Research on Typical Fault Prediction Model of the Reactor Coolant Pump Based on Multi-Information Fusion of GRU

Abstract According to the sensor type, measuring point distribution and acquisition frequency of the Reactor Coolant Pump (RCP) in the nuclear power plant, the typical failure mode of the key vulnerable parts of RCP is investigated and sorted out. Then, the simulation test bench of bearing and shaft seal components is set up respectively to simulate the operation of bearing and shaft seal components in several typical faults, and the multi-channel fault sample data is collected. In order to solve the problem of poor prediction accuracy and weak generalization ability of traditional neural network. Based on the above experimental data, a multi-information fusion prediction model of key signal of main pump based on gated circulation unit GRU is proposed in this paper. After testing the test set data, compared with the traditional deep learning algorithm, it is shown that the prediction model designed in this paper has the best prediction effect.