Application Research on Intelligent Fault-Diagnosis of Nuclear Power Plant Equipment Based on Support Vector Machine

Abstract

Large rotating machinery is very extensive and important equipment in nuclear power plant. It is necessary to carry out on-line vibration monitoring and effective fault diagnosis, which are also the basic application characteristics of intelligent nuclear power plant system. Support vector machine (SVM) can be applied to the fault diagnosis of large rotating machinery in nuclear power plant because it can achieve better classification effect with less training samples and no prior knowledge of fault classification. Therefore, firstly, an online vibration monitoring system for large rotating machinery in nuclear power plant is constructed to extract fault features. Then, according to the simulation data of vibration test-bed and the fault data of actual equipment operation, SVM algorithm is used to simulate. By comparing with the actual fault, the diagnosis result is better, which verifies the effectiveness of the method, and is intelligent for the software platform of vibration monitoring device Diagnostic function development provides support.