您的位置:首页  > 论文页面

基于信息融合的数控机床机械故障诊断方法研究

发表时间:2017-01-23  浏览量:1517  下载量:342
全部作者: 谭继文,文妍
作者单位: 青岛理工大学机械工程学院
摘 要: 数控机床结构的复杂性及子系统间的耦合关系,增加了其故障诊断的难度。研究提出一种基于多层次信息融合的数控机床故障诊断方法。利用数控机床结构组成与控制方式的特点,建立基于外部传感器、机床内部参数组成的数控机床信息多维表征体系,全面、完整地监测数控机床运行状态;提取原始采集信号和经验模态分解处理信号的时域、频域及时频域多个特征参数,挖掘故障信息;基于核主元分析(kernel principal component analysis,KPCA)获取故障敏感特征参数集合;同时启用多个分类器,分别对故障敏感特征参数进行定量分析,实现对故障的分类;最后以模糊综合评判的方式对多个分类器的结果进行全局决策融合,得到最终的诊断结果。该方法应用在数控机床滚动轴承的故障诊断中,实验结果表明该方法提高了诊断的精度与可靠性,验证了方法的有效性与应用价值。
关 键 词: 机械制造自动化;多传感器;信息融合;数控机床;故障诊断
Title: A fault diagnosis method based on data fusion for numerical control machine
Author: TAN Jiwen, WEN Yan
Organization: College of Mechanical Engineering, Qingdao University of Technology
Abstract: Fault diagnosis for numerical control machine is very difficult due to the complexity of the structure and the coupling relationship between subsystems. In order to improve the accuracy and reliability of fault diagnosis for numerical control machine, an intelligent fault diagnosis method based on multi-level data fusion is developed in this paper. The multi-dimensional information system with the external sensors data and the internal parameters of numerical control machine is established to monitor the running state of numerical control machine. Multiple characteristic parameters in time domain, frequency domain and time-frequency domain are extracted from the processed signals to extract the fault information. The sensitive parameter set is obtained by kernel principal component analysis (KPCA) method. Multiple classifiers are enabled respectively and simultaneously to analyze the characteristic parameters quantitatively and diagnose the fault types. Finally, the results of multiple classifiers are fused in the form of global decision fusion by the method of fuzzy comprehensive evaluation. Then the final diagnosis results are obtained. The model has been tested in the bearings fault diagnosis of numerical machine and the results show that the proposed model is effective and versatile.
Key words: manufacturing automation; multi-sensor; data fusion; numerical control machine; fault diagnosis
发表期数: 2017年1月第2期
引用格式: 谭继文,文妍. 基于信息融合的数控机床机械故障诊断方法研究[J]. 中国科技论文在线精品论文,2017,10(2):123-130.
 
1 评论数 0
暂无评论
友情链接