Title :
A fast combined fault diagnosis approach based on LWSVM
Author :
Wu, Dongsheng ; Yang, Qing ; Tian, Feng ; Wang, Dazhi
Author_Institution :
Sch. of Inf. Sci. Eng., Shenyang Ligong Univ., Shenyang, China
Abstract :
To fast monitor process, a combined approach of fault diagnosis approach based on Lifting Wavelets and SVM(LWSVM) was presented. Firstly the data was pre-processed to remove noise and spikes through lifting scheme wavelets, which is faster than the traditional wavelets. Then SVM was used to diagnose the faults in process. To validate the performance and effectiveness of the proposed scheme, LWSVM was applied to diagnose the faults in TE Process. The results were given to show the effectiveness of these improvements for fault diagnosis performance in terms of low computational cost and high fault diagnosis accuracy.
Keywords :
fault diagnosis; process monitoring; production engineering computing; support vector machines; wavelet transforms; LWSVM; SVM; TE process; fast combined fault diagnosis approach; lifting wavelets; Cooling; Fault diagnosis; Feeds; Kernel; Monitoring; Support vector machines; Wavelet analysis; Fault detection and diagnosis; LWSVM; Lifting wavelets; TE process;
Conference_Titel :
Signal Processing Systems (ICSPS), 2010 2nd International Conference on
Conference_Location :
Dalian
Print_ISBN :
978-1-4244-6892-8
Electronic_ISBN :
978-1-4244-6893-5
DOI :
10.1109/ICSPS.2010.5555216