Method of probability calculation of rolling bearing durability based on apparatus of nonparametric statistics

Abstract

The paper considers the solution of a problem of durability evaluation for rolling bearings with their loading mode in operation set by a specific law of radial load distribution by applying the methods of nonparametric statistics and computer simulation. It is shown that the actual laws of safety factor distribution differ significantly from normal distribution when determining the probability of bearing failure. For example, the probability of failure for the rolling bearing 2207 by dynamic load capacity at transition from the light to heavy loading mode with account of the actual law of safety factor distribution changes from 3.874% to 14.017%; at the same time in assumption of the normal law the probability of failure changes from 0.812% to 44.953%. The method of the bearing durability calculation proposed by the authors is universal and does not depend on the law of bearing loading. The method is illustrated by the example of the random load density function, which is bimodal. In the case when the loading law is assigned by the sampling obtained directly during the machine operation, the method is also implemented without changes.