DocumentCode
420809
Title
Study on chemical process faults diagnosis based on fractal geometry
Author
Chuang, Huang ; Hongbo, Shi
Author_Institution
Inst. of Autom., East China Univ. of Sci. & Technol., Shanghai, China
Volume
2
fYear
2004
fDate
15-19 June 2004
Firstpage
1658
Abstract
Fractal geometry theory cooperated with wavelet transform and neural networks were applied to chemical process fault diagnosis in this contribution. The basic idea is to diagnose faults by comparing capacity dimensions of signal curves. The effectiveness of the proposed method was tested through the study on TE problem, and the simulation was developed. The results show that the capacity dimensions are similar when the faults belong to same kind and vice versa. Therefore capacity dimensions can be an important evidence to differentiate process faults.
Keywords
chemical industry; fault diagnosis; fractals; neural nets; wavelet transforms; chemical process faults diagnosis; fractal geometry theory; neural networks; signal curves; wavelet transform; Automation; Chemical processes; Chemical technology; Electronic mail; Fault diagnosis; Fractals; Geometry; Neural networks; Testing; Wavelet transforms;
fLanguage
English
Publisher
ieee
Conference_Titel
Intelligent Control and Automation, 2004. WCICA 2004. Fifth World Congress on
Print_ISBN
0-7803-8273-0
Type
conf
DOI
10.1109/WCICA.2004.1340936
Filename
1340936
Link To Document