Hilbert-Huang Transform in Fault Detection

Failures of manufacturing systems represent a significant cost in both repairs and downtime. A considerable percentage of the maintenance cost is caused by unexpected drivetrain failures. This paper proposes a Hilbert-Huang transform (HHT)-based algorithm to effectively detect malfunctioning by revealing the instantaneous amplitude and frequency of nonlinear and nonstationary signals. The proposed algorithm is tested on synthetic signals created for this purpose. This paper presents the principles of this fault diagnosis method, results indicate that the proposed procedure can handle fault detection issues in manufacturing systems. (C) 2019 The Authors. Published by Elsevier Ltd.