Similarity-based Residual Useful Life Prediction for P artially Unknown Cycle Varying Degradation

摘要: Similarity-based approach is a popular data-driven prognostic method for Residual Useful Life (RUL) Prediction. The principle of this based on the "similarity" between monitored part, i.e., sample whose RUL has to be predicted and degradation reference trajectory patterns (or known library priori functions). challenge addressed in paper concerns estimation test using observations depending acquisition time dataset information, built knowledge endurance data. coefficient estimated here mapping function aperiodic cycle functions. Our evaluated very classical Virkler crack- growth measurements benchmark (14) experimental results show that it efficiency promising context railways maintenance. Index Terms—Prediction, Data-driven model, Aperiodic sam- pling, Rail inspection