基于小波包能量熵-BP神经网络的核电站阀门远传机构故障诊断研究

Research on Fault Diagnosis of Valve Remote Transmission Mechanism in Nuclear Power Plant Based onWavelet Packet Energy Entropy-BP Neural Network

核电站阀门远传机构长期运行在恶劣工况下,当出现故障时会使核电站中关键阀门无法正常开闭,严重时导致设备停机检修,从而造成经济损失。针对这一问题,本文提出一种小波包能量熵-BP神经网络的阀门远传机构故障诊断方法。阀门远传机构样机搭建LabWindows/CVI振动信号采集试验平台,使用小波包算法对故障信号进行分解与重构,并提取小波包能量熵来构造故障特征向量,输入到BP神经网络训练后,经MATLAB仿真运行,验证了该方法的可行性和有效性。

The valve remote transmission mechanism of nuclear power plant operates under harsh conditions for a long time.When a fault occurs,the key valves in the nuclear power plant cannot be opened and closed normally,which will lead to economic losses caused by equipment shutdown and maintenance.Aiming at this problem,this paper proposes a fault diagnosis method of valve remote transmission mechanism based on wavelet packet energy entropy-BP neural network.Based on the prototype of valve remote transmission mechanism,a LabWindows/CVI vibration signal acquisition test platform is built.The wavelet packet algorithm is used to decompose and reconstruct the fault signal,and the wavelet packet energy entropy is extracted to construct the fault feature vector.After input into the BP neural network training,the feasibility and effectiveness of the method are verified by MATLAB simulation.