Fault Diagnosis Method Based on KPCA and Selective Neural Network Ensemble

Abstract:

Aiming at the problems of less study sample, large network scale and long training time existing in current fault diagnosis field, we develop a new method based on KPCA and selective neural network ensemble. First, reducing the data size by using KPCA to extract the sample features. Then achieving a selective neural network ensemble method based on improved binary particle swarm optimization algorithm (IBPSOSEN), and combining the two methods for fault diagnosis. In selective neural network algorithm, bagging method is used to take a number of different training sets of fault samples to solve the problem of less fault samples. Finally, simulation experiments and comparisons over Tennessee Eastman Process (TE) demonstrate the effectiveness and feasibility of the proposed method.

• Periodical: Advanced Materials Research (Volumes 915-916)
• Pages: 1272-1276
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.915-916.1272
• Online since: April 2014
• Authors: Yan Yan Pang, Hai Ping Zhu, Fan Mao Liu
• Keywords: Fault Diagnosis, Kernel Principal Component Analysis (KPCA), Selective Neural Network Ensemble

References

[1] LI Y W, WEI G Z, YING L. Rolling bearing fault diagnosis based on wavelet packet-neural network characteristic entropy[J]. Advanced Materials Research, 2010, 108(1): 1075-1079.

DOI: 10.4028/www.scientific.net/amr.108-111.1075

[2] WONG W K, YUEN C W M, FAN D D. Stitching defect detection and chassification using wavelet transform and BPneural network[J]. Expert Systems with Applications, 2009, 36(2): 3845-3856.

DOI: 10.1016/j.eswa.2008.02.066

[3] GUO L J, GAO J J, YANG J F, etal. Criticality evaluation of petrochemical equipment based on fuzzy comprehensive evaluation and a BP neural network[J]. Journal of Loss Prevention in the Process Industries, 2009, 22(4): 469-476.

DOI: 10.1016/j.jlp.2009.03.003

[4] ZHOU Zhi-hua, WU Jian-xin, TANG Wei. Ensembling neural networks: many could be better than all [J]. Artificial Intelligence, 2002, 137: 239- 263.

DOI: 10.1016/s0004-3702(02)00190-x

[5] Lee H, Hong S, Kim E. An efficient gait recognition based on a selective neural network ensemble[J]. International Journal of Imaging Systems and Technology, 2008, 18(4): 237-241.

DOI: 10.1002/ima.20136

[6] Yu H, Liu D, Chen G, et al. A neural network ensemble method for precision fertilization modeling[J]. Mathematical and Computer Modelling, 2010, 51(11): 1375-1382.

DOI: 10.1016/j.mcm.2009.10.028

[7] Kennedy J, Eberhart R. Particle swarm optimization[C]/Proceedings of IEEE international conference on neural networks. 1995, 4(2): 1942-(1948).

[8] Lyman P R, Georgakis C. Plant-wide control of the Tennessee Eastman problem [J]. Computers & Chemical Engineering, 1995, 19(3): 321-331.

DOI: 10.1016/0098-1354(94)00057-u

[9] Chiang L H, Braatz R D, Russell E L. Fault detection and diagnosis in industrial systems[M]. Springer, (2001).