the cost of solar+storage on that chart these days.
Obviously there are geographic differences, but the trendline for solar is impressive.
[1] https://en.wikipedia.org/wiki/Hinkley_Point_C_nuclear_power_...
If there's a mix of say, 50% Wind + Solar, and 50% from stored solar (or wind), then at worst that's an average electricity cost of $83/MWh. That's already cheaper than nuclear, which I didn't think would happen this decade.
Natural gas produces .61 tons of CO2 per MWh [1]. If you take Exxon's planned tax amount of $80/ton of CO2, that would put natural gas at $101/MHw. This is Exxon's estimate, so I think we can safely say that $80/ton is a lower bound on the true cost of fossil fuels to the rest of the economy.
So in reality, renewables+storage are already cheaper than fossil fuels. We're just subsidizing fossil fuels by ignoring their negative externality.
[1] https://www.eia.gov/tools/faqs/faq.cfm?id=74&t=11
There is always a baseline need for energy, seems most sensible to me this comes from a modern, nuclear source and not the relics we rely on today.
The real reason to consider nuclear is that there is more than construction cost, there is also fuel cost. The fuel over the life of the nuclear plant is considerably cheaper than for a gas plant. Of course, even that isn't the whole story: I don't know how total operating cost and negative externalities balance out.
EDIT: I misread; it appears that those numbers are total cost per MWh produced, not construction cost. In that case, it appears that gas definitely beats nuclear hands down (ignoring the CO2 problem).
Yes, which is why you would want it for base load rather than responsive. If we invested in nuclear technology rather than condemn it all as a terrible venture we could produce far cheaper and cleaner nuclear plants.
Diablo Canyon nuclear plant in California is being shut down in a few years mostly due to environmental concerns but also in part because it can't be spun up and down easily to adjust for the output of the state's increasing mix of fluctuating solar and wind production.
We're headed towards a mix that looks like almost entirely renewables and peakers, or maybe eventually renewables and storage, if things like the tesla battery installation work out.
GP was expecting to see a price in terms of $/W, because that's what matters to consumers interesting in purchasing panels: https://en.wikipedia.org/wiki/Price_per_watt
The article quotes a price in terms of $/kWh, because that's how a bid from an industrial electricity provider is stated.
GP was confused / frustrated because a bid in $/kWh completely obscures the cost of the hardware, which is what he's interested in.
Governments nowadays don't buy a solar plant, they put out a tender "we're willing to buy X GWh of solar power a year for Y years. Solar panels can be placed in area A, and must be removed afterwards. Please quote us a price per kWh."
Advantage of that for the buyer is that, if they get a offer that is too good to be true, they haven't paid the full sum when the seller goes bankrupt.
Also, because the seller has to operate and maintain what he sells, he will think harder about maintenance costs, ease of removal, etc.
Another example of this is the recent auction for a sea-based wind farm in the Netherlands (http://www.offshorewind.biz/2016/07/06/borssele-12-worlds-ch..., 7.27 eurocents/kWh)
Risk for the buyer is that he commits to buy a fixed amount of power at a fixed price over the lifetime of the plant. Worst-case for the buyer, we develop cheap fusion next year, so that electricity prices drop to zero.
1 kWh is 3,600,000 Joules but it is more convenient to use some unit of watt because the power rating of the things you're attaching it to are often rated in Watts also.
Besides it is just the way things evolved. Like batteries being rated in mAh.
If they aren't perovskite (and probably can't be since those are oh-so-close but not being manufactured), that raises two exciting possiblities:
1) another jump when perovskit ~20% efficient cells are produced en masse
2) The hybrid perovskite + other types variants that give even better efficiency with mixed costs...
As I understand it, the only challenge with perovskites now is that they break down in contact with humidity IIRC. Great for deserts though?