Please wait a minute...
浙江大学学报(工学版)
J4  2010, Vol. 44 Issue (7): 1255-1259    DOI: 10.3785/j.issn.1008-973X.2010.07.004
    
Feature space dimensionreduction based process monitoring of solvent dehydration separation process
杜文莉, 王坤, 钱锋
Key Laboratory of Advanced Control and Optimization for Chemical Processes, Ministry of Education, East China University of Science and Technology, Shanghai 200237,China
Download:   PDF(0KB) HTML
Export: BibTeX | EndNote (RIS)      

Abstract  

The measurement variables of chemical process usually show the characteristic of nonlinear and nonGaussian behaviors. A novel modeling method was proposed by integrating the improved kernel principal component analysis (KPCA) with support vector data description (SVDD). The Mexican hat wavelet function was introduced to construct the kernel function by utilizing the advantage of extracting the subtle feature of nonlinear nonstationary signal. The nonlinear mapping and antinoise capability of kernel function was enhanced. Then the cluster analysis was used in the kernel feature space. The data that represented the characteristic center in every cluster were chosen, which can decrease the computational complexity and improve the results of realtime monitor. Furthermore, the SVDD was adopted to describe the feature space with dimensionreduction, and a new monitor index was constructed by SVDD to describe the nonGaussian information. The method was applied to a solvent dehydration distillation process. Simulation results demonstrate that the method can detect the fault promptly and effectively.

Published: 01 July 2010
CLC:  TP 277  
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
Cite this article:

DU Wen-Chi, WANG Kun, JIAN Feng. Feature space dimensionreduction based process monitoring of solvent dehydration separation process. J4, 2010, 44(7): 1255-1259.

URL:

http://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2010.07.004     OR     http://www.zjujournals.com/eng/Y2010/V44/I7/1255


基于特征空间降维的溶剂脱水分离过程监控

针对传统化工过程中检测变量具有的非线性和非高斯性等特点,提出将改进的核主元分析（KPCA）和支持向量数据描述（SVDD）相结合的化工过程故障诊断方法.根据Mexican hat小波在提取非线性非平稳信号细微特征方面的优势,将该小波函数引入到KPCA中以增强核函数的非线性映射和抗噪能力.在映射后的特征空间中进行均值聚类分析,选择每个聚类中展现特征中心的数据,使运算复杂度明显降低,提高了监控实时性.采用SVDD描述经过聚类降维后的特征空间分布,提出新的监控指标描述过程的非高斯特性.将该方法应用在一个实际的溶剂脱水化工精馏过程中,仿真结果验证了该方法能够及时有效地检测系统产生的故障.

［1］ QIN S J. Statistical process monitoring basics and beyond ［J］. Journal of Chemometrics, 2003, 17(8): 480502.

［2］ MISRA M, YUE H H, QIN S J, et al. Multivariate process monitoring and fault diagnosis by multiscale PCA ［J］. Computers and Chemical Engineering, 2002, 26(9): 12811239.

［3］ SCHOLKOPF B, SMOLA A J, MULLER K. Nonlinear component analysis as a kernel eigenvalue problem ［J］. Neural Computation, 1998, 10(5): 12991319.

［4］ LEE J M, YOO C K, CHOI S W, et al. Nonlinear process monitoring using kernel principal component analysis ［J］. Chemical Engineering Science, 2004, 59 (1): 223234.

［5］ CHOI S W, LEE C, LEE J M, et al. Fault detection and identification of nonlinear processes based on kernel PCA ［J］. Chemometrics and Intelligent Laboratory Systems, 2005, 28(1): 5567.

［6］ JOHN S, TAYLOR N C. Kernel methods for pattern analysis ［M］. Cambridge: Cambridge University Press, 2004: 3785.

［7］ GANEY J L, BLOCK W M, JENNESS J S, et al. Mexican spotted owl home range and habitat use in pineoak forest ［J］. Forest Science, 1999, 45(1): 127135.

［8］ ZHANG L, ZHOU W, JIAO L. Wavelet support vector machine ［J］. IEEE Transactions on Systems, Man and Cybernetics, Part B: Cybernetics, 2004, 34(1): 3439.

［9］ CHIANG L H, RUSSELL E L, BRAATZ R D. Fault detection and diagnosis in industrial systems ［J］. Technometrics, 2002, 44(2): 197198.

［10］ KRZANOWSKI W J. Betweengroups comparison of principal components ［J］. American Statistical Association, 1979, 74(36): 703707.


［11］ ZHANG Y W, QIN J S. Improved nonlinear fault detection technique and statistical analysis ［J］. Wiley InterScience, 2008, 54(12): 32073220.

［12］ 刘育明,梁军,毛勇,等.基于支持向量机的MSPC方法及其应用［J］.浙江大学学报:工学版,2006,40(10): 17211724.

LIU Yuming, LIANG Jun, MAO Yong, et al. SVMbased multivariate statistical process control method and its application ［J］. Journal of Zhejiang University: Engineering Science, 2006, 40(10): 17211724.

［13］ TAX D M J, DUIN R P W. Support vector domain description ［J］. Machine Learning, 2004, 54(1): 4566.

［14］ TAX D M J, DUIN R P W. Support vector domain description ［J］. Pattern Recognition Letters,1999,20(11/13): 11911199.
[1] DUAN Bin, LIANG Jun, FEI Zheng-shun, YANG Min, HU Bin. Nonlinear semi-parametric modeling mothed based on GA-ANN[J]. J4, 2011, 45(6): 977-983.
[2] XU Gui-Bin, ZHOU Dong-Hua. Fault prediction for state-dependent fault based on online learning neural network[J]. J4, 2010, 44(7): 1251-1254.
[3] LIN Yong, ZHOU Xiao-Jun, YANG Xian-Yong, DENG. Intelligent fault diagnosis methods based on bispectrum recognition and artificial immune network[J]. J4, 2009, 43(10): 1777-1782.