Research on Fault Diagnosis of Nuclear Power Plants Based on Genetic Algorithms and Fuzzy Logic

Based on genetic algorithms and fuzzy logic and using expert knowledge, mini knowledge tree model and standard signals from simulator, a new fuzzy genetic method is developed to fault diagnosis in nuclear power plants. A new replacement method of genetic algorithms is adopted. Fuzzy logic is used to calculate the fitness of the strings in genetic algorithms. Experiments on the simulator show it can deal with the uncertainty and the fuzzy factor.