This is probably a bad idea, but couldn't you store energy basically having wind/solar power water pumps to pull water from one location to another? Something like 2 lakes side by side with a dam in the middle pulling energy off the transfer between when energy is needed and then water being pumped back in the reservoir when energy is being claimed from wind/solar?
Yes, basically that, but using solar/wind to power the pumps, turning the lake into a giant battery to store the energy from the solar/wind power, which seems to be one of the major drawbacks of wind/solar.
That's an interesting and seemingly obvious approach, so I'm curious why it's not really discussed.
I'm guessing that at current wind/solar capacity, any fluctuations can be absorbed by just slowing down the flow out of the reservoir, rather than ever having it reverse.
You also need a downhill source of water, which means any dam that's the last one downstream can't really use this approach.
You could maybe even get away with an entirely closed-loop system, but if you look at the scale of the problem, you need either a huge pair of reservoirs (lake-sized) or a huge elevation difference (mountain-sized), or preferably both. It doesn't appear economical to build e.g. a tank on a tower.
One of the major reasons pumped storage isn't used more is because the energy loss associated with pumping and storing the water. I can't find the exact amount right now but from a lecture I watched some time back[1], pumped storage is a few orders of magnitude more expensive than using conventional fossil fuel generated electricity per kWh.
It takes a huge space. 1 kWh means moving a metric ton (a cubic meter) of water 400m of elevation.
Round trip efficiency isn't great, but also not terrible. It beats the fuel-cell/electrolysis cycle, but loses to batteries. It can be profitable though because energy rates can reach near zero (and even occasionally slightly negative) at night.
At this point, it's surprising we don't just cut off the flow of existing hydro when we have other sources peaking... e.g. it's what you describe but no need for upward pumping.
We do this to some extent, but the reality is we do need to ensure rivers continue to flow for a number of reasons (I guess).
basically the author says that what you describe is the best option right now but there are only so many places you can put those facilities because they require very specific geography.
That being said building lakes and dams is really bad for the river eco system and really expensive to buy land out from under people to flood it.
If you haven't seen already you might be interested in http://lightsailenergy.com/tech.html; they're working on compressed-air-and-water-vapour storage (also DaniFong posts on HN sometimes)
Why don't we have our energy prices directly depending on the availability ?
Each house should have a big meter showing the variable price and people will then use energy hungry appliances only when the energy is cheap.
This way we act as a buffer allowing better ressource usage.
Even better, appliances like dishwashers and washing machines could be connected to the network to read the price level and run automatically when the price is lower than a fixed amount.
Even better, for industries which are using energy automatically, they should bid for the excess energy and when there is a big surplus, the energy will be really cheap and will be used instead of going to waste.
The argument probably boils down to: the net economic loss in productivity from having every micro-power-user spend time worrying about whether it's a good or bad time to flip the switch, out-weighs the added market efficiency from what you describe.
Large users on the other hand, can devote a team to ensure optimal usage around complex / time-dependent tariffs.
Great point about smart appliances tho: that will have to wait for full integration of the smart grid and there will have to be a clear incentive to buy expensive appliances... since the overall savings could be fairly minimal (e.g. one less peaker plant) it's harder to provide benefits for millions of users. Again "demand response" e.g. turn-down-on-demand for large power users (perhaps even large building airconditioner loads) makes sense to target first.
There is no doubt a lot of interesting tech potential in this area... but utilities move so so slow.
There's already something called time-of-use metering, which is like what you're suggesting but based on long-term trends rather than a spot price.
Right now it's only cost-effective with large loads, and only economical if it's largely automated.
(By point of comparison, most people already can't be bothered to switch off a 100W room light when they leave. That means we just don't GAF about saving pennies per hour if it means we have to push a button every so often).
However, seeing how heating and cooling are the only residential loads that are really worth worrying about, and both can typically be advanced or deferred by an hour or so without any problems, you could probably solve 90% of the problem by smartening up a handful of appliances in each home.
Industrial users can often do similar things. A cold storage facility, for example, could do most of it's cooling when power is cheap and then coast through the expensive periods. And they use enough electricity to pay someone (or buy an automated system) to handle it for them.
Once you solve heating and cooling, everything other than cooking (basically lighting and entertainment) is getting more and more efficient every year. And most people don't cook enough to make that a significant load.
