Integrating nuclear and renewables

1 February 2016

The increasing penetration of renewables, especially wind generation, has dramatically changed the economics and realities of grid management in ways that now encourage some level of load-following capabilities. This case study examines the integration of a NuScale small modular reactor with a wind farm in Idaho. By D. T. Ingersoll, C. Colbert, Z. Houghton, R. Snuggerud, J. W. Gaston and M. Empey

Globally, France's PWRs routinely load-follow due to the high percentage of nuclear-generated electricity on their grid (nominally 75%). Canadian reactor units are also required to load-follow due to the percentage of nuclear power there and German reactors load-follow primarily because of a relatively high contribution of intermittent wind generation on their grid. In the United States, many nuclear plants currently operating were designed to load-follow and were originally outfitted with automatic grid control (AGC) features. However, the US Nuclear Regulatory Commission policy precluded the use of automatic dispatching for true load following, although they allow manual load-shaping if conducted by a licensed reactor operator. Load-following with nuclear plants, especially larger plants, requires complicated procedures and plant components that can tolerate thermal cycling.

The 1170MWe Columbia station in Richland, Washington, is the only commercial nuclear plant in the USA that performs routine power manoeuvring to load-shape. The load-shaping is required in spring to avoid excessive spill-over at hydroelectric plants in the Bonneville Power Authority (BPA) network. Increasing wind generating capacity in the BPA network may also require new load-shaping at Columbia.

Columbia performs short-term load shaping according to guidelines agreed by the BPA and approved by the US NRC. Generally, operators adjust reactor recirculation flow to manoeuvre the plant to 85% of full power and adjust control rods to drop power to 65%. They respond to down-power requests from BPA, which must be received at least 12 hours in advance of reduction to 85% power, 48 hours for reduction to 65% power and 72 hours for full shutdown. Power variations between 100% and 85% using reactor recirculation flow adjustments are relatively straightforward but can require many small adjustments due to the buildup and decay of xenon in the fuel, which is a strong neutron absorber. A cycle to 85% power, return to 100% power and subsequent reduction back to 85% can require as many as 17 different reactivity manipulations using recirculation flow and control rod movement.

Role of SMRs

There has been a growing interest in the USA and internationally in the developing and deploying of small modular reactors (SMRs), with capacity below 300MWe.

One SMR design has been under development in the USA since 2000 and is being commercialised by NuScale Power with the financial backing of Fluor Corporation and the US Department of Energy.

The NuScale power module is a 160MWt reactor core housed with other primary system components in an integral reactor pressure vessel and surrounded by a steel containment pressure vessel. The integral vessel (17.7m tall and 2.7m in diameter) contains the core (37 fuel assemblies and 16 control rod clusters), a central hot riser tube, a helical coil steam generator surrounding the hot riser tube, and an internal pressuriser. The surrounding steel containment vessel is 23.2m tall and 4.6m in diameter.

Up to a dozen power modules are co-located below grade in a large common pool of water, which provides passive containment cooling and decay heat removal. Each module has a dedicated turbine-generator to provide gross electrical power of 50MWe. A 12-module plant is expected to yield a net power of 570MWe to the grid.

Carbon free power project

In October 2014, Utah Associated Municipal Power Systems (UAMPS) announced a 'Carbon Free Power Project' (CFPP). UAMPS is a consortium of 44 utilities from Utah, Arizona, New Mexico, Idaho, California, Nevada, Oregon and Wyoming, and it plans to construct a NuScale plant on or near the Idaho National Laboratory Site.

Energy Northwest is expected to be the operator of the UAMPS CFPP plant in Idaho. It has experience in operating Columbia and of load-shaping on a large nuclear plant.

The nearby Horse Butte wind farm has a capacity of 60MWe. Commissioned in 2012, it is comprised of 32 Vestas V100 turbines.
The NuScale plant has features that enhance its ability to load follow. The key power management options, designated NuFollow, include:

  • Taking one or more modules offline for extended periods of low grid demand or sustained wind output;
  • Changing power levels at one or more modules to compensate for hourly changes in demand or wind generation;
  • Bypassing the module's steam turbine to respond rapidly to load or wind generation variations.

Each of these methods has a different response time and implications with respect to plant performance and operation. In general, their impacts are lower than in large plants due to the smaller reactor systems, smaller turbine-generator equipment, and fewer total systems.

Equipment in the NuScale plant is being designed for load-following operation to further reduce impacts from power cycling. For example, the module design and operating parameters allow reactor power changes using only control rod movement down to 40% reactor power, without adjusting boron concentration in the primary coolant. This improves the flexibility of the reactor without creating the additional liquid wastes associated with boron addition and dilution. The condenser is designed to accommodate full steam bypass, allowing rapid changes to system output while minimising the impact to the reactor, which can continue to run at full power. Finally, the multi-module nature of the NuScale plant and a staggered refuelling regime results in a plant configuration in which at least one module is near beginning of life. It is generally easier to perform power changes on beginning of life cores because the higher core reactivity allows better xenon override. The operator can use such modules to vary power for intermediate-term load-following, while the modules with higher burnup can be used for coarser power adjustments.

The Electric Power Research Institute (EPRI) maintains a compendium of guidelines and specifications for standardised plant designs, including specifications for load-following characteristics. This is the User Requirements Document (URD), which was updated (Rev.13)to include SMRs. The new version contains more aggressive load-following specifications to reflect the more flexible features anticipated for SMRs. The NuScale plant is able to meet all of the new Rev.13 requirements.

