Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Nuclear Powered Peaker Plants?

  1. May 22, 2015 #1
    I was reading my textbook, Power Generation, Operation, and Control, third edition, and was thinking of the possibility of supplying peak load using a nuclear fission reactor. The reasoning behind my interest is that to supply the grid with nuclear energy at anywhere near 100% from nuclear you will need to be able to supply peak energy without a huge energy storage system.

    Naturally this presents a problem because reactors (due to the high thermal inertia of the core and delayed neutrons) tend to be slow to increase and decrease in power output and as such make excellent baseload, but absolutely terrible peaker plants, or at least all of the ones I have heard of. So I wonder if there are any existing fission reactors, or as Rickover might call them, paper reactors, that could serve as a peaker plant.
  2. jcsd
  3. May 22, 2015 #2


    User Avatar
    Science Advisor
    Gold Member

    Nuclear plants can load follow just fine, they were originally designed to do so. When operated in load follow mode, the reactor's power naturally follows the demand of the turbine in a self-regulating way and can adjust to fast transients instantaneously. In practice today they almost never load follow, but only for economic reasons - reactors have effectively fixed operating costs regardless of what power they run at, therefore running at anything below 100% is simply a waste of resources. But if you want to have a grid much more or entirely dependent on nuclear power, that is perfectly feasible from a technical standpoint.
  4. May 22, 2015 #3
    The main operational issue with load following is the induced xenon transients, especially in the physically large cores used in typical commercial power units. There are technical solutions to control the xenon effects, as QP points out the original designs incorporated these features. But these are (as far as I know) designed for load changes that occur over hours, such as the daily load variation. I think the minute to minute peaking would always be done by fossil units unless you want to include small reactors with physically compact cores, maybe something similar to the naval propulsion reactors. I don't think a power generator could license reactors like that, for one thing the enrichment requirements exceed the NRC's 5% comfort range.
  5. May 22, 2015 #4


    User Avatar
    Science Advisor
    Gold Member

    Xenon transients are mitigated by core design and operating procedures. Load changes occurring over seconds or minutes are regulated by the turbine and generator. The steam generator's water level acts a buffer for quick power changes. The only limitation to load following is reactor ramp rate restrictions, which limit power increases to ~3% per hour. So as long as you don't have large instant unplanned load increases (which should be controlled by the transmission authority), there's no major barriers to nuclear load following.
  6. May 22, 2015 #5
    The newer BWR plants were designed with a 30-40% load follow profile by varying recirculation flow in the core. Power responds in seconds, and using the automatic flux control system and load control features the main steam system is allowed to draw more steam from the core during these transients to meet demand. Ramp capability was a few % per second. For large load drops, the bypass valves could momentarily open to support the Load change while the rest of the plant slowed down.

    Flow control response is very fast in BWRs. Seconds is all it takes to make large power changes.
  7. May 23, 2015 #6

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    If the primary cost is fuel, peaker plants make sense. I don't think that's the case with nuclear power.
  8. May 23, 2015 #7


    User Avatar

    Staff: Mentor

    The French utility EdF does a lot of load follow (and frequency control) with their 900 MWe units. So it's not unfeasible with PWRs, or with BWRs that have some inherent load follow capability. Xe transients are not too worrisome if the power swings are small, e.g., a few percent. EdF does something like 5 to 7%, or maybe slightly greater, in the range of 10 to 15%. However, the reactors use grey rods and quarter core reloads. This helps with fuel performance concerns, particularly PCI, which is a major concern in load-follow. I've also seen German PWRs with ramp rates much greater than 3%/hr.

    Many large PWRs now load more than 40% of the core in a single batch, and most of that fuel will only see two 18 month cycles before discharge. Load-follow could be problematic under certain conditions. Fuel deconditioning-conditioning is a critical matter.
  9. May 24, 2015 #8
    So, as gmax137 stated, naval reactors actually increase and decrease in power output according to the needs of the ship, but what of this 5% enrichment comfort zone preventing such reactors being used for civilian purposes? That seems a bit odd considering it would (from my layman's perspective) be reasonable to use fuel rods with a higher enrichment than 5% on the outer edges of the core to make the power output of each fuel rod more uniform. It also seems odd because 5%, 20% and 60% enriched uranium all have two things in common, they are uranium, and they are useless for making nuclear bombs, thus it would be silly to restrict their use.

    Also, Astronuc, wouldn't using control rods instead of chemical shim for long term reactivity control cause sub optimal fuel burnup due to the varied neutron flux throughout the core?
  10. May 24, 2015 #9


    User Avatar

    Staff: Mentor

    The 5% enrichment limit was set a long time ago, although some special test rods with higher enrichments have been fabricated in the past. The concern with 5% has to do with criticality concerns in manufacturing plants and reactivity control in power plants. The 5% enrichment gets us where we need to go in terms of fuel cycle length and burnup. Most commercial power plants put high burnup fuel on the core periphery which operate at low power in order to reduce/mitigate neutron fluence to the core barrel and pressure vessel. Neutron irradiation leads to embrittlement of steels over the lifetime of the plant.

