转子振动故障的过程功率谱熵特征分析与定量诊断

针对传统旋转机械振动故障定性诊断的不足,提出了1种以信息熵理论为基础的转子故障特征分析与定量诊断方法。在转子试验台上模拟转子振动的4种典型故障,分别得到4个测点多转速下的振动过程故障数据;对这些故障数据进行分析和处理,提取反映其振动过程的故障特征——功率谱信息熵,建立能描述转子振动过程变化规律的多转速多测点下的故障信息熵矩阵,并对振动故障进行分析;通过对转子振动故障信号的实例计算和定量诊断分析,验证了该方法在转子振动故障分类和故障严重程度判断方面是可行的。

Feature Analysis and Quantitative Diagnosis of Process Power Spectrum Entropy for Rotor Vibration Failure

To solve the shortcoming of qualitative diagnosis of the traditional rotary machine vibration failure,the method of rotor failure feature analysis and quantitative diagnosis was proposed based on the information entropy theory.Four typical failures of the rotor vibration were simulated on the rotor experiment,and vibration failure data under more points and multi-speed was obtained.The failure feature and power spectrum entropies which reflects fault features of vibration process were extracted,and information entropy matrixes which could describes the change disciplinarian of rotor vibration under more measurement points and multi-speed were built by analyzing and processing the datum.The validity of this method in distinguishing the rotor fault types and determining the fault severity was proved to be effective by example calculation and quantitative diagnosis analysis of rotor vibration fault signals.