Analysis of Reliability Correlation Degree of Rolling Bearings Based on Zero-Failure Data

Modern equipment has higher requirements for the reliability of rolling bearings. The time and economic cost of obtaining bearing failure data through test methods are getting higher. Usually, truncation time tests of small sample are used to obtain zero-failure data of bearings. Based on the zero-failure data model and multi-layer Bayesian theory, this paper improves the reliability evaluation method of rolling bearings by changing the values of hyperparameters, and calculates the estimated value of failure probability at each truncation time to obtain the reliability of the bearing. This paper adopts the theory of grey relational degree to analyze the relationship and change law of bearing reliability at each truncation time, to understand the reliability change trend of rolling bearing more comprehensively. Experiments show that the method is reasonable.