引用本文: | 杨存祥,朱 琛,解豪杰.基于RPROP神经网络算法的异步电动机故障诊断[J].电力自动化设备,2012,32(1): |
| YANG Cunxiang,ZHU Chen,XIE Haojie.Fault diagnosis based on RPROP neural network for asynchronous motor[J].Electric Power Automation Equipment,2012,32(1): |
|
摘要: |
为了更快、更精确地识别异步电动机的各种故障类型,克服BP神经网络存在的局部极小值和收敛速度慢等问题,提出了一种改进的BP算法——RPROP(Resilient PROPagation)神经网络算法。在介绍RPROP算法的基础上,建立了基于RPROP算法的异步电动机故障诊断模型,对异步电动机的定子匝间短路、转子断条、转子偏心和轴承4种故障进行识别和诊断。实验结果表明,该方法对异步电动机的故障诊断是有效的。 |
关键词: 异步电动机 故障诊断 电气故障 机械故障 BP神经网络 RPROP 神经网络 |
DOI: |
分类号: |
基金项目:河南省杰出青年科学基金资助项目(0741005100 04) |
|
Fault diagnosis based on RPROP neural network for asynchronous motor |
YANG Cunxiang, ZHU Chen, XIE Haojie
|
Key Laboratory of Information Appliance of Henan Province,Institute of Electrical and Information Engineering,Zhengzhou University of Light Industry,Zhengzhou 450002,China
|
Abstract: |
An improved BP algorithm,RPROP(Resilient PROPagation) neural network algorithm,is introduced to identify various fault types of asynchronous motor more quickly and accurately. The fault diagnosis model of asynchronous motor is established based on RPROP algorithm and four fault types of asynchronous motor are identified and diagnosed:stator winding inter-turn short circuit,rotor bar break,rotor eccentricity and rolling bearing fault . The experimental results show its effectiveness. |
Key words: induction motors fault analysis electrical fault mechanical fault BP neural network RPROP neural networks |