摘要
采煤机截割部传动系统故障源多,建立快速准确识别故障源的模型具有重要研究意义,因此提出了一种基于粗糙集-径向基函数(RBF)神经网络的故障诊断方法。首先对传动系统常见故障进行汇总分析,归纳为齿轮故障与轴承故障,通过粗糙集理论完成属性约简后得到最小条件属性集,然后根据粗糙集的最小条件属性集搭建RBF神经网络的拓扑结构。仿真结果表明,基于粗糙集-RBF神经网络的故障诊断模型结构更简单,训练效率及诊断准确性更高,在故障诊断中具有更好的应用效果。
There are many fault sources in the transmission system of the shearer cutting unit, and it is of great research significance to establish a model to quickly and accurately identify the fault source.Therefore, a fault diagnosis method based on rough set-radial basis function(RBF) neural network was proposed. Firstly, the common faults of the transmission system were summarized and analyzed, and they were summarized as gear faults and bearing faults. Then attribute reduction was completed by rough set theory, the minimum condition attribute set was obtained, and then the RBF neural network topology structure was constructed according to the minimum condition attribute set of rough set. The simulation results show that thestructure of the fault diagnosis model based on rough set-RBF neural network is simpler, the training efficiency is higher and the diagnosis accuracy is higher, so it is better applied in fault diagnosis.
作者
王海花
林邓伟
霍晓丽
Wang Haihua;Lin Dengwei;Huo Xiaoli(Jiaozuo University,Jiaozuo 454000,China)
出处
《煤矿机械》
2021年第5期175-177,共3页
Coal Mine Machinery
