By: Donald Jones, P.Eng. - retired nuclear industry engineer - 2011 January 20
The Ontario Power Authority's (OPA) 20 year Integrated Power System Plan (IPSP) was filed with the Ontario Energy Board in 2007, then withdrawn. The government has now prepared the 2010 Long-Term Energy Plan (LTEP) which it will pass to the OPA for implementation as the revised IPSP. This plan has an energy supply mix that depends on nuclear, hydro, gas, wind, solar and biofuels. The plan calls for nuclear to meet 50 percent of demand but says that having more than this would cause problems for the grid. See APPENDIX for more information. It does see a new 2,000 MW station at Darlington, hopefully two Generation III+ ACR-1000s, or failing that, four Generation III Enhanced CANDU 6s.
Nuclear is regarded as dispatchable by the Independent Electricity System Operator (IESO) but in practice the present units have shown themselves to be not flexible enough for frequent load cycling and certainly not for dispatchable load following. The grid will depend on gas and stored water hydro to provide dispatchable load following and intermediate load generation. Wind would not be practical if dispatchable gas were not available since hydro is fully committed and will become even more valuable after the loss of very flexible coal.
The Ontario grid has an installed capacity of around 35,000 megawatts and in 2009 nuclear provided 55.2 percent of Ontario's generated electricity (from around 11,400 MW of installed nuclear), hydro 25.5 percent (from nearly 8,000 MW of stored water and run-of-the-river hydro), gas 10.3 percent (from around 8,500 MW of gas and oil), coal 6.6 percent (from about 6,000 MW of coal - to be all phased out in 2014) and wind 1.6 percent (from about 1,000 MW of nameplate wind). Other fuel types (biomass, solar etc) gave 0.8 percent. Wind generation is expected to drastically increase over the next few years under Ontario's Green Energy Act to around 5,000 MW and eventually to more than 8,500 MW in the early years of the 20 year LTEP when transmission connections are available.
Over the next 15 years or so Pickering A and B will be permanently shutdown and Darlington, Bruce B and units 3 and 4 of Bruce A will be refurbished. Refurbished Bruce A units 1 and 2 will come back on line in late 2011 and operate for 25 to 30 years. Since the government mandate under the LTEP is to limit nuclear to 12,000 MW in order to provide 50 percent of Ontario's electricity it means that only half the output of Pickering A and B, 2,000 MW, will be replaced by new nuclear build on the Darlington site. This new build, which will be operating for 60 years, must at least be able to provide daily and weekend load cycling by manoeuvring reactor power, combined with steam bypass if necessary, and ideally be able to respond to frequent load following dispatching. According to AECL its new reactor, the ACR-1000, can cycle daily down to 75 percent of full power, periodically down to 60 percent and when required, for example on weekends, down to 50 percent.
Having the capability to operate the grid at the expected minimum demand loads will be more difficult than meeting the maximum loads. The LTEP points out that during periods of Surplus Baseload Generation (SBG) having too much nuclear will be a problem. SBG occurs when baseload generation, from nuclear, must-run hydro, combined-heat-and- power, and wind, that cannot be reduced for technical or contractual reasons, exceeds demand. Let's test this statement. Instead of adding 2,000 new megawatts at Darlington let's see how the grid would handle 4,000 new megawatts made up of four ACR-1000s, for a total nuclear capacity of 14,000 MW. Assuming no load cycling capability of the refurbished 10,000 MW and assuming that must-run hydro is around 2,000 MW this would give a total baseload generation of 16,000 MW from nuclear and must-run hydro combined. This ignores intermittent wind and an expected 1,000 MW or so of combined-heat-and-power which would be added to the nuclear and must-run hydro so increasing the amount of surplus baseload supply. Water spilling, if necessary to reduce hydro-electric generation, may be restricted by water management regulations, environmental concerns and public safety around the spillways which would increase the amount of must-run hydro generation. The overnight output of the ACR-1000 generation can be reduced to 75, or maybe to 60 percent of full power, so the combined overnight baseload generation could be reduced to 15,000 MW or even to 14,400 MW. This means that under SBG conditions (typically overnight in the spring, when minimum demand has been around 11,000 MW) nuclear units would have to shutdown if this total generation, which ignores wind, still exceeded demand. Of course in fifteen years or so, after refurbishment and new build, it is more than likely there will be an increase in minimum demand but this does show why adding more nuclear could make operation of the grid more difficult. Although the existing 1,200 MW of wind on the grid operates under must-accept contracts new wind supply under the feed-in-tariff program will provide incentives for wind to be shutdown during periods of SBG.
