Score-based Bayesian network structure learning algorithms for modeling radioisotope levels in nuclear power plant reactors

Radioactive corrosion products released into the primary coolant loop dominate final shutdown radiation fields of pressurized water reactors. Thus, reducing concentration these is a paramount duty in optimization process reactor performance. However, complexity and uncertainty present this make it difficult to predict their evolution theoretical way. We propose application structural learning Bayesian networks discover complex relations between most relevant variables loop, giving rise probabilistic models that obtain accurate reliable predictions products. Our analysis 5 power plants demonstrates our approach results simpler more models. Additionally, we conclude learned structures may represent an interpretable tool for plant technicians since they reveal useful information can be directly employed improve operation.