针对目前滚动轴承可靠性评估中性能指标的概率模型大多为单性能指标的静态概率模型,且与实际概率模型误差较大的问题,提出了基于按类统计的滚动轴承可视化可靠性动态评估技术.利用轴承可靠性下降时,其空间状态发生跃迁、概率模型类别会逐渐增多的特点,以滚动轴承的均方根和峭度这两个性能指标为分析对象,采用核密度法建立起初始按类概率模型.对初始按类概率模型进行可视化处理得到初始按类概率图像模型,计算非正常类的图像分布区域面积占总图像分布区域面积的比值得到故障率,进而得到可靠性指标.当数据样本不断累积更新,可实现滚动轴承可靠性的动态评估.对美国智能维护系统中心提供的滚动轴承寿命实验数据进行分析,结果表明,提出的动态按类概率模型能够及时跟踪滚动轴承的退化过程,可靠性指标能够实时反映滚动轴承的退化程度,具有较强的工程实用价值.%Most of the probability models of performance indexes are static probability models of single performance index in the study of the reliability evaluation of rolling bearings,which have large errors compared with the actual probability models.Therefore,a visualized dynamic assessment technology of rolling bearing reliability based on class statistics is proposed.When the bearing reliability declines,there may be a transition in its spatial state,and the probability model category will increases gradually.Selecting the root mean square and kurtosis value of rolling bearings as the analysis object,and the initial class probability model is established by using the nuclear density method.Then the initial class probability model is visualized to obtain the initial class probability image model.The failure rate is the ratio of the abnormal area to the total image area,and the reliability index is obtained.As the data are continuously updated,a dynamic reliability evaluation of the rolling bearings can be achieved.The rolling bearing life test data provided by the American intelligent maintenance system center are analyzed,and the results show that the new dynamic probability model can track the degradation process of rolling bearings in time,so as to reflect the degradation degree of the rolling bearings.
展开▼
