Presenter: Matthew A. Jessee
Advisor(s): Paul Turinsky
Author(s): Matthew A. Jessee
Graduate Program: Nuclear Engineering
Title: Fuel Rod Optimization for Boiling Water Reactors
Abstract: The performance and energy production of a boiling water nuclear reactor depend on many design parameters. Three essential design parameters include the fuel bundle loading pattern, the control rod program, and the fuel rod configuration in each fuel bundle (i.e. the bundle design). Over the past fifteen years, major advancements have been made in optimization codes that determine loading patterns and control rod programs that maximize reactor performance and minimize fuel manufacturing costs. These codes do not attempt to optimize the bundle design and typically only consider two to three different bundle designs. The objective of this work is to incorporate the bundle design into these existing optimization codes. The bundle design optimization problem is very unattractive to solve. It is an integer programming problem with over thirty decision variables, ten design constraints, and approximately 10^46 feasible solutions. In addition to the immense size of problem, the direct evaluation of each bundle design candidate requires three, computationally intensive, reactor analysis calculations. For this work, an approximation method was employed to rapidly estimate the worth of each bundle design candidate in terms of a formulated objective function. A simulated annealing algorithm was employed to search the bundle design space using the developed approximation model to determine the optimal solution. Application of the method demonstrates that improved core designs are achieved that include six to ten different bundle designs, previously unattainable with existing optimization software. The increase in the number of bundle designs allows for better fuel utilization equating to annual electric energy generation cost savings of $1M for the Progress Energy Brunswick Nuclear Plant, and a potential decrease in manufacturing costs of 2%-4% for the GNF fuel fabrication plant located in Wilmington, NC.