A hybrid feature selection and health indicator construction scheme for delay-time-based degradation modelling of rolling element bearings

2017 2nd International Conference on Mechatronics and Electrical Systems

机械制造;无线电电子学

Zhang, Bin^1 ; Deng, Congying^1 ; Zhang, Yi^1

School of Advanced Manufacturing Engineering, Chongqing University of Posts and Telecommunications, Chongqing

400065, China^1

关键词: Condition monitoring sensors;    Construction scheme;    Hybrid feature selections;    Maintenance efficiency;    Mechanical components;    Operational reliability;    Rolling Element Bearing;    Rotatory machines;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/339/1/012026/pdf
DOI  :  10.1088/1757-899X/339/1/012026

来源: IOP

【 摘 要 】

Rolling element bearings are mechanical components used frequently in most rotating machinery and they are also vulnerable links representing the main source of failures in such systems. Thus, health condition monitoring and fault diagnosis of rolling element bearings have long been studied to improve operational reliability and maintenance efficiency of rotatory machines. Over the past decade, prognosis that enables forewarning of failure and estimation of residual life attracted increasing attention. To accurately and efficiently predict failure of the rolling element bearing, the degradation requires to be well represented and modelled. For this purpose, degradation of the rolling element bearing is analysed with the delay-time-based model in this paper. Also, a hybrid feature selection and health indicator construction scheme is proposed for extraction of the bearing health relevant information from condition monitoring sensor data. Effectiveness of the presented approach is validated through case studies on rolling element bearing run-to-failure experiments.