Slashdot Mirror
A Tower of Molten Salt Will Deliver Solar Power After Sunset (ieee.org)
schwit1 sends this report from IEEE Spectrum: Solar power projects intended to turn solar heat into steam to generate electricity have struggled to compete amid tumbling prices for solar energy from solid-state photovoltaic (PV) panels. But the first commercial-scale implementation of an innovative solar thermal design could turn the tide. Engineered from the ground up to store some of its solar energy, the 110-megawatt plant is nearing completion in the Crescent Dunes near Tonopah, Nev. It aims to simultaneously produce the cheapest solar thermal power and to dispatch that power for up to 10 hours after the setting sun has idled photovoltaics. ... [The system] heats a molten mixture of nitrate salts that can be stored in insulated tanks and withdrawn on demand to run the plant’s steam generators and turbine when electricity is most valuable. ... Eliminating the heat exchange between oil and salts trims energy storage losses from about 7 percent to just 2 percent. The tower also heats its molten salt to 566 degrees C, whereas oil-based plants top out at 400 degrees C.
If this company thinks they can operate this plant, I see no reason to stop them.
I see no reason why you keep mentioning birds like it's some sort of game changer. In Canada between 16-42 million birds are killed each year through collisions with buildings. Should we stop building houses? http://www.ace-eco.org/vol8/is... North America wide that number may rise as high at 1 billion. http://www.flap.org/faqs.php Not to mention that you conveniently left out the death toll on all animals from pollution/habitat loss from the fossil fuel generators which far exceeds the numbers of 'streamers' that these plants will generate.
Improvements on all fronts, should not be abandoned because those improvements are not perfect.
Why do you need to smooth pricing? By allowing prices to rise and fall throughout the day in response to supply and demand, you don't need to add supply between the time the sun goes down and the time people go to bed at night.
In other words, you can treat it as an economics problem and save your customers a lot of money on power plants and fuel. This is why the world is switching from flat rate pricing to time-of-use pricing.
The goal isn't to smooth pricing...it's to smooth the peaks and valleys of demand.
The grid has to be built for peak, not average...in other words, if 5% of the time, the total load of a utility's customer base is 4.5 gigawatts, then they have to be able to provide 4.5 gigawatts, even though 95% of the time the demand is half that, at most.
Ideally, power demand would be flat and constant...the same amount, all the time. Steam plants experience metal fatigue when they throttle up and down, and this is already a major problem with most utilities now. It's also way harder to regulate an efficient burn at multiple rates...which in turn, means it's harder to regulate emissions, which leads to limits on capacity if they exceed emissions of certain sorts (and, just to make it fun, those standards have just been tightened...a LOT). Those are both a big deal: too much leaking in the heat exchange coils in the boiler, and the whole plant has to come offline. Even getting close to the limit on emissions for a period, and the plant comes offline to avoid overshooting it...the plant goes into reserve mode, needed only for emergencies. And this, in turn, increases the impact of the peaks/valleys situation on the rest of the utility. And what I just described assumes 100% controllable, fuel-based generation (nuclear, petroleum, coal, gas). Now, these peaks are predictable (and predicted...there's a whole industry around the metrics and predictive load management involved), but it still poses a challenge. The steeper the walls of the peak, the faster and harder you have to spin up the generators, and the greater the stress, as well.
Renewable energy is great, except that it throws another wrench into the works. Let's say you're getting a lot of your power from solar...but then clouds move in. Effectively, for your non-renewable generation, you've just introduced a peak because it has to throttle up to take up the slack. So you end up with lots of peaks of various sizes, instead of the one or two big peaks per day. And even worse, these peaks aren't predictable.
If, however, your solar generation capacity includes a way to continue generating power after the clouds roll in, you've done two things. One, if the cloud layer is short-lived, you're able to simply disregard it and life goes on. Two, if it isn't, then you've bought more time to spin up capacity more slowly...which means less stress on the boilers, and also more options to choose from. Maybe you fire up a CT peaker, which has less trouble with variable load but takes 20-30 minutes to come online, for example. But ultimately, what you've done is taken one of the biggest problems with renewable generation and dramatically reduced it.
For your security, this post has been encrypted with ROT-13, twice.