Oil Drum Debate with Nick on Intermittency

I tend at times to get involved in online dabates. At the very least dentes are methods of self education. I do listen to what the other side says, and sometimes I change my mind, if confronted with a question to which I have no good answer. I have, for example, modified prior views on the genetic basis of behavior, although the issues are far more complex than the genetic determinist would allow for.

My oil drum debate with Nick, was triggered by claims of another commenter, Chris. Chris claimed that renewable power sources were inherently more efficient than nuclear. Part of my responce was to point to the problem of intermittency with renewables.

Nick responded to that comment with a statement that began:
"Intermittency is greatly exaggerated as a problem for renewables. That said, I have to say I find this argument silly, and intellectually dishonest on both sides."

"Nuclear advocates point to non-problems (like intermittency) . . ."

I responded to Nick:

In my book claiming that intermittency is a non-problem is where the intellectual dishonesty lies. Intermittency is a non-problem, only for supplemental power. Intermediacy becomes a problem once renewables are offered as base or even peak power. There are three methods of coping with the problem of intermittency, back up generation with fossil fuels, energy storage, or massive renewable generation redundancies. Each method requires the building and maintenance of duplicate facilities, in order to cope with the intermittency.y.

Nick responded back by apologizing for saying that my point about intermediacy is intellectually dishonest, He then stated that:

There are more, and better, methods of coping than that. First is connecting grids at the edges in order to reduce the variance (aka intermittency) of various generation sources. If you make your grid effectively large enough, variance becomes manageable. This is relatively low cost, as you don't need the kind of earth-girdling cables envisioned by another TOD post recently, you just need to connect grids at the edge.

2nd is demand management: if you make meters dynamic and time-of-day sensitive, and have a large % of demand which is flexible and schedulable, generating variance becomes much easier to deal with. For instance, if you have 100M EV's (as we will in the US, in one form or another), you have a very large demand which can be shifted at essentially no cost.

3rd is location and source tuning: using both the predictable patterns of certain negatively correlated wind locations, and the negative correlation of wind and solar to reduce system variance. This costs nothing but the time spent in planning and regulating installations, something system operators should be doing now. - Nick

I responded to Nick:

Nick, I am pleased that you are willing to dialogue about this. Let me address you your preferred methods of coping with intermediacy. I will take them out of order. You suggest demand management as a coping method. Demand variation is something of a problem for nuclear power, so I have looked at the issue from that viewpoint. There are flexible demands, that is demands that can be shifted to other times, and methods of altering consumer behavior, however some demands will remain inflexible, For example the demand for power for air conditioning in the summer. Texas is as you are probably aware quite hot in the summer, and the heat is actually a health hazard. There is no why you can get people at home to turn off their air conditioners on hot summer nights in Texas.

You suggest connection to the grid as a means of providing electricity, but how is the power going to be generated. The sun is down so solar sources are out of the picture. Texas summer wind potential drops too. I believe that the capacity factor for Texas wind generators is below 17%. But worse, the wind is highly variable, and wind output may drop to zero all over the state. Thus wind generation would provide Texas with the worst of all possible worlds, high generator redundancy plus unreliability. If your wind system is built to compensate for a capacity factor of 17%, you will build so many windmills that it would be far cheaper to just build all nuks. Thus the grid seems to provide no solution to the problems of renewables.

Finally you suggest location and source tuning. I assume that in my thought experiments on renewables. I assume that wind generators will be located in the best locations, and that solar generation will be located in the Southwest. But putting solar in the Southwest will not stop the sun from going down. And building windmills in the Texas Panhandle is not going to give you the power you need to keep Texas Air Conditioners running all night long on Texas Summer nights.

My argument for charging the energy input of backup fossil fuel plants to renewables is this. A renewable power system, as I believe I have just successfully demonstrated, is not capable of always meeting electrical demand. Ergo, it requires alternative sources of electricity, lets call them the green crutch, as backups. Now if you went with an all nuclear system, the crutch would not be needed. We can throw the crutch away, by recognizing the limits of renewables, and not expecting more than they can deliver.

The limits are these. Wind will never be reliable enough to serve as base or peak power. Therefore wind of of very limited utility outside reducing CO2 emissions. Wind can only serve as a supplement to fossil fuel power, If fossil fuel power needs to go away, there is no rational for wind.

