基于振动信号时域特性的传动系统故障分类与诊断  

Fault classification and diagnosis of transmission system based on time domain characteristics of vibration signal

作　　者：张秋昕 吕渊 简红英 袁俊杰[2] 张西良[1] ZHANG Qiuxin;LV Yuan;JIAN Hongying;YUAN Junjie;ZHANG Xiliang(School of Mechanical Engineering,Jiangsu University,Zhenjiang 212013,China;College of Agricultural Engineering,Jiangsu University,Zhenjiang 212013,China)

机构地区：[1]江苏大学机械工程学院,江苏镇江212013 [2]江苏大学农业工程学院,江苏镇江212013

出　　处：《机电工程》2024年第11期1995-2002,共8页Journal of Mechanical & Electrical Engineering

基　　金：江苏仅一联合智造有限公司委托项目(20200495)。

摘　　要：包装机长期工作在往复变速运动及高粉尘、高水汽浓度环境下,其传动系统中的滚动轴承极易出现腐蚀故障和疲劳故障。针对这一问题,基于振动信号的时域特性,对包装机传动系统的滚动轴承进行了故障诊断研究。首先,以深沟球轴承6002为研究对象,对不同封装速度下的振动信号进行了采集,结合包装机传动系统的构造与工作机理,研究了其常见故障特点及类型,并从振动信号中选取了受外部干扰最小的数据段;然后,对选取的数据段进行了幅值变化和功率谱分析,验证了其具有相同的运动特性或运动过程,从而进一步得到了任意数据段之间在同一时间节点上的转速都一致,此时这些信号之间处于“相对平稳运动状态”;最后,利用偏度、峭度、峰值等时域指标对数据段进行了分析,得到了故障特征区间,将其用于故障诊断。研究结果表明:该状态下数据段之间处于“相对稳态运动”,解决了传统变速运动时域信息不可用的问题,提升了时域指标的诊断效率与稳定性,特别是峰值指标诊断准确率达到80%以上。该研究方法满足了工业生产中实时在线诊断的要求,同时也可为其他旋转机械的故障诊断研究提供参考价值。The packaging machine works for a long time in reciprocating variable speed movement and high dust and water vapor concentration environment,and the rolling bearings in its transmission system are prone to corrosion failure and fatigue failure.Aiming at this problem,based on the time domain characteristics of vibration signal,the fault diagnosis of rolling bearing in the transmission system of packaging machine was studied.Firstly,taking the deep groove ball bearing 6002 as the research object,the vibration signals under different packaging speeds were collected.Combining with the structure and working mechanism of the packaging machine transmission system,the characteristics and types of common faults were studied,and the data segment with the least external interference was selected from the vibration signal.Then,the amplitude change and power spectrum analysis of the selected data segments were carried out to verify the same motion characteristics or motion process,so as to further obtain the rotational speeds of any data segments at the same timline were consistent,and these signals were in arelatively stable motion state.Finally,the time domain indexes such as skewness,kurtosis and peak were used to analyze the data segment,and the fault feature interval was obtained for fault diagnosis.The research results show that the data segments are in arelative steady-state motionin this state,which solves the problem that the time-domain information of traditional variable motion is not available,and improves the diagnostic efficiency and stability of the time-domain index.In particular,the diagnostic accuracy of the peak index reaches more than 80%.The research method meets the requirements of real-time online diagnosis in industrial production,and also provides reference value for fault diagnosis of other rotating machinery.

关 键 词：包装机传动系统 滚动轴承 故障诊断 腐蚀故障 疲劳故障 振动信号动态特性 时域指标 

分 类 号：TH133.3[机械工程—机械制造及自动化] TB486[一般工业技术—包装工程]