    Fuel assemblies are shuffled to different locations in subsequent cycles. Controlled assemblies will have lower burnups than their co-resident uncontrolled siblings, so the amount of control time is a factor. If load follow is restricted to weekends, then control would be less frequent, than say for daily load follow. I'd have to look into the details of grey control designs and use of chemical shim in conjunction with load follow to get a sense of burnup penalties. But I have seen examples of load follow in French and German PWRs and Swedish BWRs, and I'm somewhat familiar with load follow trials in the US back in the 1970s or early 1980s.
  11. May 24, 2015 #10
    Raising enrichment on the outer peripheral rods can affect overall core loading pretty significantly. Many plants were designed with the idea that the outer region will have low flow, and have core orifices installed to compensate for flow differences, so these would have to be replaced. Today's modern core designs for BWRs utilize the outer periphery as a low leakage neutron reflector which also happens to breed plutonium. If you had higher enrichment fuel on the periphery, it would have an effect on the next inner ring, which I've seen is typically higher output fuel. I'm not saying it can't be done, but the economics of fuel loading want you to use fuel for as long as possible, and the oldest fuel needs some place to go. Older fuel is more limiting on MCPR/DNBR and is also much more limiting on LHGR. (New BWR fuel can handle 14+ kw/ft, old fuel can't handle more than 5 kw/ft. PWR fuel starts in the 22-24 kw/ft range and old fuel can't handle more than 12-14 kw/ft). So the outer periphery being the oldest fuel allows you to use the fuel longer, minimize the number of fresh bundles you have to load, and improve cycle economy/efficiency.

    Due to the 5% enrichment limit, fuel designers have had to use other methods to extend cycle length. The simplest ones are using burnable poisons and part length rods to help load more fuel in areas that typically have thermal peaking limits. The use of water rods flattens the flux profile, and some fuel designs use two water rods now (these are rods that only pass water through them, creating a bypass flow path in the center of the bundle). More extreme techniques include thinning the pellet cladding gap and thinning the inner cladding itself to physically load larger pellets (hence more fuel by mass) into the same fuel rod.
    Last edited: May 24, 2015
  12. May 24, 2015 #11


    User Avatar

    Staff: Mentor

    An LHGR of 22 to 24 kW/ft would be for a transient condition. In a 17x17 fuel rod, that would be in dryout space. The smaller diameter 17x17 rods (0.360 inch/9.14 mm OD) have local peaks around 10.7 kW/ft, and the larger 0.374 inch/9.50 mm OD rods can go a little over 11.2 kW/ft with the same heat flux. In the upper half of the core, that could produce nucleate boiling. The even larger 14x14 or 15x15 fuel rods (~.422 inch / 10.72 mm OD) can go to higher LHGRs (~12.5 kW/ft) for the same heat flux.

    At high burnup (>55 GWd/tU), rod internal pressure and cladding corrosion/hydriding become concerns.
  13. May 24, 2015 #12

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2016 Award



    pretend i'm a dispatcher

    i have a nuke that makes kilowatts for two cents apiece
    and a coal plant that makes them for five cents

    I also have non-heat recovery type gas turbine were it costs me fifteen cents to make them.. but it can start up and shut down in just an instant.

    and i sell them all for the same twelve cents

    I'll cycle the gas turbine first and the coal second.....
  14. Jun 1, 2015 #13


    User Avatar
    Education Advisor
    Gold Member

    Folks are re-discovering in this thread why nearly all utilities want a mixture of different types of generation. Different types are handy for different contexts.

    In Ontario there is one pre-eminent form of power for load following. It is water-generated. You can go from 100% to 0% in one dispatch time unit. IIRC it's 10 seconds. And you can come back from 0% to 100% in the next. Sadly there is not a lot of new generation from water available. The waterfalls that can be used have nearly all been used, and people would be sad if we shut off Niagara. So the load-following plants tend to be natural gas. Before the coal was shut down it was coal.

    Load following certainly can be designed in to a nuclear plant. However, it is easier to leave out such a feature. A base-load-only design is likely to be cheaper and easier to operate. Margins are often tight in the electrical industry, so it is often not used.

    One interesting trick for some designs is to accommodate decreases in demand by dumping heat. So if the electrical demand falls faster than the reactor wants to decrease power, you send heat to the lake instead of to the turbine-generator. The usual CANDU design includes the ability to do this with something like 50% of full reactor power. So in principle a CANDU can load-follow up and down between 50% and 100% fairly easily. We don't like it though. It can be difficult to maintain our temperature limits in the lake where we get our coolant water.
  15. Aug 12, 2015 #14
    I think a lot of it has to do with the fact that fuel is a minor cost for a nuclear plant. Throttling down to save fuel consumption has only small effect on the Expenses side of the plants financial balance sheet. Therefore, reducing the number of MWhs the plants sells per month would require them to increases the per MWh price. If a plant runs at 100% half the time and 50% half of the time (75% capacity) then they'd have to bump up their rates by 1 / 0.75 = 1.33, or 33%, to compensate.

    So using nuclear for load following would be technically viable, but fiscally silly. It's best to run them balls-to-the-walls 24/7 and do load following with the cheaper plants those per-MWh costs are more determined by fuel expenses.

    (Yes, stretching fuel assembly life may allow you to run longer between outages; to the figure would be a little better than 33%. But not by much.)
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Nuclear Powered Peaker Plants?