基于小波包和EMD的断路器机械故障诊断研究

以ZN63A-12型高压真空断路器为研究对象,针对处理高压断路器振动信号时单独使用小波包特征熵或经验模态分解(EMD)特征熵作为特征向量进行诊断正确率低的缺点,将高压断路器振动信号的小波包能量熵、经验模态分解能量熵、经验模态分解能量相结合作为特征向量,采用马氏距离判别法进行模式识别,实现对断路器两种机械故障模式的判别.实验结果表明,该方法准确率达97.40％,具有较高的实用价值.

Research on Mechanical Fault Diagnosis of Circuit Breaker Based on Wavelet Packet and EMD

Mechanical fault diagnosis based on vibration signal of ZN63A-12 high voltage vacuum circuit breaker is researched.In order to overcome the shortcoming of low accuracy of using wavelet packet characteristic entropy or empirical mode decomposition (EMD) entropy as a feature vector alone,a novel feature vector which combined with the wavelet packet energy entropy,empirical mode decomposition energy entropy,and EMD energy is proposed.With the advantage of this feature,the Mahalanobis distance discrimination method is used for pattern recognition to realize the classifications of two kinds of mechanical fault modes.The experimental results show that the accuracy of this method can reach 97.40%,which is feasible enough for practical application.