PROJECT TITLE :

A Classification Framework for Predicting Components' Remaining Useful Life Based on Discrete-Event Diagnostic Data

ABSTRACT:

In this paper, we tend to propose to outline the problem of predicting the remaining helpful life of a component as a binary classification task. This approach is particularly useful for problems in that the evolution of the system condition is described by a mixture of a big range of discrete-event diagnostic data, and for which various approaches are either not applicable, or are solely applicable with vital limitations or with a massive computational burden. The proposed approach is demonstrated with a case study of real discrete-event information for predicting the occurrence of railway operation disruptions. For the classification task, Extreme Learning Machine (ELM) has been chosen as a result of of its sensible generalization ability, computational potency, and low needs on parameter tuning.