Work

Reliability Assessment Method Based on Small Sample Accelerated Life Test Data

Year: 2024

Type: article

Abstract: Aiming at the problems of small sample size, few test failure data and low reliability evaluation efficiency in accelerated life test of high-reliability and long-life products, an improved virtual augmentation Bootstrap-Bayes reliability evaluation method based on small sample accelerated life test data was proposed. Firstly, the test data under various stress conditions are augmented by virtual fusion. Secondly, the empirical distribution function and bootstrap sampling method are improved, and the kernel density estimation method is used to fit the density distribution of test data as prior information. Then, the parameter estimates of the test data are obtained by Gibbs sampling combined with Bayes formula. Finally, the reliability index under normal stress level is obtained by accelerating model extrapolation. The feasibility of the proposed method is verified by the accelerated life test data of a type of Ship communication equipment.

Source: Eksploatacja i Niezawodnosc - Maintenance and Reliability

DOI: 10.17531/ein/192170

Authors: Dazhang YOU, Shan LIU, Fangke LI, Hang LIU, Yepeng ZHANG

Institution: Hubei University of Technology

Language: English

Topic: Reliability and Maintenance Optimization

Subfield: Safety, Risk, Reliability and Quality

Field: Engineering

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: gold

Estimated APC paid: $442