Solve the storage problem first, then everything else follows. Otherwise this is a recipe for disaster.
The problem with building out solar or especially wind power generation is that it's unreliable, it doesn't produce power when you want it. This means that you need to supplement power generation with low latency backup systems. Unfortunately, those systems tend to be gasoline-powered generators. Building those and operating them is expensive and also not so beneficial from a total CO2 emissions perspective. More so when you compare the whole system to simply replacing the windmill and the generator with a natural gas generator, which is more reliable, cheaper, and produces less total net CO2.
You completely ignored my points, some additional facts (Germany is becoming an electrical power importer, relying on France's nuclear plants, is relying more and more on coal, and may end up increasing its CO2 emissions. Meanwhile, the US has actually decreased its CO2 emissions mostly by relying more heavily on natural gas and should hit 1990 levels soon, despite population growth.):
Yes, but that doesn't contradict InclinedPlane's point in the least.
It's easy to generate a lot of renewable energy when you've got the old power plants ready to cover any shortfalls in generation capacity. Just because you can acheive 25% under those conditions by prioritizing renewables does not imply that you can push renewables up to 90%+.
The wind can stop blowing for quite some time, even on a country wide scale. You either have to store the energy, import it from other countries or have backup power stations to cover the shortfall.
The (free) book Without Hot Air is a fantastic read for more information. It's written by a no nonsense physicist that dives pretty deep into possible plans for a future energy grid, including all the messy details.
And the details are messy. Any renewable energy plan that doesn't sound like a momentus undertaking is cheating by 1) Only talking about electricity, not total energy 2) Discounting embodied energy of imports 3) Keeping a country sized fossil fuel infrastructure around to fill in the gaps 4) Importing energy
Why not handle the daytime excess problem first, and keep shutting down more coal plants as the storage gets better? Summer requires more energy than winter (good for solar), and daytime requires more power than nightime. (Again, good for solar.) Spread the solar plants around enough, modernize the grid to work nationally/internationally (US, EU) and take advantage of what we can do now.
Besides, as we bring more cars off of fossil fuels, we will need the electricity on the grid anyway.
That's not really true for electricity though. In the US we use more electricity in the summer for ac than than heating in the winter. Partly this is based on where people live, but also because people in the north use fuel oil etc in the winter for heating and people in the south have little need for winter heating and less need for AC. There is some regional spikes in demand though unusual cold snaps in southern regions, but as the overall grid has lower demands they just import power from other regions.
As to why AC uses so much power, a lot of it relates to dehumidifying air which takes a lot more power than you might expect.
PS: In most areas you can substitute solar hot water heaters for the vast majority of home heating needs so long term it's not really a problem.
I said in most of the world. The US uses more energy on AC than the rest of the world combined. It is true that gas heating is common elsewhere, but so is electric heating. Fuel oil globally is no longer that important an energy source.
Having read through their workflow diagram, I'm really unclear why they can achieve 90% efficiency. Isn't that basically just a Carnot engine? If not, why not? If it is, how can they possibly achieve 90% efficiency when the best Carnot-like engine I've seen is around 35%?
There is a way to feed the energy you make using renewable sources into the grid so you can use them when they are around. Since you're still connected to the grid, conventional energy sources or better yet, renewable energy being generated elsewhere, can take over when the renewables aren't available.
The problem is still severe. In a country the size of the US it is still possible for there to be days of very little wind across the whole country. And, of course, once the Sun goes down on the West coast there won't be any power coming from solar for many hours. Which means that in a worst case scenario you need to rely on nuclear, coal, or hydrocarbon based power generation. And it means that you can't avoid having to build out conventional power generation capacity to match peak loads (which can occur when solar and wind power production is effectively zero).
Why is storing energy at hydroelectric dams not sufficient? It's been done for ages.. (ie. pumping water back up the dam when there is an energy surplus, and using the dam when there is a deficit)
Because it's not enough and it's not everywhere. Hydroelectric power is pretty good, but it's only about 7% of the US's total electrical power production. There are parts of the country where there aren't hydropower stations, and you also can't use hydropower storage to generate 100% of the US's base power load.
The study just handwaves storage, in reality it will be an enormous problem to tackle. If it were so easy it would be done already, due to the ability to play the "buy low, sell high" game on the spot power market.
