Harnessing Variable Renewables: Where is the Beef?

The International Energy Agency (IEA) has just published a book titled "Harnessing Variable Renewables: a Guide to the Balancing Challenge." Variable renewables refers to solar and wind power generated electricity. Balancing refers to making the grid stable when solar and wind generated electricity are plugged in.

In the best of all possible worlds such a book would be available for free down load on the Internet, so that retired guys like me, who study reports about energy technology, but who cannot afford expensive books, can look at them. Unfortunately downloading this book from the IEA will cost me €80, and 80 Euros is a little steep for me, especially as I am probably going going to find the download useless once I finish writing my review.

Why so expensive? The Press Release answers that question nicely,

Written for decision makers . . .

That means that tax payers, and rate payers will be expected to pay the bill. Hay even the IEA is not above grubbing for money when it finds deep pockets.

"Harnessing Variable Renewables" appears to be a will it work study, intended for decision makers. A will it work study is one which examines a concept and determines whether it is viable in the real world. The IEA press release explains,

Power systems must be actively managed to maintain a steady balance between supply and demand. This is already a complex task as demand varies continually. But what happens when supply becomes more variable and less certain, as with some renewable sources of electricity like wind and solar PV that fluctuate with the weather? To what extent can the resources that help power systems cope with the challenge of variability in demand also be applied to variability of supply? How large are these resources? And what share of electricity supply from variable renewables can they make possible?

There is no one-size-fits-all answer. The ways electricity is produced, transported and consumed around the world exhibit great diversity. Grids can cross borders, requiring co-ordinated international policy, or can be distinct within a single country or region. And whether found in dispatchable power plants, storage facilities, interconnections for trade or on the demand side, the flexible resource that ensures the provision of reliable power in the face of uncertainty likewise differs enormously.

Thus the question decision makers who are addressed in "Harnessing Variable Renewables" will be asking is can the variable electrical output from renewables be balanced on the grid, how can it be balanced. First the conclusion of "Harnessing Variable Renewables" that some variable renewables can be balanced is not new. Even many renewable critics acknowledge that. The real question is not will it work, but how much will it cost to make it work. Costs are the beef in the part of the title that asks, "Where is the Beef." The IEA press release is silent about costs, and so we are left to wonder if the IEA book addresses cost issues. So far none of the reviews of "HVR" I have run across mention costs, yet I would hope that decision makers would would want to know how much a balanced variable renewable grid system would cost, before the give the go ahead to implement such a system.

Reviews of "HVR" are suggestive. For example offshoreWIND.biz tells us,

Assessing flexible resources

Harnessing Variable Renewables: a Guide to the Balancing Challenge lays out a four-step method for assessing existing flexible resources, which can then be used to balance increasingly variable supply and demand. Step one of this Flexibility Assessment (FAST) method assesses the ability of the different flexible resources to change their production or consumption; step two examines the aspects of the power system that will constrain them from doing so; step three calculates the maximum requirement for flexibility of a given system resulting from fluctuating demand and output from wind plants and the like; and step four identifies how much more variability can be balanced with existing flexible resources.

The book features eight case studies in which the FAST Method is applied to eight geographic areas with very different characteristics. The resulting analysis shows that each region has the technical resources to balance large shares of variable renewable energy.

Potentials range from 19% in the least flexible area assessed (Japan) to 63% in the most flexible area (Denmark). The IEA also assessed the resources of the British Isles (Great Britain and Ireland together), 31%; the Iberian Peninsula (Spain and Portugal together), 27%; Mexico, 29%; the Nordic Power Market (Denmark, Finland, Norway and Sweden), 48%; the Western Interconnection of the United States, 45%; and the area operated by the New Brunswick System Operator in Eastern Canada, 37%.

This range of results is due to the different flexible resources found in these areas. Norway, for example, has extensive hydropower, which is a very flexible resource; while Japan’s power plants, many of which run on nuclear and coal, are not as flexible (e.g. it takes longer for these sources to respond to fluctuations in demand).

