PROJECT TITLE :

A New Degradation Indicator Based on a Statistical Anomaly Approach

ABSTRACT:

This paper presents a brand new technique for combining measured parameters into one indicator for monitoring the condition of systems subjected to degradation effects. The proposed approach integrates the employment of nonparametric density estimation techniques into Runger's U2 methodology, which allows for the separation of variables that are directly connected to degradation effects from those that are not. 2 simulated case studies are presented for illustration, particularly, the monitoring of a flap extension and retraction system, and a gas turbine used as an auxiliary power unit. For comparison, degradation indicators are also calculated by using Hotteling's T2 and Runger's U2 strategies, with a nonparametric technique without separation of variables. In both case studies, the proposed method provided the simplest ends up in terms of fault detection performance and suitability for remaining useful life prediction.