Qualitative fusion technique based on information poor system and its application to factor analysis for vibration of rolling bearings

Abstract

Poor information means incomplete and insufficient information for the characteristic presented in the subject investigated. The qualitative fusion technique is proposed by this paper to solve some problems of statistics under the condition of poor information. Based on the grey system theory which allows the probability distribution unknown and the data number small, the concept, qualitative fusion, is defined and its calculating method is described via factor analysis of the influence on the performances of rolling bearings. Experiments on factor analysis for vibration of the tapered roller bearing show that the main factors of the influence on vibration of the tapered roller bearing are linearity of the outer ring raceway, squareness of the bore surface, waviness of the outer ring raceway, roundness of the inner ring raceway, and roughness of the inner ring raceway. Hereby vibration of the tapered roller bearing can be improved by means of controlling over these 5 quality indexes of the rolling bearing rings.