Thus the assumption by the the IEA researchers is that grid managers will draw on existing grid resources to back up renewable energy. The study finds in effect that their are limitations to renewable grid penetration posed by reliance on existing grid resources. Indeed in most countries renewable grid penetration of less than 50% will be possible are using existing grid resources for balancing.

This means that for most countries a high renewables grid penetration system discussed in "HVR" is only at best a bridge to the 4/5th fossil fuel reduction that climate scientists envision us requiring by 2050. But can a renewables only system take us all the way to an energy future than produces only 20% of the CO2 produced today by fossil fuels?

There are clearly grounds for doubt, and clearly grounds for wondering it is going to be possible to produce 80% of all energy resources through reliance on renewables, how much will it cost to do so.

As I have pointed out decision makers need to know how much a balanced high renewables penetration system will cost. If "HVR" addressed the cost issue and offered good news, the pro-renewable reviewers, and indeed the press release would have mentioned that fact. If it did not address cost issues, then decision makers lack important information that are required to make appropriate decisions or the future sources of post-carbon energy.

Since I have no information on costs related to the "HVR" case studies, I will have to go with my own case studies, and since the United States Western Interconnect is one of the cases studied, I will note a previous Nuclear Green post, "The cost of carbon mitigation with renewables." In that post I discussed the Eastern Wind Integration and Transmission Study and , How do Wind and Solar Power Affect Grid Operations: The Western Wind and Solar Integration Study.These studies looked at 30% to 35% renewablea penetration of the two largest North American interconnects. Not quite as high as the 48% maximum penetration which "HVR" envisioned for the Western Interconnection. I noted that there were both significant connections about the actual CO2 mitigation that

As with all National Renewables Energy Laboratory reports, the WWSIS made no attempt to compare renewables costs and performance with nuclear power. But a relatively simple thought experiment can yield some very telling results. First we can assume that nuclear power will displace coal rather than CCGT. The Energy Information Agency estimates that the levelized cost of Advanced Nuclear will be 119.0, or about 12 cents per kWh. If nuclear displaces coal at that cost, the cost of displacing one ton of CO2 would be $119. Now let us take the 11% renewables case. The 2016 levelized cost of wind is 149.3, while the levelized cost of solar thermal is 256.6. Thus the average levelized cost of the 11% renewables is 159.08, and the cost of displacing a ton of CO2 with renewables is $159.0 + transmission costs and other hidden cost of wind generation systems, and the added CO2 emissions of fossil fuel wind backups kept spinning. plus the added CO2 efficiencies of fossil fuel generators used in load leveling and load following roles. Since wind is displacing relatively carbon efficient CCGTs rather than carbon inefficient coal fired generating plants. each MW of CCGT power displaced would produce 800 pounds of CO2, rather than a ton of CO2 produced by the equivalent electrical output of a coal fired power plant. Thus carbon mitigation with the 11% wind April scenario will cost about $400 + hidden costs or over three times as much as nuclear power would costs.

In the April 35% penetration case, wind becomes the predominate source of electricity on most days, and it displaces 2/3rds of coal generation capacity and all the CCGTs. Yet for the July 35% penetration case, wind failed to displace most CCGTs and no coal. Thus the WWSIS study data reported provided in sufficient information for understanding the the potential carbon mitigation costs . However it should be noted that the DoE study, Eastern Wind Integration and Transmission Study(EWITS) found that the cost of total system electrical output increased dramatically as wind penetration rose to 30%.

I concluded that,

Clearly then increasing wind penetration in the West will increase the level of carbon mitigation as well as its costs. It is also clear that wind displaces CCGTs before it displaces coal, and this increases the cost of carbon mitigation by wind significantly. Carbon mitigation with conventional nuclear would thus appear well over 3 times more cost effective compared to carbon mitigation with wind. The true cost effectiveness advantage of nuclear cannot be gaged until we know more about the hidden costs of wind, but the hidden costs appear to extract greater cost penalties at higher levels of wind grid penetration.

These conclusions suggest that Wind and Solar energy at high penetrations may not be cost effective tools for carbon mitigation. Taken all together, what we know about the contents of "HVR" is consistent with variable renewables being a questionable and expensive bridge to a low carbon future.