And I haven't read it yet, but I guess storage costs rise very sharply for every additional 9 you want in your reliability. 90% could mean a month without power each year. To get from there to 99% or a week one very two years by improving storage capacity, you would likely need capacity for at least three weeks of full load.
The problem with this article, and nearly all others pushing for adoption of wind power, is that it ignores reality.
>>When there is not enough renewable energy direct from source, and the stored energy reserves are insufficient to bridge the shortfall, top up the remaining few percent of the demand with fossil fuels.<<
We don't have any way to "top up" the remaining few percent of demand with fossil fuels. You can't simply flip a switch on a coal fired or nuclear power plant and have it instantly produce power. These massive processes need tremendous lead time to start producing electricity. So, in theory wind could provide a lot of power, but since we have no idea when the wind will die and when we will need to "top off" the shortfall, we keep coal fired plants churning.
This is one reason why, as the largest utility in Germany states[1], it takes roughly 24MW of wind power to take just 1MW of fossil fuel power offline:
"As a result, the relative contribution of wind power to the guaranteed capacity of our supply system up to the year 2020 will fall continuously to around 4% (FIGURE 7). In concrete terms, this means that in 2020, with a forecast wind power capacity of over 48,000MW (Source: dena grid study), 2,000MW of traditional power production can be replaced by these wind farms."
Storage, and the near-immediate ability to use what is stored, would help this tremendously, but we don't have that capability, and we don't seem to be well on our way to getting it, even by 2030.
First, you need to build wind/solar plants, which tend to be more expensive than conventional power plants to start with.
Then, you either need to invest in R&D for power storage systems and build out those facilities or, more likely, you need to build out conventional power capacity to accommodate for demand when solar or wind is not producing. In the end you've spent quite a lot more money and likely you haven't actually reduced the number of conventional power plants you've built or even reduced CO2 emissions substantially (due to the reliance on less clean conventional power plants during solar/wind down times as well as the CO2 cost of constructing those plants).
The takeaway is electricity prices will fall significantly and the worst climate change effects will be avoided as the market rapidly replaces our fossil fuel infrastructure with a renewable one.
It's good to see that there's an optimal solution that gets us reasonably economical power from 99.9% renewables.
What I'm wondering is whether we can get there from here. We can't simply dictate the end result we want and have it appear.
What we could use now is some kind of economic simulation, taking into account the cost of shutting down fossil plants early, the capital cost of building all that wind and solar (and the upfront capital is the bulk of the cost, since operating costs are low), the actual utilities and regulatory structures currently in place, etc.
Then we can see whether what sort of policy changes we might need, to get an end result like this, and what it will cost to put all this in place.
Wind and solar can be complimentary: Watch the Cal-ISO renewables graph (second graph on on this page: http://www.caiso.com/outlook/SystemStatus.html ; the first graph is demand and reserves ).
On many days, the wind power goes up at night and down during the day, very much in complement, though in larger scale than, solar generation. More solar will more nearly equalize the two -- on good days.
California is nowhere near having an overcapacity of renewable sources to worry about; just look at the first graph where the lowest demand (2 AM to 4 AM) is 22 GW on 12/14/2012. The highest the solar gets is not quite 1 GW. The wind is more variable, anecdotally as high as 3+ GW and as low as a flatline near zero for the whole state all day and night.
One great idea, is to store energy by ammonia NH3, split water into H2 and O (brown or grey, and you can actually clean the water this way) by electrolysis at as little as 1.2 V (Stan Meyer) and then using a reverse Fuel cell to bind the N in the air (N is about 80%) to store NH3 at as low as 120 PSI/10 Bar in existing nursing tanks. It can be used to run an existing (lightly converted) diesel generator or a fuel cell again at a higher efficiency level. No need to use H2 which requires massive amounts of energy to be cryogenically frozen or compressed in carbon fibre tanks. And you can run your diesel engine car/truck/train/ship.
Another novel and new invention is www.aquionenergy.com who has an environmental battery made of carbon and salt :)
Essentially its best to start capping our consumption levels and increasing efficiency. Have a look at the current wasted energy in our system: https://flowcharts.llnl.gov
43 comments
[ 2.9 ms ] story [ 77.1 ms ] threadhttp://en.wikipedia.org/wiki/Helms_Pumped_Storage_Plant
I'm guessing that at current wind/solar capacity, any fluctuations can be absorbed by just slowing down the flow out of the reservoir, rather than ever having it reverse.
