摘要
滚动轴承失效是最常见的传动系统故障,滚动轴承一旦发生故障会造成极大的经济损失和安全风险。因此,在故障发生的早期进行故障诊断可以有效地避免损失,具有很大的实际意义。应用BP神经网络的相关理论,以滚动轴承工作过程中产生的摩擦因数、振动值和噪声值作为输入变量,通过训练获得的BP神经网络求得轴承的不同工作状态,从而实现轴承的故障诊断。
Rolling bearing failure is the most common transmission system fault,once the failure of rolling bearing occurs,it will cause great economic and security risks.Therefore,fault diagnosis in the early stage of rolling bearing failure can effectively avoid losses,which has great practical significance. The theory of BP neural network is mainly applied.The friction coefficient,vibration value and noise value produced in the working process of rolling bearings were taken as input variables,and the corresponding states of bearings were obtained by training the BP neural network,so as to realize the fault diagnosis of bearings.
作者
曹智军
Cao Zhijun(Henan Polytechnic,Zhengzhou 450007,China)
出处
《煤矿机械》
北大核心
2019年第1期146-148,共3页
Coal Mine Machinery
关键词
BP神经网络
滚动轴承
故障诊断
BP neural network
rolling bearing
fault diagnosis
