On The Cost of Wind Reliability

Wind advocates point to a study by CRISTINA L. ARCHER AND MARK Z. JACOBSON
“Supplying Baseload Power and Reducing Transmission Requirements by Interconnecting Wind Farms”.

This study shows, wind advocates claim that systems of windmills, given proper grid interconnection, can supply base power. I will not dispute this conclusion, however, I do wish to point to some aspects of the Archer and Jacobson study that wind advocates often ignore, and which demonstrated that base wind power has undesirable features.

Archer and Jacobson based their model on wind data for the southwest. They chose 19 sites in Kansas, Oklahoma, New Mexico and Texas. They calculated that the average capacity factor for each site was around 0.45. Thus given the installation of a standard 1.5 MW windmill, the average output per windmill would be around 670 KWs. They were able to calculate that 79% of the time windmills from the combined sites would be able to produce at least 312 KWs, or 47% of the average output. Thus the base wind production from the 19 facility system would be .45 X .47 of the rated outputs of the 19 wind facilities or a little over 21% of rated capacity. Producing electrical energy 79% is about as good as your average coal fired steam plant. So this puts wind into the base capacity ball park. But consider how much this is going to cost.

Yesterday I used the figure of $2.5 Million per megawatt for wind costs. We would have to tack on the cost of the interconnecting grid, but we can ignore that for right now. How much would our wind base power cost? Well, if base power is 21% of rated capacity, we can get the figure by deviding the cost of rated capacity, by .21. At $2.5 million per MW, with a 1.5 MW system our wind mills will cost $3.75 million a pop. So how much is our base power going to cost us. Well $3.75 divided by .21 = $17.86 per base MW. Wow, now that is expensive!

Now Google wants to have 380 GWs of wind generators by 2030, how much of that if going to count as wind basic? The answer is 380 x .21 - 79.8 GWs. That seems like a very credible addition to our wind generations system until we realize that there is a serious performance fly in the ointment.

The capacity factor of wind generators drop on hot days. Really drop, and the hotter the day, the bigger the drop. This is a problem because it is hotter during the summer than during the winter. To make matter’s worse, electrical demand goes up during the summer. How bad does it get? In Texas and California on the hottest days, wind capacity factors drop to as low as .02 during periods of peak electrical demand. Thus when the system needs reliable base load capacity the most, wind base capacity is unavailable. If the .02 capacity factor for very hot days held for the entire Google 380 GW national wind system, the combined electrical output of the entire system during the hottest hours of the day would be 7.6 GWs. About the amount of electricity produced by 4 very large nuclear plants.

In order to have its desired wind generation system by 2030 we will have to build 360 Billion GWs of windmill generating capacity. This will cost about $900 billion. The same amount of money will buy 112 reactors. And those reactors will have a .9 capacity factor. Instead of the average output of 79.8 GWs of base power, you will get an average output of 101 GWs of base power from the reactors. But instead of only 7.6 GWs output during the summer peak demand periods, the reactors will and average of 109.76 GWs of output at any given time during the period of summer peak demand.

Summer wind power will never be able to compete with nuclear power as a reliable source of electricity, and any money spent on windmills would buy far more reliable power if spent on nuclear power.