基于Beta分布自学习预警控制的轴承故障诊断智能监测  

Intelligent Monitoring of Bearing Fault Diagnosis Based on Beta Distribution Self-Learning Early Warning Control

作　　者：王飞飞[1] 张明[2] 王强 WANG Feifei;ZHANG Ming;WANG Qiang(Xinxiang Vocational and Technical College,Intelligent Manufacturing College,Xinxiang Henan 448000,China;College of Mechanical Engineering,Henan Polytechnic University,Zhengzhou 450000,China;Luoyang Huaguan Gear Co.,LTD.,Luoyang henan 450000,China)

机构地区：[1]新乡职业技术学院智能制造学院,河南新乡448000 [2]河南理工大学机械工程学院,郑州450000 [3]洛阳市华冠齿轮股份有限公司,河南洛阳450000

出　　处：《机械设计与研究》2022年第3期104-108,共5页Machine Design And Research

基　　金：国家自然科学基金资助项目(51905496)。

摘　　要：滚动轴承在初始运行阶段形成的故障信号强度较弱,较易与外部噪声信号形成互混,从而导致识别难度的大幅增加。为了提高机械传动部件的前期故障诊断效率,本文根据振动信号监测结果,综合运行小波包分解、动态核主成分分析等方法,建立了初期故障检测模型,由此实现通过数据驱动的方式完成初期设备故障的测试报警功能。研究结果表明:与采用峭度以及有效值方法对665报警点对照,以本方法进行故障检测可以缩短近30 h,满足实际要求。应用实例证明采用本设计的初期故障测试预警模型能够更快速、精确探测出轴承产生的初期缺陷。与传统固定阈值报警线方式比较,可以缩短近17天的故障检测预警时间。The intensity of fault signals from rolling bearings in the initial operation stage is relatively weak,and it is difficult to distinguish from other external noise signals,which leads to difficulty in identification.In order to improve the efficiency of early fault diagnosis of mechanical transmission parts,an early fault detection model is established based on the monitoring results of vibration signals,integrated operation wavelet packet decomposition and dynamic kernel principal component analysis,so as to detect early equipment failure using a data-driven approach.The results show that compared with the neurosis and effective value method for 665 alarm points,the proposed method can shorten nearly 30 hours for fault detection and meet the actual requirements.The application example proves that the incipient defect of the bearing can be detected more quickly and accurately using the designed incipient fault testing and warning model.Compared with the traditional fixed threshold alarm line method,it can shorten the fault detection and early warning time by nearly 17 days.

关 键 词：小波包分解 自学习预警 故障检测 预测 

分 类 号：TH17[机械工程—机械制造及自动化]