This means that if more new nuclear is to be added to the grid, to replace non-renewable fossil fuels, the turndown has to be significantly improved to handle periods of SBG even if this goes against the economics of nuclear operation. Any engineering opportunity to improve the flexibility of the 10 units at Darlington and Bruce while they are being refurbished should be pursued. Bearing in mind they are going to be around for another 25 to 30 years after they restart such work could pay off if it will make room for more nuclear units on the grid that could bring nuclear up to at least 14,000 MW. Amongst other things this could include improving the robustness of the steam bypass system so plant outputs can be considerably reduced during SBG periods. Steam bypass is not the best way of reducing plant output but as Ontario demand builds up over the years the bypass system would be used less and less. Over the next fifteen years or so periods of SBG will be fewer anyway as nuclear undergoes refurbishment. During past periods of SBG Bruce B units have reduced output to around 60 percent of full power using steam bypass even though the bypass system was not designed for this kind of frequent use. The amount of power reduction using steam bypass is restricted by environmental regulations that limit the temperature difference between lake water entering and leaving the station. The new CANFLEX fuel may be of some help in the future.
Alternatively nuclear can remain at full power, where it is most economic, and any surplus generation can be used for demand response loads like thermal storage and hydrogen generation. Hydrogen would have many uses including day time electricity generation to help meet peak loads, and by using hydrogen produced overnight from cheap electricity this may be economic especially when increasing natural (with more and more shale gas and liquefied natural gas) gas prices will put up the price of generation during the day. This would be a massive undertaking but at least it deserves an economic study, even using today's reactors. Small modular thermal reactors may be on the scene within the 20 year time frame of the LTEP or shortly thereafter and they would have to be considered. Commercial Generation IV fast reactors are unlikely to be on line until the second half of this century but other Generation IV reactors could be available sooner, including those high temperature reactors that could be used to produce hydrogen more efficiently, but who knows exactly when either would be ready for deployment in Ontario. All this and more needs to be looked at particularly as the natural gas supply cannot be relied upon indefinitely. The LTEP shows that lack of nuclear flexibility is resulting in the use of more polluting natural gas-fired generation and the amount of this generation must include that needed during the long nuclear refurbishment outages. (Aside: This need for more gas-fired generation could be used as a rationale to incorporate more wind into the system in the future.)
The availability and cost of conventional and controversial shale gas in the next few years is unknown yet the Ontario government is betting our future on enough affordable gas being available to power our electricity generators, heat our homes, supply our petrochemical industry, supply our industrial sector, meet our potential transportation needs, and meet the demands of all other north American users. Shale gas, and imported liquid natural gas, will only replace the declining reserves of gas from north America's conventional sources, and at higher cost and with higher life cycle greenhouse gas emissions. Shale gas in particular has life cycle emissions that can exceed those of coal. As supplies run down costs will spiral upwards. Without gas (or coal) wind would be useless for generating grid electricity.
With nearly 80 percent of their electricity coming from nuclear the French seem to handle the economics and technicalities of load following and load cycling reasonably well, using their fleet of 58 pressurized water reactors that are very flexible during the early part of the interval between refueling outages, which occur every 12 months to two years. Refueling outages are managed to ensure overall grid flexibility.
The LTEP is much too sanguine on nuclear. In the immediate future the Ontario government should instruct the OPA to look at what can be done during the refurbishments of the 10 units to improve unit flexibility. It must also produce a Long-erTerm Energy Plan to look into how to go about integrating much more nuclear into the system. New plants should be built so that they are ready for when demand builds up, as long as they can be turned down at night. The grid can rely on precarious natural gas for only so long.
The following is taken from the section on nuclear supply in Ontario's Long-Term Energy Plan - 2010 November.
Ontario will continue to rely on nuclear power - at its current level of contribution to the supply. Nuclear generation is ideally suited for providing baseload generation because of its unique economic and operating characteristics. Nuclear plant operational design and economics depend on the plants being able to operate steadily throughout the year. A generation mix of 50 per cent nuclear combined with baseload hydroelectric generation is sufficient to meet most of Ontario's baseload requirements. If nuclear capacity beyond this were added, the hours in the year in which nuclear capability exceeded Ontario demand could substantially increase. Under such surplus conditions, some nuclear units might need to be shut down or operate differently than intended. This could lead to significant system and operating challenges and so therefore, generating too much nuclear is undesirable.