Abstract
Aiming at the problem of continuous model updating for fault recognition in the time-varying process, a novel method called the Procrustes analysis–based Fisher discriminant analysis was proposed. First, each class of the training data was preprocessed by Procrustes analysis. Second, the new test data were aligned with each class of the training data by Procrustes analysis. Then, all the data were reduced to a low-dimensional space using Fisher discriminant analysis. Finally, the Euclidean distance between the test data and the training data after the Procrustes analysis was calculated, and the class recognition was achieved based on the discriminant principle of Fisher discriminant analysis. Two case studies show that the proposed Procrustes analysis–based Fisher discriminant analysis is superior to the traditional method based on Fisher discriminant analysis, and it can be used for fault recognition in a new and efficient way.
Highlights
Fault detection and accurate fault diagnosis of chemical processes are critical to guarantee process safety and improve production quality
Based on the proposed Procrustes analysis–based Fisher discriminant analysis (PFDA) model, a fault classification scheme was constructed in this work by adjusting the online process data to make it fit the structure of the modeling data before performing Fisher discriminant analysis (FDA)
Based on the presented results, it can be concluded that the PFDA model outperforms the FDA-based classification for most type of faults, especially for fault 13 and fault 15, and a large amount of improvement is obtained
Summary
Fault detection and accurate fault diagnosis of chemical processes are critical to guarantee process safety and improve production quality. Based on the proposed PFDA model, a fault classification scheme was constructed in this work by adjusting the online process data to make it fit the structure of the modeling data before performing FDA. The PFDA was applied to deal with the plant operating data mixed with both normal and various process faults.[33] In the case studies, we first constructed the classification model for the fault data based on the training data and aligned the test data to the training data to optimally match the fault type. In order to simulate time-varying behaviors, a linear increment of 0.0005 was added to t after the operating mode had been carried out Such types of fault data were stored as the test set.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
