In a recent column, solar advocate Paul Wolff proposes a solution to the "variability" problem with Solar PV: “The truth is that Georgia’s wealth of solar and offshore wind capacity, supplemented by natural gas, is our best option. Unlike nuclear or coal plants, which are cumbersome and expensive to power up and down to meet variable demand, natural gas can be turned on and off virtually at the flip of a switch to offset periods when the sun and wind don’t provide enough power.”
Wolff thus would construct a hybrid power system, with a natural gas turbine system to provide the base load power and thus fill in the gaps left by variable (green, as in solar) power.
Just one question: Where's the beef? 9/27/12 Update: Here's a NYT article on new gas plant technology that goes to this issue.
I’m requesting solid, dollar-and-cents evidence to support this “hybrid” brown/green power plant proposal. My own experience convinces me he’s got none. I hope I’m wrong. Some background: I installed a 10KW, grid-tied, Solar Photovoltaic (PV) system in Gillis Springs, GA., for $3.5/watt up-front price ($1.4/watt tax-credited subsidized price). It's featured here.
I've also written about the economics and politics behind grid-tied, individually owned, PV systems.
Part of my research involved interviewing current and retired power company executives. The prime obstacle to PV, and the reason why power companies don't want it, is variability: Customers want their A-C to work all night, and their alarm clocks to go off at the right time. Coal and nuclear powered plants (brown power) produce a steady, “base-load” of power. Green (PV, wind, etc.) does not. Green (PV, specifically) at best supplies primarily “peak-load” power (sunny days when demands peaks) but remains too variable (cloudy days) and thus is worth less to electric power companies (would you buy a reliable or unreliable car?).
So, utilities want brown, not green power. But greenies got legislation passed forcing utilities to buy green anyway, and that’s the only reason why my utility buys my PV-generated electricity (I have negative-$100/month power "bills"). You can see how Georgia’s legislation works in my open letter to my utility.
Paul Wolff's solution here is fetching: Create a hybrid system. He would “light-switch-flick” a gas-turbine powered plant off and on throughout the day and night as PV, wind, and any other 100%-variable alt-energy sources kick on and off. The turbine would fill in the gaps left by the variable sources. This hybrid system would thus produce a continuous flow of electricity. More sun and more wind would result in less gas (a brown fuel) consumption. Stretched out over a large area and smart-grid coordinated, less and less gas (brown) power would be needed (stretched out PV would be especially useful because as clouds descend over one area of PV arrays, sun would shine over another; the same process could be envisioned for wind turbines).
I ask this question again, here. More on this topic here.
Here's another hybrid (but no cost data, which I find irritating).
Ditto for this geothermal hybrid. A friend writes about that: "One of the smart strategies with large solar plants is to co-locate them with existing power plants - generally natural gas. That way you can piggyback on some of the existing infrastructure such as power-lines, security, operators etc. Now we know solar + geothermal works too." My question, though, is this: I'm interested in the execution -- can, for example, gas turbines be tuned fine enough to stop when Solar PV electricity flow starts? Or, is there substantial -- and wasted -- overlap of some sort?
July, 2012 update on that geothermal hybrid plant here. And yes, of course it is being subsidized: "The DOE also has another reason for touting the project. The Stillwater plant received $40 million from the federal government's American Recovery and Reinvestment Act -- the same program that was put through the wringer last year following the Solyndra debacle. After receiving a $535 million loan guarantee through the Recovery Act in 2009, Fremont, California-based Solyndra filed for bankruptcy two years later."
Since we are co-investors in it, the government should require timely release of independently audited cost-data to evaluate whether it makes any economic sense. I sure hope it does.
July 2014 Update: Here's my email exchange with a grid operator friend:
Q. Is this bullshit or the potential standard nationwide?
A. That's standard. There are several ways to balance solar or wind but natural gas is towards the top.
Q. I guess my question is this: Is there an unacceptable amount of "on and off" or overlap waste?
Maybe I'm thinking simplistically. It seems to me that, while the solar arrays are feeding power into the Florida utility's grid, the utility must also run a gas turbine to fill in the variability gaps.
So, can it flick the turbine on and off all day as the clouds pass over, or must it run the turbine at a "best guess" constant level and somehow blend the turbine-generated electricity (base flow) with the solar arrays' electricity (variable)?
Whether the utility uses the flick on and off, or continuous run approach, is there not some inevitable wasted power, not to mention undue wear and tear cost on the natural gas turbine if it's being flicked on and off all the time?
I wish I knew more about the mechanics of this process.
A. There are a lot of moving parts to this question. It would be easier to explain over the phone but I'll try to give you some of the high level stuff.
When you imagine solar as a resource don't think of one plant by itself - think of all the plants in an area. I'd recommend looking up papers by Richard Perez and Tom Hoff. These guys explain how distributing solar regionally tends to smooth out the net output from a collection of plants. They find that weather isn't correlated once you get outside of a local area... This means variability isn't correlated. This means solar is more manageable than you might think. Germany has been dealing with peak solar events exceeding 50% and their grid hasn't crashed.
Here's are some of their papers and I attached a picture as well. I'd recommend the second paper for a good overview.
I'd argue that a healthy amount of PV on the grid would not ask natural gas plants to operate any differently than they are now. In this case we're asking them to operate not as a base load or peaker but as a mid-merit plant. There's nothing at all usual about this. Is there a little bit more wear and tear on the equipment? Maybe but think about it this way... If PV gets to be cheaper than natural gas then it's not PV's fault that NG has to move out of the way when PV is on the grid. We don't blame coal for being cheaper than natural gas?
What the conventional power plant operators have been finding is that they're ramping up and down a lot more than they used to and it's not nearly as bad as they expected.
As I said... there are a lot of moving parts here and I've only scratched the surface here.