Royal Military College of Canada

Energy and Nuclear Science

Optimal Design of CANDU Fuel Bundles for Advanced Fuel Cycles

By: Dr. Hugues W. Bonin

The general objective of this research is to use modern optimization techniques for the design of CANDU fuel elements (bundles) intended for the use of advanced fuels in the next generation of CANDU nuclear reactors. Although specifically designed to use natural uranium dioxide as fuel, the CANDU reactor was seen from the start as being able to use other kinds of fuels such as thorium and plutonium oxides mixed with natural or enriched uranium. The presence within the fuel bundles of mixed oxides brings significant benefits in terms of residence time of the fuel elements and the average discharge burn-up (energy released by the bundle). On the other hand, the advanced fuel compositions in the bundles cause undesirable effects such as excessive power and temperature densities in parts of the bundles and large surface heat fluxes through the cladding of the fuel rods when the present 37-rod fuel bundle design is used with these advanced fuels. These problems can be addressed through derating the whole reactor power output, but this alternative is economically unacceptable. An alternate approach is through a redesign of the CANDU fuel bundle, by changing the number of fuel rods, their dimensions, their positions within the bundle and their fresh fuel compositions, either singly or all at the same time. AECL has already initiated research and development of a new, improved CANDU fuel bundle, resulting (through design methods not based on modern optimization techniques) in a 43-rod bundle called CANFLEX. This bundle was designed primarily for Low Enrichment Uranium (LEU) fuels, but was found to be acceptable for some other types advanced fuel cycles.

The purpose of this research is to go beyond the CANFLEX bundle design and to use optimization methods to arrive to new CANDU bundle designs that would be based upon fuels such as DUPIC (Direct Use of Uranium from Pressurized Water Reactors (PWRs) In CANDUs), thorium-based fuels and advanced fuels containing plutonium recycled from discarded nuclear warheads from the USA and Russia. The research would use and refine the steepest-descent-based optimization methods used in the past successfully to determine the optimal fresh fuel composition distribution of LEU-based fuel and plutonium-based mixed oxide fuel in 37-rod bundles to the case of the advanced fuel cycles of today's and tomorrow's interest: LEU, DUPIC, weapons' plutonium and thorium. It is planned to extend the optimization method to the geometry of the bundle (number, dimensions and placement of the fuel rods) in addition to solving the optimization problem of determining the best fresh fuel composition of the various fuel rods within the bundle. The optimization problems are formulated to include safety-related constraints such as maximum reactivity insertion, and thermal-hydraulic based constraints such as maximum power densities, radial power distribution and maximum surface heat fluxes within the cladding.

The simulation of the CANDU bundles will be carried out using the WIMS-AECL computer code already available at RMC. Depending on the resources available, it is among the objectives of the project to extend the computer modelling of the bundle to include thermal-hydraulic parameter calculations using specialized codes such as ASSERT for the analysis of the heat and flow transfer properties of the new bundle designs. In addition to the DUPIC fuel, the study aims at investigating the cases of several uranium enrichments for the LEU fuels, several fresh fuel compositions for the mixed oxides based on plutonium recycled from discarded nuclear warheads and a range of fresh fuel compositions based on thorium. It is expected that for each type of advanced fuel, a different optimal CANDU bundle design will be obtained, but the research may well identify a near optimal design that may be suitable for groups or even most of these new fuels.

SAMPLE PUBLICATIONS:

  • H.W. Bonin, "Optimization of the Fuel Bundle Design for CANDU Nuclear Reactors Using Advanced Fuels Cycles," Proc. Topical Meeting on Advances in Fuel Management, Pinehurst, N.C., U.S.A., 2-6 March 1986, pp. 441-450.
  • H.W. Bonin, "Optimization of the Fuel Bundle Design for CANDU Nuclear Reactors: Phases I and II: Fuel Composition and Geometry Optimization Problems," Proc. 12th Annual Canadian Nuclear Society Symposium of Reactor Dynamics and Plant Control, Hamilton, Ontario, 1986.
  • H.W. Bonin, "Enrichment Distribution Optimization in CANDU Reactor Bundles for Advanced Fuel Cycles," Proc. International European Nuclear Society- ENC-4, Genève, Switzerland, June 1986, Vol. 4, p. 341.