The Chinese Academy of Sciences developed intelligent fault diagnosis methods for complex systems in nuclear power plants

The research group of Wang Jianye, a researcher at the Institute of Nuclear Energy Safety Technology, Chinese Academy of Sciences, has made progress in the research of intelligent fault diagnosis methods for complex nuclear power plants.

Based on the data-driven method, the fault diagnosis model of complex systems in nuclear power plants is established, and a set of adaptive fault diagnosis method based on the combination of non-dominated genetic algorithm with elite reservation strategy and convolutional neural network algorithm is developed, which provides theoretical and method support for fault diagnosis of complex systems in nuclear power plants. The findings were published in Annals of Nuclear Energy.

According to the report, the researchers analyzed the characteristics of nuclear power plant data, established a data-driven fault diagnosis model for complex systems of nuclear power plants, and developed an adaptive fault diagnosis method based on the combination of non-dominated genetic algorithm with elite reservation strategy and convolutional neural network algorithm. The algorithm has been applied to the platform of China lead-based research reactor under the strategic leading science and technology project of Chinese Academy of Sciences.