PSO-BP神经网络采煤机截割系统故障诊断  

PSO-BP Neural Network Shearer Cutting System Fault Diagnosis

作　　者：贾凯[1] 李浩 熊晓航[1] 

机构地区：[1]辽宁工业大学机械工程与自动化学院,辽宁锦州121001 [2]辽阳石化公司芳烃厂仪表车间,辽宁辽阳111000

出　　处：《辽宁工业大学学报（自然科学版）》2010年第3期173-175,共3页Journal of Liaoning University of Technology(Natural Science Edition)

摘　　要：截割系统是采煤机重要的组成部分,如果发生故障将导致连续采煤机停止工作,严重影响其生产率。采用粒子群算法(PSO)优化BP神经网络的方法,对采煤机截割系统故障诊断进行研究,与传统BP网络比较后,结果表明这种方法可确定连续采煤机截割系统的运行状态、降低故障率、提高生产率和使用寿命。Cutting system is an important part of shearer,in the event of failure which will cause the miner to stop working constantly,seriously affecting its productivity.The use of particle swarm optimization（PSO） optimized the BP neural network,then the research was made on the cutting system fault diagnosis with the optimized BP neural network.Compared with the traditional BP network,the results ineicate that this method can determine the continuous miner cutting systems state,reduce the failure rate,improve the productivity and highlight the service life.

关 键 词：采煤机 神经网络 故障诊断 PSO-BP 

分 类 号：TD421.6[矿业工程—矿山机电]