基于PSO-BP神经网络的采煤机电动机故障诊断研究  

Research on fault diagnosis of shearer motor based on PSO-BP neural network

作　　者：姜磊[1] 叶圣超 李飞龙 JIANG Lei;YE Shengchao;LI Feilong(Zhejiang Industry Polytechnic College,Shaoxing 312000,Zhejiang,China;Beijing Information Science&Technology University,Beijing 100101,China;China Mining Longke Energy Technology Co.,Ltd,Beijing 101300,China)

机构地区：[1]浙江工业职业技术学院,浙江绍兴312000 [2]北京信息科技大学,北京100101 [3]中矿龙科能源科技股份有限公司,北京101300

出　　处：《矿山机械》2020年第9期59-64,共6页Mining & Processing Equipment

基　　金：2020年浙江省自然科学基金项目(LGG20F010002);北京市科技提升计划项目(PXM2016_014224_000021)。

摘　　要：针对因采煤机电动机超长时间运行与矿井极端工作环境而引起的故障问题,结合异步电动机数学模型及其常见故障机理分析,在分析 BP 算法存在缺陷的基础上,提出一种用于电动机故障诊断的 PSO-BP 神经网络算法,以实现对采煤机运行状态的实时监测。将 PSO 算法与 BP 算法相结合,共同优化神经网络连接权值,用电动机故障训练样本对 PSO-BP 神经网络进行训练并进行网络测试。结果表明,与 BP 神经网络相比,PSO-BP 神经网络能更快速、准确诊断电动机的健康状态,及时采用有效措施可降低电动机故障率,从而保障矿井人员作业安全,提高生产效率。Aiming at the faults caused by the ultra-long running duration of the shearer motor and the extreme working environment of mine,in combination with the mathematical model of the asynchronous motor and its common fault mechanism analysis,based on the analysis of the defects of BP algorithm,a PSO-BP neural network algorithm for motor fault diagnosis was proposed to realize the real-time monitoring for the operation status of the shearer.In the paper,PSO algorithm and BP algorithm were combined to optimize the connection weights of the neural network,and the training samples of the motor fault were applied to train the PSO-BP neural network.And then,the network was tested.The results showed:compared with the BP neural network,the PSO-BP neural network diagnosed the health status of the motor more quickly and accurately.Timely application of effective measures reduced the fault ratio of the motor,so as to ensure the safety of the mine staff and improve the productivity.

关 键 词：采煤机电动机 PSO-BP神经网络 故障诊断 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置] TD421[自动化与计算机技术—控制科学与工程]