Bearing fault detection using relative entropy of wavelet components and artificial neural networks
Bearing fault detection using relative entropy of wavelet components and artificial neural networks
复制标题
DOI：
10.1109/demped.2013.6645767
复制
发表时间：
2013-10
期刊：
影响因子：
--
通讯作者：
H. L. Schmitt;L. R. B. Silva;P. Scalassara;A. Goedtel
中科院分区：
文献类型：
--
作者：
H. L. Schmitt;L. R. B. Silva;P. Scalassara;A. Goedtel
电机的故障检测一直是研究人员广泛探索的领域，特别是轴承故障，约占总故障的40%至60%。由于这类故障可通过定子电流的特定频率检测到，因此现在已成为研究的一个来源。因此，这项工作提出了一种预测性分析方法的基础上，相对熵的措施估计从小波包分解组件获得的重构信号。使用添加了与轴承故障相关的频率分量的真实的电机电流信号来模拟信号。使用三个人工神经网络拓扑结构，这些熵的措施分为两组：正常和故障信号的高性能率。
Fault detection in electrical machines have been widely explored by researchers, especially bearing faults that represents about 40% to 60% of the total faults. Since this kind of fault is detectable by particular frequencies at the stator current, it is now a source of investigation. Thus, this work presents a predicability analysis method based on relative entropy measures estimated over reconstructed signals obtained from wavelet-packet decomposition components. The signals were simulated using a real motor current signal with addition of frequency components related to the bearing faults. Using three ANN topologies, these entropy measures are classified in two groups: normal and faulty signals with a high performance rate.