Tuesday, August 18, 2009

The future cost of nuclear power considered.

Critics of nuclear power keep up a constantly attack focusing on the alleged cost of nuclear power. Recent critics have pointed to the cost of Canadian estimate of nuclear cost as proof that nuclear is too expensive. But such cost estimates are based on projections of the inflation in all power plant construction costs from 2002 to 2008 into the middle of the next decade. Between 2002 and 2007 the cost of wind generation facilities double. However inflationary pressures on materials and labor costs collapsed in 2008-2009. The 2002 to 2008 trend can be demonstrated by the following chart of the cost of wind turbines between 1997 and 2007 which demonstrates the effect of the inflation on wind costs:
Installation cost increases for wind installations also kept pace during the same period. Offshore wind is even more expensive than on shore, and if anything even more subject to inflation. Rod Adams points out that the recently completed German off shore Alpha Ventus wind farm, built for $357 Million and having a pathetic name plate capacity of 60 MWs, and a capacity factor that would be
- roughly the same amount of power that could be generated by a 20 - 25 MWe nuclear plant -
would seem to suggest that wind inflation, at least continues into 2009.

Secondly, I would like to point out that renewables are more vulnerable to materials and labor cost inflation than nuclear is. Last year in Bratislava two European energy researchers, J. Mišák, and F. Pazdera reported on the materials inputs into various energy technologies. (Comparison of Electricity Production from the Nuclear Energy and from the Renewable Energy Sources) They reported on materials input in kilograms per GWh of electricity produced in a year. The money figures were as follows
Technology Iron Copper Bauxite
kg/Gwh kg/Gwh kg/Gwh
Nuclear 457 6 27
PV 4969 281 2189
Wind (case 1) 3066 52 35
Wind (case 2) 4471 75 51

The Mišák & Pazdera report did not include concrete, but it is known that 500 tons of concrete go into the base of a 1.5 MW wind generator. The cost of construction materials reversed its historic upward trend in 2008. A future return of materials inflation would clearly impact renewables cost far more than nuclear cost.

Data om energy related labor input into the construction of power generating facilities is not easy to come by. I have seen estimates that conventional nuclear facilities require 10,000,000 to 20,000,000 hours of on site labor per billion watts (GW) of reactor capacity. Wind project labor input is difficult to come by. I reviewed the scheduled labor input for the Kittitas Valley Wind Power Project, a 181 MW wind facility built in Washington state in 2003. My interpretation of the Kittitas Valley facility labor plan suggests about 1300 hours of on site labor for every 1 MWs of wind generation capacity installed. The data for nuclear power suggests that about 10,000 labor hours are required per nuclear MW. This would suggest that wind has a significant labor cost advantage over nuclear but this is far from the whole story. First onsite manufacturing patterns are different for nuclear and wind. Many componants of wind systems, including the generator, the turbine blades, and the tower structure are manufactured off site. Labor input into component manufacture is not included in this assessment. A second issue is durability. While wind facilities can be expected to produce power from 15 to 20 years existing nuclear facilities are projected to produce power for 60 years, and it is anticipated that new nuclear facilities will produce power for 80 or even 100 years. A second consideration is te average time of power production. Typically nuclear facilities produce power 90% of the their generation capacity, while wind turbines typically generate power for a little less than 30% of their generation capacity. When these two factors are taken into consideration the labor input advantage for wind seems less obvious. There are significant issues of wind reliability, and in order to make wind generating facilities approach the reliability of nuclear facilities added investments involving large labor inputs would be required.

There are actually two reliability issues for wind power. The first is the intermittent nature of wind generation, and the hact that it cannot produce electricity on demand. The second is the relatively short lived nature of wind generation technology, the fact that wind generators need to be replaced every 15 to 20 years. We have seen that the first issue tends to increase wind related labor costs. The second issue means that there will be an unrelenting cycle of wind rebuilding with attendant labor costs. Renewables plans featuring large wind deployment, seldom include the cost of energy storage necessitated by the use of wind. The never include the cost of unceasing rebuilding of wind generation facilities of up to 5 times in a century. Thus in the long run the labor costs of wind construction will hugely exceed the labor costs that of nuclear power construction per MWh of power produced.

