Reliability statistical inference method based on multivariate accelerated degradation data

Abstract

The invention relates to a reliability statistical inference method based on multivariate accelerated degradation data. The method mainly comprises the following steps: (1) using a stochastic process to respectively establish an accelerated degradation model for accelerated degradation data of each performance parameter; (2) using a maximum likelihood estimation method to obtain a parameter value of each accelerated degradation model; (3) respectively establishing a marginal survival function based on a stochastic process model for each performance parameter; (4) using a Copula function to establish a correlation model between the marginal survival functions; (5) using a Monte Carlo method to estimate a correlation model parameter value; (6) constructing a reliability function under a rated stress for inferring the reliability of a product. The reliability statistical inference method based on multivariate accelerated degradation data has a wider application scope and better engineering applicability.

Description

technical field [0001] The invention belongs to the field of reliability engineering and relates to a reliability statistical inference method based on multivariate accelerated degradation data. Background technique [0002] In today's society, product replacement is accelerating, how to determine the reliability index of the product in a short period of time is very important to enhance the competitiveness of the product. Accelerated degradation tests have been widely used in product reliability assessment because they can obtain enough performance degradation data in a shorter period of time. However, how to use the measured multivariate accelerated degradation data to evaluate the reliability of such products is currently a difficult problem . Most of the existing methods assume that there is correlation between the degradation quantities at the same measurement time, and can only use the same type of degradation model to model each degradation process, which limits the ...

AI Technical Summary

Problems solved by technology

Accelerated degradation test has been widely used in product reliability evaluation because it can obtain enough performance degradation data in a shorter time. However, how to use the measured multivariate accelerated degradation data to evaluate the reliability of such products is currently a difficult problem

Most of the existing methods assume that there is correlation between the degradation quantities at the same measurement time, and can only use the same type of degradation model to model each degradation process, which limits the application range of such methods

Images