The company OSC (https://www.osc.cz/) develops the simulator for the Dukovany Nuclear Power Plant. Given recent advancements in artificial intelligence, OSC would like to analyze the application of machine learning methods to classify abnormal events in the nuclear power plant.

The primary goal of this thesis is to review existing work in the field, perform a wide range of experiments, and synthesize the findings that will serve as a solid foundation for future research. The experiments will involve training machine learning models found in the literature on simulator-generated data and evaluating their performance, limitations, and real-world applicability.

Finally, the thesis will assess the general feasibility of applying machine learning to abnormality classification and determine whether further research is justified.