Title :
Multi-step ahead fault prediction method based on PCA and EMD
Author :
Wang, Shu ; Zhao, Zhen ; Wang, Fuli ; Chang, Yuqing
Author_Institution :
Key Lab. of Integrated Autom. of Process Ind., Northeastern Univ., Shenyang, China
Abstract :
In recent years, fault prediction method, which means forecast process fault in an early time based on the current condition of the system, has attracted more and more attention by companies and scientists. However, it still has many problems in this area, especially for its application in industrial process. In the present work, a multi-step ahead fault prediction method combining principle component analysis, empirical mode decomposition and extreme learning machine are developed to realize early prediction of fault. The application of the presented method is illustrated with respect to simulated data collected from the Tennessee Eastman process. The experimental results demonstrate the effectiveness of the proposed method.
Keywords :
condition monitoring; fault diagnosis; learning (artificial intelligence); principal component analysis; EMD; PCA; Tennessee Eastman process; empirical mode decomposition; extreme learning machine; fault forecast process; industrial process; multistep ahead fault prediction method; principle component analysis; Fault diagnosis; Indexes; Predictive models; Principal component analysis; Process control; Time series analysis; empirical mode decomposition (EMD); extreme learning machine (ELM); fault diagnosis; fault prediction; principle component analysis (PCA); signal processing;
Conference_Titel :
Control and Decision Conference (CCDC), 2011 Chinese
Conference_Location :
Mianyang
Print_ISBN :
978-1-4244-8737-0
DOI :
10.1109/CCDC.2011.5968743