Nuclear however, does require a large number of skilled laborers to build. Compared to the relatively simple construction tasks required to build wind facilities, the tasks required to build conventional nuclear facilities are large and complex. In 2005 the NRC published a report on resource and labor inputs into new nuclear construction. This report indicated that of the 2400 workers who would be employed in the construction on a new nuclear plant, only 10% would be common laborers.

The rest included:
Electricians/Instrument Fitters
Iron Workers
Concrete Masons
Operating Engineers
Sheetmetal Workers

This would not be all of skilled professions represented on the site. Vendors and subcontractors will require their own staff on site. Quality control inspectors and Nondestructive Testing specialist required at each construction site.

Nuclear plant construction represents a monumental organization task. Past studies have indicated that almost 27% of worker's time on reactor sites is lost because of labor disorganization. The NRC report assumed
five year construction schedule from site preparations to commercial operations is assumed for each plant with 12 to 18 months for site preparation, 36 to 42 months for construction (first concrete to fuel load), and 6 to 12 months for testing and commissioning activities.
Thus labor costs present a mixed picture wit short run labor costs for wind running lower than nuclear, but long term labor cost running significantly higher. Furthermore, the introduction of off site reactor manufacture would create a revolutionary change in the nuclear labor cost picture.

Thirdly numerous nuclear critics have charged that the cost overruns problems of the Finnish Olkiluoto-3 reactor prove that nuclear power is too expensive. But does arguing from the Finnish experience in the construction of one reactor to all future reactors is to commit the logical fallacy of hasty generalization, and to discount the learning curve for nuclear construction.

Fourthly the Asian reactor construction experience suggests that reactor construction with reasonable cost is the rule rather than the exception. For example the cost of the South Korean Generation 3 APR-1400 currently runs $2330 per kW of capacity. Unlike the Finns, the South Koreans have built 12 reactors during the last twenty years, suggesting that the learning curve is alive and well in South Korea.

Fifthly I will argue that new reactor construction approaches now being developed in Asia - factor assembly of reactor modules and field assembly of modules - will lower reactor construction price and time. The American firm of Babcock and Wilcox, an old hand at reactor manufacture, plans to factory manufacture small reactor, saving both time and money in the process.

Sixthly, we need new rules on reactor financing that will lower interest costs. Electric companies should be allowed to charge customers for capital investment funds prior to the building of reactors. This would actually save customers money in the long run, and the need for new carbon free energy is urgently required to fight global warming.

Finally, the trend to smaller reactor will open multiple routes to cost lowering, and a shift to Generation IV reactors will accelerate this trend.


Soylent said...

Did you catch this post over at the capacity factor blog?: http://uvdiv.blogspot.com/2009/08/red-hot-irony.html

Duncan said...

You write these great analyses, but then leave them full of tpyos and/or grammar errors that make them frustrating to read.

Your blog is authoritative enough that it deserves to be checked for careless mistakes.

Anonymous said...

Every technology has a learning curve but it's difficult to determine that for nuclear in most countries due to the obscuring effect of hostile regulatory environment. Delays during construction, for example, can result in huge cost overruns. France shows that a favorable legislative environment cuts total project cost dramatically.

For investors, one other problem is that the price for nuclear projects appears to be on a diverging course, making it more and more difficult to determine the competitiveness and increasing risk of large cost overruns.

This also means that some projects are likely to be much cheaper than the figures that many critics use today.


Blog Archive

Some neat videos

Nuclear Advocacy Webring
Ring Owner: Nuclear is Our Future Site: Nuclear is Our Future
Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet Free Site Ring from Bravenet
Get Your Free Web Ring
by Bravenet.com
Dr. Joe Bonometti speaking on thorium/LFTR technology at Georgia Tech David LeBlanc on LFTR/MSR technology Robert Hargraves on AIM High