摘要
重型刮板输送机链传动系统故障中影响最大的就是断链事故,不仅影响煤矿企业生产计划,同时还可能造成安全事故。断链故障主要是由于链环所受应力突增导致材料出现塑性变形和开裂,因此从链环和与其啮合的链轮轮齿接触部位的应力分布、应变等动态特性出发,通过ANSYS分析并获取其应力变化特征,并以此作为样本,通过BP神经网络构建和训练预测模型,实现断链故障的预警及诊断。仿真结果表明,该方法对链轮、链条的应力变化跟踪较好,精度较高,可以满足断链故障的诊断及预警要求。
Chain breaking accident is the most important fault in chain drive system of heavy scraper conveyor, which not only affects the production plan of coal mining enterprises, but also may cause safety accidents. The chain breaking fault is mainly caused by the plastic deformation and cracking of the material caused by the sudden increase of the stress on the chain link. Therefore started from the dynamic characteristics of the stress distribution and strain at the contact part of the chain link and its meshing sprocket teeth, analyzed and obtained the stress variation characteristics through ANSYS, and took this as a sample, constructed and trained the prediction model through BP neural network, so as to realize the chain breaking fault early warning and diagnosis. The simulation results show that the method can track the stress change of sprocket and chain well with high accuracy, which can meet the requirements of diagnosis and early warning of broken chain fault.
作者
马艳芳
刘雪贞
邓小飞
Ma Yanfang;Liu Xuezhen;Deng Xiaofei(Jiaozuo University,Jiaozuo 454000,China)
出处
《煤矿机械》
2021年第4期181-183,共3页
Coal Mine Machinery
基金
河南省重点研发与推广专项(科技攻关)项目(202102310204)。
关键词
刮板输送机
链传动系统
断链故障
神经网络
scraper conveyor
chain drive system
broken chain fault
neural network