To understand how well a NuScale plant can mitigate variability from a wind farm, an analysis was conducted using wind generation data from Horse Butte. It is an especially challenging case study because its total generating capacity is small, which can result in short-term changes in generation that are significant fractions of the farm's total output. Figure 1 shows the frequency of 5-minute changes in output from the Horse Butte, expressed as percent per minute and normalized to the maximum wind generation during a seven-day period. This frequency distribution is compared to similar seven-day results for wind generation across the entire BPA system, which was roughly one hundred times larger than Horse Butte output. Most of the ramp rates for the larger BPA system were on the order of 1% per minute. In contrast, Horse Butte experienced a significant number of ramp rates up to 5% per minute and it requires a higher level of agility from the NuScale load-following response.

It should be noted, however, that the substantially larger total output from the BPA system results in a different challenge -- one of bulk replacement power. Over the same seven-day period, the absolute BPA wind output varied from zero to over 4.2GWe, and the largest five minute change was 136MWe. Output changes of this magnitude require a combined response from nuclear, hydroelectric and fossil.

Figure 2 shows a hypothetical scenario to demonstrate the integration of a NuScale plant with the Horse Butte wind farm. Included in the graph are: the US-averaged daily electricity demand profile (arbitrary normalisation) showing typical morning and evening demand peaks, the actual generation from Horse Butte for a single day in November, 2014, and the output from a single NuScale module that would be needed to meet the grid demand beyond what Horse Butte can provide.

Figure 3 shows two examples of how the NuScale module might yield the desired output. In one case the variation in NuScale output is entirely a result of turbine bypass, i.e. the reactor continues to operate at full power. The wasted power resulting from dumping main steam directly to the condenser is also plotted and closely tracks the power produced by the wind farm. Another approach that achieves the same demand-matching is to change the module's reactor power for coarse-level load shaping and use the turbine bypass equipment to provide the balance of load-following. This option is shown in the lower portion of the figure. This scenario reduces wasted energy and cycling of the power conversion equipment. However, the dispatcher must have an accurate forecast of wind power and the operator must be allowed to make changes in reactor power with minimal notice. Forecast and dispatch adjustments would need to be made hourly to support these types of operations.

Varying the reactor power introduces a number of operational considerations. It is better not to throttle back the nuclear plant or dump steam, but instead sell the excess electricity from the combined output of the wind farm and the gross output from the NuScale module to neighbouring utilities. However, this may not be an option. One method for selling such excess power, the Electricity Imbalance Market (EIM), is in the early stages of deployment. This new market was established to help balancing authorities cope with increased penetration of non-dispatchable energy. Participation in this market requires the unit to have automatic generation control (AGC), among other features. Adapting this to nuclear is not new, but it will require new approaches to accommodate regulatory policies.

An alternative to selling excess capacity is to use the power, either as electricity or steam, to support non-grid applications such as water desalination or chemical production. Using this hybrid approach, the combined wind and nuclear output can be optimised to meet grid demand and provide other valuable products without requiring the nuclear plant to vary its output.

Even with the NuFollow features, load following with a nuclear plant has operational and economic impacts. Reactor operations are affected least when changes in electrical output are accomplished by closing or opening the bypass valve to redirect main steam flow from the turbine to the condenser. This can be done much more quickly than adjusting reactor power. The drawback is that energy is wasted in the form of turbine bypass flow, and extended periods of high bypass flow to the condenser will increase wear on the equipment, resulting in increased maintenance and equipment replacement.

Adjusting reactor power for partial or full load-following requires a reliable wind forecast such that reactor power can be scheduled for daily or even hourly dispatch while remaining at a power level reasonably above that required to generate the expected electrical output. Turbine output is then trimmed via the turbine bypass valves for fine-tuned matching of output to demand. This option minimises the amount of wasted energy, which minimises the excess loading of the bypass equipment, including the condenser.

Other issues that must be considered when preparing for load following:

  • The fuel design must be optimised for resilience.
  • Routine thermal and operational cycling will cause components to degrade faster and may result in increased maintenance and lower module availability.
  • Although the reactor module is designed to vary power using only control rods, extended periods of low power operation may require boron adjustment.
  • Operator workload and maintenance, and hence overall staffing requirements, may increase.
  • Sustained operation using turbine bypass will place additional requirements on the cooling tower capacity.
  • Sustained operation of the module at low power may affect the refuelling schedule. This is less of an issue for a NuScale plant, because of its staggered refuelling strategy and the fact that refuelling activities will be conducted by permanent, in-house staff.

Ultimately, it will be economics, policy mandates and regulatory requirements that will drive the decision regarding the extent of load-following by nuclear plant.

NuScale is considering a move into other markets outside the Unites States. The current focus is on the UK where the company has opened an office and has been building collaborative relationships, including with Ultra Electronics, the Nuclear Advanced Manufacturing Research Centre in Sheffield and the National Nuclear Laboratory. It is also assisting the government in its assessment of SMRs for future deployment in the UK.

Further information

This article is based on a paper presented at ICAPP 2015 May 03-06, 2015 - Nice (France). The material is based on work supported by the US Department of Energy.

For more information, please contact Daniel T. Ingersoll ([email protected]), NuScale Power, LLC 1100 NE Circle Blvd, Suite 200, Corvallis, OR 97330. J. W. Gaston is with Energy Northwest and M. Empey is with Utah Associated Municipal Power Systems.

Figure 2. Example of NuScale module load-following to compensate for generation from the Horse Butte wind farm and daily demand variation
Figure 3. Two load-following options to achieve the NuScale module output shown in Figure 2 use only turbine bypass (upper graph), or use combination of reactor power manoeuvring and turbine bypass (lower graph)
Location of Horse Butte wind farm and Idaho National Laboratory
NuScale Power Module
Figure 1. Impact of wind farm size on relative generation changes

Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.