论文部分内容阅读
电机的振声信号包含了丰富的状态信息,利用电机的声音信号进行故障检测,并提出了综合小波包能量相对熵的概念。对采集的声音信号进行小波包分解,利用重构系数计算故障信号与正常信号小波能量包相对熵,并得到综合小波包能量相对熵,确定电机是否有故障;并通过比较各频带的综合小波包能量相对熵大小判断故障所处的频带位置,从而确定电机为何种故障。电机振声信号诊断实例验证了该方法的有效性。
The vibration signal of the motor contains rich state information, uses the motor sound signal to carry on the fault detection, and puts forward the concept of synthesizing the relative entropy of the wavelet packet energy. Wavelet packet decomposition of the collected sound signal is carried out, and the relative entropy of the fault signal and the normal signal wavelet energy packet is calculated by using the reconstruction coefficient, and the relative entropy of the energy of the wavelet packet is obtained to determine whether the motor has a fault. By comparing the comprehensive wavelet packet The relative entropy of energy determines the location of the frequency band in which the fault is located to determine what kind of fault the motor is. The vibration signal diagnosis example of the motor verifies the effectiveness of the method.