Trajectory Similarity-Based Prediction with Information Fusion for Remaining Useful Life

摘要

Prediction of remaining useful life (RUL) has widely application in industrial domain, especially for aircraft where safety and reliability are of high importance. RUL Prediction can provide the time of failure for a degrading system, so that there are high requirements of its accuracy. In this paper, we propose a new trajectory similarity-based RUL prediction approach with an information fusion strategy (named IF-TSBP) in the similarity measure step. The novel information fusion strategy allows us to get more precise trajectory similarity degree than traditional similarity measure strategy which contributes to the prediction result. The experimental results show that the prediction accuracy of our proposed algorithm IF-TSBP outperforms the traditional trajectory similarity-based prediction approach and some common machine learning algorithms.