Lifetime Evaluation Method Based on Small Samples and Multi-Source Data

Abstract

For the spaceborne components with long lifetime and high reliability, we may get a relatively small number of lifetime samples, but the sample may have more sources. Therefore, how to effectively use these small sample multi-source data is a difficult problem. This paper proposes a comprehensive lifetime assessment method for the spaceborne components which is based on small sample multi-source data and uses Bayesian theory to fuse all data. Firstly, the small sample data of the spaceborne components is regarded as the sample information in the Bayes theory, and the other information of the lifetime data is regarded as the prior information of the Bayes theory. Secondly, according to the method of determining the weight value of the second-type maximum likelihood estimation to obtain the best prior distribution, Multi-source prior information is weighted fusion. Then, the best prior distribution is combined with small sample data to obtain the Bayes posteriori distribution. In the end the comprehensive lifetime assessment of spaceborne component is completed. Finally, the proposed method is used to estimate the lifetime data of a certain type of device.