Sunday, July 17, 2011

The Problem with Intermittent Sources of Electricity

As I've mentioned in prior posts, we should not compare electricity generating technologies based off of their Levelized Cost of Electricity. The reason is that different types of electricity can sell for different amounts of electricity. The goal of this post is to describe a little bit about how an efficient electricity grid operates, and then one can see the inherent problems of intermittent sources of electricity, especially wind.

There are requirements for the electricity grid to maintain a set AC frequency. The exact requirements vary from region, in most places the frequency is 60 Hz and the variation in the frequency can be no greater than roughly +/- 0.1 Hz. This frequency is maintained as the demand changes through a combination of day-ahead contracts to large power producers (such as nuclear and coal) as well as minute-to-minute contracts to small-scale peak-following power plants (such as hydro-electric and natural gas). In addition, the grid is maintained at 60 Hz through the use of spinning reserve. This is mostly steam and gas turbines that can pick up or take both real and imaginary power (Voltage and Vars) on a second-to-second basis.

This means that a natural gas power plant can generate revenue if it produces no electricity, and vice versa, a wind farm can produce electricity and generate no revenue. This is the main reason why the Rate of Return on Investment is a much better figure of merit when comparing competing electricity generating technologies.

The main reason why we are seeing so many wind farms is that electricity markets are being forced to include 'renewable electricity' into the electricity grid. The amount of 'renewable energy' changes from one state to another, and from one country to another. On a per kW-hr basis, wind is proving to be a cheap solution; however, this wind-generated electricity is not particularly useful. If there were no requirement for renewable electricity, then we would probably see most wind farm projects stop immediately. Intermittent electricity is not as valuable as electricity that can be generated predictably.

If I told you that I was selling apples at a fruit stand, but that I might not always have apples available when you come by the fruit stand. Why would you come to the fruit stand at all if the apples were the same price as at the grocery store?  The only reason why you would come by the fruit stand is if the price I charged for apples was so low that it outweighed the chance that you would have driven to the fruit stand as there were no apples.

Here, you can see the problem. How can a wind-farm bid into the day-ahead electricity market? Sure, you can predict the temperature somewhat, but it's really hard to predict wind speed. And since the power that a wind farm generates is proportional to the speed of the wind raised to the third power, any deviation from the day-ahead weather prediction is quickly amplified. This means that intermittent sources of electricity, like wind, will have problems competing in free markets until we have more sources of electricity storage. [Note that the intermittent nature of solar is not as bad as wind because solar panels generate electricity during the day when we would be needing to bring on peak-following power plants anyways. This is a different market than the day-ahead market, and the price of electricity is higher than the day-ahead market.] Also, countries like Denmark can survive with ~20% electricity from wind because they have large sources of electricity storage, i.e. the hydroelectric generators in Norway, Sweden & Finland. Some places in the US, such as Idaho and Washington, could possibly handle 20% electricity from wind.

20% wind power on the East Coast could be a significant problem because there are not that many sources of large scale electricity storage. Instead, including more than ~5% intermittent wind would require more natural gas peak-following / spinning reserve power plants. This means that we will have natural gas power plants spinning idling just so that we can claim to meet renewable energy standards. I see this as silly because wind turbines plus idling natural gas turbines will probably not bring down emissions of greenhouse gases. In fact, a study in 2009 by Kent Hawkins suggest that the greenhouse gas emissions of a wind turbine power plants (that requires natural gas turbines as back-up) is greater than the GHG emissions of just having the natural gas turbines provide. 

In Kent's own words, "The general conclusion is clear: industrial wind power does not produce the claimed benefits of reductions in fossil fuel consumption and CO2 emissions when up-and-down backup generation inefficiencies are taken into account."

Similar results were also found by Warren Katzenstein and Dr. Jay Apt of CMU, depending on the type of wind turbine and the location of the wind turbine.

But remember, the real figure of merit is the risk-adjusted, internal rate of return on investment when the economic damage from emitting pollution are included in the calculation. If wind turbines can eventually achieve high rates of return on investment (without requiring feed-in tariffs or other government mandates), then more power to them. I like cheap, clean electricity. We should be investing in those technologies that achieve high rates of return on investment, and by 'high' I mean greater than 10%/yr after adjusting for risk and inflation. But what I don't like is black outs, and I'm afraid that government mandates for intermittent sources of electricity will lead to more blackouts.

I see the main options for bringing down greenhouse gas emissions are: 1) nuclear 2) hydroelectric 3) geothermal 4) coal gasification with CO2 capture & sequestration underground, and 5) some small-scale local biomass combustion plants. Natural gas turbines are certainly economically viable without a CO2 tax  and provide less emissions of greenhouse gases than coal power plant, but the long term viability of natural gas is unclear right now until we figure out what is the economic and environmental damage of emitting greenhouse gases.

So, related to this discuss on electricity grids, I'd like to share some websites for the company that runs the largest wholesale electricity market in the world (which also happens to be electricity market for where I live currently.) The electricity grid is called the PJM (which stands for the original three states involved in forming the grid...Pennsylvania, New Jersey, and Maryland. The PJM website has free online training about: how electricity grids operate, how day-ahead electricity markets work, how the PJM maintains 60 Hz grid frequencies, how the PJM determines the marginal price of electricity, etc...

I think that the PJM has pretty much got things right. There is a free market for electricity generation, but there is a regulated market for electricity transmission and distribution equipment. The reason, I suspect, that the PJM is growing is that they seem to have solved the problem of how to run an efficient electricity grid that is cost effective.

The main thing I want to point out in this post here is that it's not as simple as just supplying power to the electricity grid whenever you want. You can damage people's electronics if you change the frequency of the electricity grid. We as a society are faced with the problem: Are we willing to trade off electricity grid stability (hence increase the chances of black-outs) for renewable electricity generation?

My own belief is that mandates and feed-in-tariffs for renewable sources is a really bad idea. In fact, I suggest that we need to remove laws requiring renewable electricity before we do some significant damage to our electricity grid. There are much better ways of lowering the emissions of greenhouse gases.

Instead of mandating renewable electricity, we need to figure out what is the economic damage of emitting greenhouse gases, and then we need to impose a fee per ton of CO2 equivalent emitted into the atmosphere. Of course, right now, the problem is that determining the economic damage per ton of CO2 is still an open question, but that's a story for another post.

So, here's the point of this post:  the main problem with wind turbines is that the wind provides an intermittent supply of electricity and the options for making it a base-load source of electricity (such as batteries) are prohibitively expensive (i.e. negative rates of return on investment.) Relying on an intermittent source of electricity for more than 5% of demand is not a viable option for a society that wants to avoid brown outs or black outs, unless they happen to live in a location with plentiful hydroelectric dams or are willing to have natural gas turbines sit on idle or stand-by. We should instead be smart. We should phase out feed-in-tariffs for renewables, tax pollution, and then force the electricity generators to compete on the open market.

1 comment:

  1. i think we need to focus more on producing kinetic energy as a constants resource for energy/power