Solar is good at providing daytime power. But solar power imposed penalties and expenses related to materials demand, land use and security. At present the capitol costs of solar exceed that of nuclear, and the cost of materials for building solar facilities are rising. Therefore solar is a candidate for daytime peak power, but may never fulfill its expectations. Furthermore, if solar power is marginal in the Southwest, its performance will be far worse in other parts of the country.

My conclusion then is that at present, only nuclear power can fully substitute for fossil fuel power sources. Furthermore, without some significant breakthroughs in the generation and storage of solar power, the substitution of nuclear for fossil fuels is inevitable.

Afterward: I by no means think that Nick's remedies are bad in themselves, it is just that they don't do for renewables, what renewables advocates say they will. Interconnecting the grid is already working for much of the United States. For example the Southeast exports power to the North in the winter, while importing power from the North during the summer. That helps both areas get by without adding seasonal capacity to meet peak demand. Such a swap is not so easy for Texas. Much of our peak generating capacity is from natural gas. Texas consumers are burning gas in the winter to keep warm. The gas producers would rather sell gas to consumers in the winter, than to sell it to electric companies. The fact that the gas producers have a surplus of natural gas in the summer means that gas is on the market for power production during the summer in Texas. Adding an interstate grid connection to the Texas power system does not help Texas meet peak summer demand. Summer is a peak demand time for the most likely exporters of power to Texas, while power generation by wind drops all over the country.

Update: “What we know, is the wind blows sufficient for these windmills to be producing about 35%, perhaps 40% of the time. So the paradox of building windmills is that you have to build a lot of ordinary power stations to back them up and those are going to be almost certainly gas in the short to medium term and that’s what’s required. If you ask the question who’s making sure that there’s enough gas stations out there to back up the windmills the answer is nobody.” - Dr Dieter Helm, Energy Economist and Fellow in Economics, New College, Oxford

Renewable Energy: Wasted Days and Empty Nights

One of the I had to learn when I began to read discussions of renewable sources of electricity is the vocabulary of power. Generators of electricity can be divided into three different types. There are base power generators, these are the units that can be counted on to generate electricity, all the time, except during periods of maintance. A base power unit can be counted on to generate electricity 60% to 95% of the time. Examples of base electrical generators include coal and natural gase fired electrical generating plants, nuclear power plants, geothermal power plants, hydroelectric power generators.

Peak power generators represent a second class of generators. Peak power generators can provide electricity when demand is high. Usually peak demand is during the day time, when stores, offices and factories are using electrical energy for their operation. In the North peak demand may occure on winter nights, when cold waves drive demand for electrical heating. The popularity of lighted Christmas displays is another source of winter night time peak demand. In the South peak demand is associated with summer use of air conditing. Although the absolute peaks of air conditioning driven summer demand are in day time, the air may remain hot at night, bringing about elevated demand for electricity after dark. Sources of peak demand electricity may include pump storage facilities, as well as reserve conventional electrical generators.

Supplemental power is a third category of electrical generator. Supplemental power cannot be counted on to be avaliable on demand. Some renewable sources of energy like wind powered generators cannot be counted on on to provide electricity on demand. Wind power is avaliable when the wind is blowing. Other supplemental generators, for example solar voltaic panels, are only avaliable at spacific times of the day. Since the "fuel" for wind generation is free, and they do not generate greenhouse gases, the use of supplemental renewable power sources, may be desirable to replace base power when the supplementals electricity is avaliable.

Electricity from variable renewable energy sources has to be ballanced, Electrical demand stays constant whether the sun is shining or behind a cloud. For a simple solar or wind generating system a battery might do. Feed the electricity into the battery and then draw it out when you need powet. pump storage facilities are sort of super batteries. You can pump water up a mouintain when the sun is shining, or the wind is blowing, and then let it flow down the mountain through a generating turbine, when consumers want electricity. A more common way to ballance variable renewable power is by using power from a generating source that can quickly be drawn on as the wind drops, or clouds cover the sun. Denmark draws on electricity generated at Norwegian dams to ballance its wind generated electricity.

Electrical generation has to be able to come on line quickly in order to ballance the variations of wind generated electricity. If you have access to hydroelectric power and plenty of water, that will do fine. but California does not have enough hydroelectrical power to ballance its wind generators. Balancing power usually come then from convintional fossil fuel powered electrical generators. These are called back up generators. In order to be able to supply power quickly, the back up fossil fuel generators must be kept running all the time. Power companies call running back up generators "spinning" them. Of course when you run a back up generator, when you spin it, you are burning fossel fuel, and you are producing geenhouse gas.