Keeping up with China: The Economic Advantage of Molten Salt Nuclear Technology

Last Friday, Brian Wang called attention to a Boomberg's article on Chinese nuclear cost. The Bloomberg's story reported that the French designed EPR would cost 40% less to build in China than in Europe:

Areva SA said the EPR nuclear reactor costs 3 billion euros ($4 billion) to build in China, 40 percent less than the price tag Electricite de France SA has put on building one in Normandy.

Nuclear Townhall, on the 26th, called attention to the competitiveness of both the Chinese and the Russian Nuclear Industries.  In addition to Russia and China, Nuclear Green has repeatedly called attention to the cost competitiveness of South Korea, and Indian nuclear technologies.  The Indians especially are following a well-charted path to an innovative low cost nuclear future.

American and European nuclear development can either proceed by following the cost-lowering paths being pioneered in Asia, or begin to develop low cost innovative nuclear plans.  Since low labor costs represent the most significant Chinese and Indian cost advantage, it is unlikely that European and American reactor manufacturers will be able to compete with the Asians on labor costs.  Labor costs for conventional reactors can be lowered by factory construction of reactor component modules, but the Chinese are clearly ahead of the West in that game.  Yet the weakness of the Chinese system is the relatively large amount of field labor that the manufacture of large reactors requires.

The Chinese system is to introduce labor saving devices wherever and whenever possible, but clearly shifting labor from the field to a factory still offers cost advantages.  The more labor which can be performed in the factory, the more labor cost savings are possible.  Other savings advantages are possible by simplifying reactor design, and lowering materials input.  Building a reactor with less materials and fewer parts lowers nuclear costs directly and indirectly.  Decreasing core size per unit of power output also can contribute a cost advantage.  Direct saving relate to the cost of parts and matetials, but fewer parts and less material also means less labor is required to put things together, since there is less to put together.  In addition a small reactor core structure, would, all other things being equal, require a smaller housing.  Larger cores mean more structural housing expenses.

While the Pebble Bed Modular Reactor has a relatively simple core design, the actual core is quite large, because of the cooling inefficiency of helium.  Thus, the simplicity of the PBMR core is balanced by its size, its total materials input, and the size of its housing.  The large core and housing requirements of the PBMR also adds to its labor costs, especially its field labor cost.  Thus while the simplicity of the PBMR core design would seem to suggest a low cost, this expectation is unlikely to be born out in practice.

Transportation limits ability to shift production from the field to the factory.  An analysis performed by the University of Tennessee's, and the Massachusettes Institute of Technology's Departments of Nuclear Engineering looked at the 335 MW Westinghouse IRIS reactor. The analysis found,

A rough estimate of the weight for a 1000 MWt modular reactor and its secondary system, similar to the Westinghouse IRIS plant, is taken as the summation of all of the major components in the analysis. Many of the smaller subcomponents have been neglected. The containment structure contributes ~2.81E6 kg (3100 tons). The primary reactor vessel and the turbo-generator contribute ~1.45E6 kg (1600 tons) each. The heat exchange equipment and piping contribute ~6.78E5 kg (747 tons). Therefore, the total weight of the major plant components is~ 6.39E6 kg (7047 tons).

The weight and width of the IRIS would be constrained by limits of barge transportation on the Tennessee and Ohio Rivers.  The report stated,

The Westinghouse barge mounted IRIS reactor modules were limited in size based on input from the University of Tennessee. The barge dimension limitations were established to be 30 meters (98’-5”) wide, 100 meters (328’-1”) long, with a 2.74 meter (9’) draft. These dimensions establish the barge maximum displacement at 8,220 metric tons. In addition, the barge(s) are limited to ~20 meters (65’-7”) in height above the water surface, so that they fit under crossing bridges and can be floated up the Mississippi, Ohio, and Tennessee Rivers as far as the city of Chattanooga, Tennessee. Further movement above Chattanooga is currently limited by the locks at the Chickamauga Reservoir dam.

