Life estimation based on unbalanced data for hydraulic pump

摘要

This paper focuses on statistic method under small test samples through re-sampling for aeronautics pump. Based on the limit accelerated life testing samples and regular life testing samples, this paper presents a life evaluation method combing the Synthetic Minority Over-Sampling Technique (SMOTE) algorithm, Kolmogorov-Smirnov (KS) test and accumulated damage theory. SMOTE algorithm is used to solve the imbalance between sample groups, while KS is a classic way to evaluate the goodness of fitting. The maximum likelihood estimation of proposed method shows that the sample group can be expanded efficiently with the goodness-of-fit being guaranteed.