Opperating an electrical system that draws on wind generated electricity or solar pannel generation, requires generating back ups. This is called redundancy. How much redundancy do you need? That depends on how much electricity you are trying to get from your vaiable renewables. If you are just operating a few wind generators, you may need quite a lot. If you have a lot of wind, wind generators will produce a lot of electricity. Wind generators are rated by the mazimim amount of power they can generate on a very windy day. Unfortunately in most places the wind rarely blows that hard. In a light brieze, your wind generator may only produce 20% of its rated power. If you are commited to produce a certain percentage of your electricity by wind generation that may mean that you want to buy more wind generators to back up your generating capasity. Five wind generators operating in a light breeze will supply the same amout of power as one generator operating in a heavy wind. If you have light breezes 25% of the time, then you may need to buy five wind generators in order to provide the amount of wind generated electricity you want.

Now suppose no wind at all blows 25% of the time. How many wind generators do you need? Well no number of wind generators will get you what you need. So you will need 100% of your electricity from back up generators. How often do you need back up generators? How often have you been out in the wind, and saw a hard wind turn into a breeze, or even simply die away to nothing? You basically have to keep 100% of your back up capacity spinning all the time with wind. In desert environments the sun shines almost all the time, clouds are fairly predictable.

There is usually some warning if the sky is going to cloud over. So spinning back up is not nearly as critical with Solar power as it is with wind. You still need back up capacity with solar, and you need variable online power for ballancing, even if there are only a few clouds in the sky.

There sunlight is much more tricky than you might imagine. The problem of SV intermittency is discussed in "The Character of Power Output from Utility-Scale Photovoltaic Systems," by Aimee E. Curtright and Jay Apt, (Carnegie Mellon Electricity Industry Center Working Paper CEIC-07-05: http://wpweb2.tepper.cmu.edu/ceic/papers/ceic-07-05.asp). The paper is unfortunately password protected.

A quote from Curtright and Apt,
"The intermittency of large-scale PV power for four sites in the American southwest desert is significant, even during daylight hours. These data also imply that site diversity over a ~280 km range does not dampen PV intermittency sufficiently to eliminate the need for substantial firm power or dispatchable demand response. The high correlation between geographically dispersed arrays may indicate that high, widespread clouds are responsible for a portion of the intermittency. Observed rapid and deep fluctuations at time scales of 10 seconds to several minutes may indicate that a component of the intermittency is due to low, scattered clouds with significant opacity. We observe a number of examples of output power rising above nameplate capacity before and after deep drops in power. This may be due to focusing of sunlight around the edges of low clouds."

If the problem exists in the relatively cloudless southwest, how much worse is it going to be in the South East, when there can be a lot of clouds in the sky on "sunny" summer days?

The use of solar power for peak power is further complicated by demands for peak power at night and under adverse weather conditions. You also have to factor in the demand for peak electricity on cold, snowy winter days, and after dark in much of the country. Solar cells don't generate electricity when covered with snow. On hot summer days in Texas, AC demand continues all night. It is often in the upper 90's at midnight after hot Texas Summer days. We need peak power at night.

One other factor that can effect the reliability of solar power is air born dust. Dust can effect solar power both by blocking sunlight on dusty days, and by dust deposits both on solar panels, and on mirrors in Solar thermal mirror arrays. NASA has predicted a return to dust bowl conditions in the Southwest.

You need back up generators for the 12 hours a day the sun does not shine. Although advocates of solar power talk about night time backup systems, they require redundant generating capacity. Thus with stored solar back up, capital costs will be more than doubled. If you use none solar backup those generators will either burn fossil fuels, and thus they will emit CO2, or they will be nuclear powered. But why, if you built nuclear power plants, would you only want to run them at night? It would be cheaper to run them all the time and not worry about solar power.

Either a ST or an SV system ends up being expensive, vulnerable to clouds and dust storms, and is going to occupy a lot of territory, and damage a whole lot of vulnerable desert habitat. But hay, damaging the environment is no big deal to greens.

Building so much redundancy into the electrical generating system costs money. Taking land from other uses and committing it to renewable generated electrical power also has economic costs. Electrical grid systems that gather and transmits electrical energy from far flung renewable energy "farms" and transmits it to cities also costs money. And there is of course the cost of fuel to keep back up generators spinning. Is there a cheaper way to do thing?