针对振动信号的高压断路器机械故障诊断-20111011.pdfVIP

针对振动信号的高压断路器机械故障智能诊断方法 A Method of Mechanical Fault intelligent Diagnosis for High Voltage Circuit Breakers Based on Vibration Signals 沈阳工业大学 辽宁省电网安全运行与检测重点实验室 刘洋，林莘，徐建源，李斌，杨壮壮 （Shenyang University of Technology, Liaoning Province Key Lab of Power Grid Safe Operation and Monitoring ,110870, Shenyang） Liu Yang, Lin Xin , Xu Jianyuan, Li Bin, Yang Zhuangzhuang [摘要]以高压断路器振动信号为基础，提出一种小波包能量熵与支持向量机相结合的高压断路器智能 诊断方法。该方法根据自建的检测平台通过加速度传感器测得高压断路器不同状态下的振动数据，利 用小波包分解技术分析断路器故障时的振动信号，对振动信号进行一维小波包三层分解，以各频段的 小波包能量熵作为断路器故障模式的特征向量，采用支持向量机理论建立智能诊断网络。最后，通过 对断路器的脱扣器卡涩，真空灭弧室固定螺栓松动，传动机构摩擦力过大三种故障模式进行识别分析， 验证了该方法相对于传统的神经网络诊断方法具有更优的分类能力和更高的准确性。 [Abstract] Based on high voltage circuit breaker’s vibration signal, take the high voltage circuit breaker’s vibration signal as the foundation, a new method of intelligent fault diagnosis based on wavelet packet energy entropy and support vector machine is presented. This method association with the examination platform applies the acceleration sensors to collect vibration data of high voltage circuit breaker’s different state. Take the wavelet packet analyzing the vibration signal, and do three layers decomposition of vibration signal, take the each frequency band’s wavelet packet energy entropy as characteristic vector of circuit breaker failure mode, and by using support vector machine theory establishment intelligent diagnosis network. Finally, distinguishing the circuit breaker tripping device card acerbity, vacuum arcing chamber fixed bolt looseness, transmission mechanism too much friction three failure mode, verify that this method compared with the traditional neural network diagnosis method has better capability of classified and higher accuracy. [关键词]支持向量机；能量熵；高压断路器；智能诊断 [Key Words] support vector machine；energy entropy ；high voltage circuit breaker ；Intelligent diagnosis 0 引言