You also need a downhill source of water, which means any dam that's the last one downstream can't really use this approach.
You could maybe even get away with an entirely closed-loop system, but if you look at the scale of the problem, you need either a huge pair of reservoirs (lake-sized) or a huge elevation difference (mountain-sized), or preferably both. It doesn't appear economical to build e.g. a tank on a tower.
It's just not very cost effective right now.
[1] http://www.youtube.com/watch?v=MoAFiL1UdSg
I can certainly believe the capital requirements are much higher for pumped storage than for natural gas peakers.
Round trip efficiency isn't great, but also not terrible. It beats the fuel-cell/electrolysis cycle, but loses to batteries. It can be profitable though because energy rates can reach near zero (and even occasionally slightly negative) at night.
99% of installed energy storage is pumped hydro.
We do this to some extent, but the reality is we do need to ensure rivers continue to flow for a number of reasons (I guess).
http://www.inference.phy.cam.ac.uk/withouthotair/c26/page_18...
basically the author says that what you describe is the best option right now but there are only so many places you can put those facilities because they require very specific geography.
That being said building lakes and dams is really bad for the river eco system and really expensive to buy land out from under people to flood it.
This way we act as a buffer allowing better ressource usage.
Even better, appliances like dishwashers and washing machines could be connected to the network to read the price level and run automatically when the price is lower than a fixed amount.
Even better, for industries which are using energy automatically, they should bid for the excess energy and when there is a big surplus, the energy will be really cheap and will be used instead of going to waste.
Large users on the other hand, can devote a team to ensure optimal usage around complex / time-dependent tariffs.
Great point about smart appliances tho: that will have to wait for full integration of the smart grid and there will have to be a clear incentive to buy expensive appliances... since the overall savings could be fairly minimal (e.g. one less peaker plant) it's harder to provide benefits for millions of users. Again "demand response" e.g. turn-down-on-demand for large power users (perhaps even large building airconditioner loads) makes sense to target first.
There is no doubt a lot of interesting tech potential in this area... but utilities move so so slow.
Right now it's only cost-effective with large loads, and only economical if it's largely automated.
(By point of comparison, most people already can't be bothered to switch off a 100W room light when they leave. That means we just don't GAF about saving pennies per hour if it means we have to push a button every so often).
However, seeing how heating and cooling are the only residential loads that are really worth worrying about, and both can typically be advanced or deferred by an hour or so without any problems, you could probably solve 90% of the problem by smartening up a handful of appliances in each home.
Industrial users can often do similar things. A cold storage facility, for example, could do most of it's cooling when power is cheap and then coast through the expensive periods. And they use enough electricity to pay someone (or buy an automated system) to handle it for them.
Once you solve heating and cooling, everything other than cooking (basically lighting and entertainment) is getting more and more efficient every year. And most people don't cook enough to make that a significant load.
The problem with building out solar or especially wind power generation is that it's unreliable, it doesn't produce power when you want it. This means that you need to supplement power generation with low latency backup systems. Unfortunately, those systems tend to be gasoline-powered generators. Building those and operating them is expensive and also not so beneficial from a total CO2 emissions perspective. More so when you compare the whole system to simply replacing the windmill and the generator with a natural gas generator, which is more reliable, cheaper, and produces less total net CO2.
http://en.wikipedia.org/wiki/Renewable_energy_in_Germany
It's possible
http://www.bloomberg.com/news/2011-05-30/germany-becomes-net...
http://www.bloomberg.com/news/2012-08-19/merkel-s-green-shif...
http://www.spiegel.de/international/germany/new-coal-fired-p...
It's easy to generate a lot of renewable energy when you've got the old power plants ready to cover any shortfalls in generation capacity. Just because you can acheive 25% under those conditions by prioritizing renewables does not imply that you can push renewables up to 90%+.
The wind can stop blowing for quite some time, even on a country wide scale. You either have to store the energy, import it from other countries or have backup power stations to cover the shortfall.
The (free) book Without Hot Air is a fantastic read for more information. It's written by a no nonsense physicist that dives pretty deep into possible plans for a future energy grid, including all the messy details.
