摘要
针对感应电机中多类型故障的检测问题,提出一种基于经验模态分解(EMD)和改进多重信号分类(MUSIC)算法的检测方法。通过采集器采集电机定子的稳态电流信号。利用EMD对稳态电流信号进行时域分解,获得包含丰富故障信息的两个本征模态函数(IMF)。利用MUSIC算法对IMF进行频域分析,获得各类故障的特征频率,同时引入了Prony算法来确定IMF频谱中的幅值。基于故障频域特征,对转子断条、轴承缺陷和转子质量不平衡这3种故障进行检测。实验结果表明,该方法不仅能够应用于单一故障检测,还能够在多故障并发时准确检测出所有故障。
For the issues that the multi-type fault detection of induction motor, a detection method based on empirical mode decomposition (EMD) and improved multiple signal classification (MUSIC) was proposed. The steady-state current signal of the motor stator was collected by a collector. The EMD was used to decompose the steady-state current signal in time domain, and two intrinsic mode functions (IMF) with rich fault information were obtained. The MUSIC algorithm was used to analyze the frequency domain of the IMF, and the characteristic frequencies of various faults were obtained, and the Prony algorithm was introduced to determine the amplitude of the IMF spectrum. The fault frequency domain, the broken rotor bar, bearing defects and rotor mass imbalance fault were detected based on the characteristics. Experimental results show that this method can not only be applied to single fault detection, but also can detect all faults accurately when muhi- ple faults are concurrency.
出处
《微特电机》
北大核心
2017年第7期37-40,53,共5页
Small & Special Electrical Machines
基金
河南省科技攻关重点项目(162102310442)
关键词
感应电机
故障检测
经验模态分解
多重信号分类
PRONY算法
induction motor
fault detection
empirical mode decomposition
muhiple signal classification
Prony al-gorithm