Rolling bearings fault assessment based on quantitative correlation model

TAN Xiao-dong1,3, ZHANG Yong2
(1.School of Mechanical and Electrical Engineering, University of Electronic Science and Technology of China,
Chengdu 611731, China; 2.Science and Technology on Integrated Logistics Support Laboratory,
National University of Defense Technology, Changsha 410000, China; 3.Department of Force
Management, Officers College of PAP, Chengdu 610213, China)

Abstract: Aiming at improving condition monitoring for rolling bearings fault evolution process, the relationships among fault signal resources, condition indices and fault severity which impact the accuracy of fault severity assessment for rolling bearing were studied, and a fault severity assessment method based on quantitative correlation model optimization was proposed. Original signals of rolling bearings in various fault severity were collected by available tests deployed on equipment, and condition indices of original signals for each fault severity were calculated. Thus those complex relationships among multiple test points, condition indices and fault severity were built by quantitative correlation models, and the fault severity assessment results with various tests and condition indices were studied. Finally, the minimum RMSE of assessment outcomes was used respectively to select optimal assessment models for rolling ball, inner race and outer fault. The results indicate that the proposed method can capture more original fault information, and accurately describe the relationships among fault signal resources, condition indices and assessment models, and the assessment models using multiple regression model optimization can increase the accuracy of quantitative assessment.
Key words: rolling bearings; fault assessment; quantitative correlation model; condition indices; root mean square error（RMSE）