基于自学习SOM和ARMA算法的数控机床滚动轴承健康预警研究  

Research on Health Warning for Rolling Bearing of CNC Machine Tool based on Self-learning SOM and ARMA Algorithm

作　　者：夏筱筠 林浒[2] XIA Xiao-jun;LIN Hu(University of Chinese Academy of Sciences,Beijing 100049,China;Shenyang Institute of Computing Technology,Chinese Academy of Sciences,Shenyang 110168,China)

机构地区：[1]中国科学院大学,北京100049 [2]中国科学院沈阳计算技术研究所,沈阳110168

出　　处：《小型微型计算机系统》2019年第1期215-220,共6页Journal of Chinese Computer Systems

基　　金：国家科技重大专项课题项目(2017ZX04011004)资助

摘　　要：随着我国智能制造技术的发展,预测性设备维护在工业环境中扮演着日益重要的角色.目前大多的健康预警算法针对性较强,自学习能力不足,导致算法的适用性、灵活性存在较大的局限性.为此,本文以滚动轴承为研究对象,基于状态检测的设计策略,完成了滚动轴承健康维护的实施方案;根据以上实施方案,提出并实现了自组织特征映射网络的故障诊断算法及自适应ARMA故障预警算法,提高了滚动轴承故障诊断与预警的智能化水平及对健康预警的处理效率.实验结果表明,所研究的算法对于实现可靠的滚动轴承故障诊断及预警具有良好的应用效果.With the development of intelligent manufacturing technology,predictive equipment maintenance plays an increasingly important role in the industrial environment. At present,most of the health warning algorithms have strong pertinence and lack of selflearning ability,which leads to the limitation of applicability and flexibility of the algorithm. Therefore,taking the fault diagnosis and early warning of rolling bearing as an example,the implementation scheme of the healthy maintenance of rolling bearing based on state detection has been designed. Combining with the characteristics of fault early warning and fault diagnosis of rolling bearings,the fault diagnosis method of self organizing feature mapping network and the adaptive ARMA fault warning algorithm have been proposed and realized. The intelligent level and the efficiency of health early warning has been improved. The experimental results showthat the algorithm is effective in achieving reliable fault diagnosis and early warning of rolling bearings.

关 键 词：状态维修 自组织特征映射网络 线性回归预测模型 故障诊断及预警 自学习能力 

分 类 号：TD324[矿业工程—矿井建设]