Nuclear Science and Engineering / Volume 155 / Number 3 / March 2007 / Pages 378-385
Technical Paper / Mathematics and Computation, Supercomputing, Reactor Physics and Nuclear and Biological Applications / dx.doi.org/10.13182/NSE07-A2670
An optimization method has been developed to determine the optimal fresh fuel rod configurations, fresh streams, and fresh bundle design placements given a known exposed fuel loading pattern and operational strategy for boiling water reactors. The optimization method is based on a first-order approximation of various core parameters, such as hot excess reactivity and critical power ratio, using fuel rod perturbations to the reference fresh bundle designs. A simulated annealing optimization algorithm is shown to produce fresh bundle designs, consisting of rods selected from a user-defined set of rod types that optimize the core design with respect to its design constraints.
The method utilizes a linear superposition method based upon sensitivity coefficients to approximate core parameters. A parallel computing system was implemented to decrease wall clock time for the numerous lattice physics and core simulator calculations. A periodic update of the reference bundle design, without the computational burden of updating the sensitivity coefficients, was introduced and is shown to significantly improve the accuracy of the approximation model. Application of the method demonstrates that improved core designs are achieved when a many-fresh bundle design (i.e., stream) solution is considered as part of the design space. Six-stream (and higher) core designs that increase fuel utilization while simultaneously reducing manufacturing costs through reduction of fuel rod types fabricated, previously unattainable with existing methodologies, are now possible.