IEEE Access, Early Access, September 23, 2019, DOI:10.1109/ACCESS.2019.2943076

A Weighted Deep Domain Adaptation Method for Industrial Fault Prognostics according to Prior Distribution of Complex Working Conditions

Zhenyu Wu1, Shuyang Yu2, Xinning Zhu2, Yang Ji2, and Michael Pecht3

1Engineering Research Center of Information Network, Ministry of Education, Beijing Univ. of Posts and Telecommunications, Beijing, 100876 P.R. China.
2Key Laboratory of Universal Wireless Communications, Ministry of Education, Beijing Univ. of Posts and Telecommunications, Beijing, 100876 P.R. China.
3Center for Advanced Life Cycle Engineering, University of Maryland at College Park, College Park, MD 20742, USA

Abstract:

In modern industrial engineered systems, variant working conditions disturb the distributions of machines' operational data, which results in different feature distributions (DFD) problems for fault prognostics. Domain adaptation (DA) have been proved good at dealing DFD problems, and several deep DA-based methods have been also proposed in fault prognostics filed. However, existing methods refer to the DA tasks from one working condition to another, without considerations of transferring between datasets under complex working conditions. The prior distribution of working conditions will influence the distributions of machines' operational data, and few studies take prior distribution of working conditions into consideration of DA for fault prognostics. Thus, in this paper, a workingcondition-based deep domain adaptation network (Deep wcDAN) is proposed to overcome the DFD problems caused by variant complex working conditions. In the proposed method, CNNs combines LSTMs with domain adaptive transfer technique to minimize the distribution discrepancy between training and testing datasets. Furthermore, a working-condition-based MMD (wcMMD) is proposed to optimize the DA process based on the prior distribution of each working condition. The performance of proposed model is evaluated and the negative transfer effects have been analyzed based on C-MAPSS datasets. The results show that the proposed method performs better than baseline methods on predicting remaining useful life (RUL) with DFD problems.

This article is available online here.

[Home Page] [Articles Page]
Copyright © 2019 by CALCE and the University of Maryland, All Rights Reserved