摘要
滚动轴承在现代机械装备中得到了广泛应用。轴承的早期故障监测对于降低维护检修成本,提高机械系统的效率和可靠性具有重要意义。为此,提出了一种基于统计特征加权融合和图建模的滚动轴承早期故障监测新方法。对采集的轴承原始振动信号提取多个统计特征;采用自适应加权的方法对统计特征进行融合,对融合后的特征进行图建模,以提高其鲁棒性;然后进行图模型距离度量得到异常得分;最后采用检验假设对轴承早期故障进行监测。将此方法应用于滚动轴承数据集,实验结果证明了该方法的有效性,同时表明该方法在滚动轴承实时在线监测应用中具有良好的潜力。
Rolling bearings are widely used in modern mechanical equipment.The early fault monitoring of bearing is of great significance to reduce the maintenance cost and improve the efficiency and reliability of mechanical system.Therefore,a new method based on statistical feature weighted fusion and graph modeling for early fault detection of rolling bearing was proposed.Several statistical features were extracted from the original vibration signal of the bearing;the adaptive weighting method was used to fuse the statistical features,and the fused features were modeled by graph to improve their robustness;then the graph model distance measurement was used to get the abnormal score;test hypothesis was used to monitor the bearing early fault.This method is applied to the rolling bearing data set,and the experimental results show that the method is effective,and it has good potential in the application of rolling bearing real-time online monitoring.
作者
李苏
叶新来
卢国梁
Li Su;Ye Xinlai;Lu Guoliang(AtomHorizon Electric,Jinan 250061,China;School of Mechanical Engineering,Shandong University,Jinan 250061,China;Key Laboratory of High-efficiency and Clean Mechanical Manufacture(Shandong University),Ministry of Education,Jinan 250061,China)
出处
《机电工程技术》
2021年第5期227-231,共5页
Mechanical & Electrical Engineering Technology
关键词
滚动轴承
故障监测
特征融合
图模型
rolling bearing
fault detection
feature fusion
graph model
