Independent component analysis based on data‐driven reconstruction of multi‐fault diagnosis

Skip to Navigation

EarlyView Article

  • Published: Aug 23, 2017
  • Author: Lin Feng, Yingwei Zhang, Xuguang Li, Yuanjian Fu


Independent component analysis based on data‐driven reconstruction has been widely used in online fault diagnosis for industrial processes. As an alternative to conventional contribution plots, the reconstruction‐based fault diagnosis method has been drawing special attention. The method detects fault information with a specific reconstruction model based on historical fault data. In this paper, a novel method was proposed that focuses on handling multiple fault cases in abnormal processes. First, reconstruction‐based fault subspaces were extracted based on monitoring statistics in 2 different monitoring subspaces to enclose the major fault effects. Independent component analysis was then used to recover the main fault features from the selected fault subspaces, which represent the joint effects from multiple faults for online diagnosis. The simulation results showed the feasibility and performance of the proposed method with simulated multi‐fault cases in the Tennessee Eastman (TE) benchmark process.

Social Links

Share This Links

Bookmark and Share


Suppliers Selection
Societies Selection

Banner Ad

Click here to see
all job opportunities

Most Viewed

Copyright Information

Interested in separation science? Visit our sister site

Copyright © 2018 John Wiley & Sons, Inc. All Rights Reserved