摘要
电机低频运行时发生故障的特征频率与电源频率之间的差值较小,采用传统的3层小波包分解法不能满足频率细化的要求。对电机在正常和故障两种运行状态下的定子电流信号进行8层小波包分解,细化了电机电流低频段序列,提取出能够较为准确反映电机故障的特征向量。将其特征能量进行了对比分析,从正常和故障信号的特征能量对比结果证明了利用8层小波包分解和重构的方法能够更准确地判断电机故障。
The difference between failure occurs when the motor is running low frequency characteristic frequency and power frequency between the smaller, traditional three-layer wavelet packet analysis method can not meet the requirements of the frequency of refinement. The motor under normal operating status and fault two stator current signal 8 layer wavelet packet decomposition, refined motor current low frequency sequence extracted motor failure can be more accurately reflects the characteristics of the vector, and were characterized by energy comparative analysis, comparing the energy characteristics of normal and fault signals from the results demonstrate the use of wavelet packet decomposition and reconstruction of eight layers can more accurately determine motor failure.
出处
《电机与控制应用》
北大核心
2015年第4期32-36,共5页
Electric machines & control application
基金
国家高技术研究发展计划(863计划)课题项目(2013AA06A409)
中国煤炭科工集团有限公司科技创新资助项目(2012MS018)
关键词
小波包分解
低频运行
特征频率
低频段序列
wavelet packet decomposition
low frequency operation
characteristic frequency
low frequency sequence
