Design of Boiling Water Reactor Equilibrium Cycles Using Genetic Algorithms with PARCS and Neural Network Surrogates
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
We have developed an application to autonomously design boiling water reactor equilibrium fuel cycles. The application designs the loading pattern, control rod schedule, and core flow schedule for multiple cycles using a genetic algorithm. The fitness function is evaluated using neural network surrogate models for speed with periodic adjustments from PARCS. The program can design an equilibrium cycle from scratch using only a set of fresh fuel designs or can transition a depleted core to an equilibrium solution populated by the given fresh fuels. The application is applied to Edwin I. Hatch Nuclear Power Plant Reactor 1 Cycle 3 to transition the core to GE8X8 fuels through an equilibrium cycle design.
Fields of Study & Subjects
Nuclear reactor physics and engineering
Nuclear Engineering Thermal-Hydraulics
Nuclear and radioactivity studies
Publication details
- Journal / source
- Nuclear Technology
- Pages
- 1–19
- Publication date
- 26 March 2026
- Language
- EN