The above barge displacement limitation will impose severe limits on how much structural support and shield concrete can be placed in the barge modules at the shipyard.  For example, the estimated weight of concrete in the IRIS containment and the surrounding cylindrical shield structure alone greatly exceeds the total allowable barge displacement. This however does not mean that barge- mounted pressurized water reactors (PWRs) are not feasible.  It does mean that barge-mounted PWRs need to employ steel structures that are then used as the forms for the addition of needed concrete after the barge has been floated into its final location and founded.

Thus for the IRIS, barge transportation presented problems, and rail transportation was unthinkable.  The core of the 125 MW B&W mPower reactor is rail transportable, but final onsite mPower assembly/construction became a significant undertaking, with a consequent increase in overall cost.  The core unit does include a pressure vessel and heat exchange mounted above the actual reactor, but many other mPower component modules must be transported seperately and assembled on site.

The IRIS project demonstrates the unlikelihood of whole small reactors being transported to the field ready for energy production without some field construction.  This might be possible, however, for mini reactors that are too small to be viewed as a plausible substitute for the fossil fuel powered electrical plants currently supplying electricity for the grid.  This then leaves us with a gap between the cost savings potential of factory manufacture, and the costly process of onsite assembly.  B&W the manufacturers of the small 125 MW MPower reactor still has not clarified what percentage of the manufacturing process would be factory based.  It is clear, however that B&W knows where it is coming from and what its problems are, as Rod Adams tells us:

I spoke in more detail to Chris Mowry and listened as he explained how his company's research on the history of the nuclear enterprise in the US had revealed that 30% of the material and labor cost of the existing units came from the supplied components while 70% was related to the site construction effort. He described how the preponderance of site work had influenced the cost uncertainty that has helped to discourage new nuclear plant construction for so many years.

What Mowry did not tell Adams is what percentage of the materials and labor costs will be shifted to the factory as mPower reactors are produced.  There have been hints that a significant percentage of the mPower manufacturing process, perhaps as much as 50% will still take place on site.  B&W still is working on the design of their manufacturing process, and thus do not yet know all of the details.  Clearly then more work needs to be done on controlling onsite costs.

Finally, a shift to advanced technology can lower manufacturing costs.  Compared to Light Water reactors, Liquid metal cooled reactors use less material and perhaps less labor, but pool type liquid metal reactors are not compact.  Compared to Liquid Metal cooled reactors, Molten Salt cooled reactor will have more compact cores.  Shifting to closed-cycle gas turbines will decrease construction costs.  The added safety of Molten Salt cooled reactors will increase reactor simplification, and thus further lower labor and materials related construction costs.

The recycling of old power plant locations will also offer some savings.  Decreasing manufacturing time will lower interest costs.

All in all there are a lot of reasons to expect lower nuclear manufacturing costs with Generation IV nuclear power plants, and at present no one has come up with a good reason for expecting Molten Salt cooled reactors to cost more than traditional NPPs.  The argument, however, is not iron clad.  Even if no one has pointed out plausible errors in it, we need to introduce the caveat that expectations frequently are not met.  It is possible, for example that the NRC might impose unreasonable expectations on molten salt cooled reactors.  Demanding, for example, that they include the same safety features as LWRs, even though they do not have many LWR safety problems.  But the potential savings on the cost of energy by adopting molten salt nuclear technology is substantial, and should not be ignored.

To return to the problem posed by Brian Wang, the problem of lower Asian nuclear construction costs.  If Europe and the United States cannot meet the Asian energy cost challenge, their economies will encounter a significant decline.  Because of Labor cost advantages, it is unlikely that Generation III nuclear plants will ever cost less to build in the United States or Europe than in Asia.  In order to keep the American and European economies competitive, the United States and Europe must adopt a low cost, factory manufactured nuclear technology.  Molten Salt nuclear technology represents the lowest cost approach, and is highly consistent with factory manufacture and other cost lowering approaches.  Couple to that the outstanding safety of molten salt nuclear technology, the potential for dramatically lowering the creation of nuclear waste, and the obstacles to nuclear proliferation posed by molten salt nuclear technology, and we see a real potential for keeping the American and European economies competitive, at least as far as energy costs are concerned.