And the details are messy. Any renewable energy plan that doesn't sound like a momentus undertaking is cheating by 1) Only talking about electricity, not total energy 2) Discounting embodied energy of imports 3) Keeping a country sized fossil fuel infrastructure around to fill in the gaps 4) Importing energy
Sometimes all four.
Besides, as we bring more cars off of fossil fuels, we will need the electricity on the grid anyway.
As to why AC uses so much power, a lot of it relates to dehumidifying air which takes a lot more power than you might expect.
PS: In most areas you can substitute solar hot water heaters for the vast majority of home heating needs so long term it's not really a problem.
The problem is still severe. In a country the size of the US it is still possible for there to be days of very little wind across the whole country. And, of course, once the Sun goes down on the West coast there won't be any power coming from solar for many hours. Which means that in a worst case scenario you need to rely on nuclear, coal, or hydrocarbon based power generation. And it means that you can't avoid having to build out conventional power generation capacity to match peak loads (which can occur when solar and wind power production is effectively zero).
It'd be nice if the actual study were online somewhere without a paywall.
And I haven't read it yet, but I guess storage costs rise very sharply for every additional 9 you want in your reliability. 90% could mean a month without power each year. To get from there to 99% or a week one very two years by improving storage capacity, you would likely need capacity for at least three weeks of full load.
>>When there is not enough renewable energy direct from source, and the stored energy reserves are insufficient to bridge the shortfall, top up the remaining few percent of the demand with fossil fuels.<<
We don't have any way to "top up" the remaining few percent of demand with fossil fuels. You can't simply flip a switch on a coal fired or nuclear power plant and have it instantly produce power. These massive processes need tremendous lead time to start producing electricity. So, in theory wind could provide a lot of power, but since we have no idea when the wind will die and when we will need to "top off" the shortfall, we keep coal fired plants churning.
This is one reason why, as the largest utility in Germany states[1], it takes roughly 24MW of wind power to take just 1MW of fossil fuel power offline:
"As a result, the relative contribution of wind power to the guaranteed capacity of our supply system up to the year 2020 will fall continuously to around 4% (FIGURE 7). In concrete terms, this means that in 2020, with a forecast wind power capacity of over 48,000MW (Source: dena grid study), 2,000MW of traditional power production can be replaced by these wind farms."
Storage, and the near-immediate ability to use what is stored, would help this tremendously, but we don't have that capability, and we don't seem to be well on our way to getting it, even by 2030.
[1]http://www.wind-watch.org/docviewer.php?doc=eonwindreport200...
First, you need to build wind/solar plants, which tend to be more expensive than conventional power plants to start with.
Then, you either need to invest in R&D for power storage systems and build out those facilities or, more likely, you need to build out conventional power capacity to accommodate for demand when solar or wind is not producing. In the end you've spent quite a lot more money and likely you haven't actually reduced the number of conventional power plants you've built or even reduced CO2 emissions substantially (due to the reliance on less clean conventional power plants during solar/wind down times as well as the CO2 cost of constructing those plants).
https://sites.google.com/site/verticalwindfarm/
it still doesnt matter because everyone gets screwed.
http://www.csmonitor.com/layout/set/print/content/view/print...
apparently there is no value in ideas.
also the only scholarship i ever got was from Chevron. the irony will save us!
What I'm wondering is whether we can get there from here. We can't simply dictate the end result we want and have it appear.
What we could use now is some kind of economic simulation, taking into account the cost of shutting down fossil plants early, the capital cost of building all that wind and solar (and the upfront capital is the bulk of the cost, since operating costs are low), the actual utilities and regulatory structures currently in place, etc.
Then we can see whether what sort of policy changes we might need, to get an end result like this, and what it will cost to put all this in place.
On many days, the wind power goes up at night and down during the day, very much in complement, though in larger scale than, solar generation. More solar will more nearly equalize the two -- on good days.
California is nowhere near having an overcapacity of renewable sources to worry about; just look at the first graph where the lowest demand (2 AM to 4 AM) is 22 GW on 12/14/2012. The highest the solar gets is not quite 1 GW. The wind is more variable, anecdotally as high as 3+ GW and as low as a flatline near zero for the whole state all day and night.
Another novel and new invention is www.aquionenergy.com who has an environmental battery made of carbon and salt :)
Essentially its best to start capping our consumption levels and increasing efficiency. Have a look at the current wasted energy in our system: https://flowcharts.llnl.gov