Entropy-Clustering and K-means based Kernel partial least squares soft-sensing method
详细信息 查看官网全文
- 英文论文题名：Entropy-Clustering and K-means based Kernel partial least squares soft-sensing method
- 论文作者：Bin Chen ; Yaqiong Wu ; Bo Tao ; Ying Zheng
- 英文论文作者：Bin Chen ; Yaqiong Wu ; Bo Tao ; Ying Zheng ; National Key Laboratory of Science and Technology on Multispectral Information Processing ; School of Automation ; Huazhong University of Science and Technology ; College of information Science & Technology ; Beijing University of Chemical Technology ; the State Key Laboratory of Digital Manufacturing Equipment and Technology ; School of Mechanical Science and Engineering ; Huazhong University of Science and Technology
- 年：2017
- 作者机构：National Key Laboratory of Science and Technology on Multispectral Information Processing,School of Automation,Huazhong University of Science and Technology;College of information Science & Technology,Beijing University of Chemical Technology;the State Key Laboratory of Digital Manufacturing Equipment and Technology,School of Mechanical Science and Engineering,Huazhong University of Science and Technology;
- 英文论文关键词：entropy clustering ; partial least squares regression ; K-means algorithm
- 会议召开时间：2017-07-26
- 会议录名称：第36届中国控制会议论文集（B）
- 英文会议录名称：Proceedings of the 36th Chinese Control Conference（B）
- 语种：英文
- 分类号：TP274
- 学会代码：KZLL
- 会议名称：第36届中国控制会议
- 会议地点：中国辽宁大连
- 主办单位：中国自动化学会控制理论专业委员会
- 学会名称：中国自动化学会控制理论专业委员会
- 页数：5
- 文件大小：313k
- 原文格式：D
- 会议级别：全国
摘要
Kernel partial least squares(KPLS) is widely adopted for soft-sensing in nonlinear industrial process. For KPLS method, the determination of central nodes and kernel width in the kernel function will affects generalization ability and predictiability. This paper proposes an entropy-clustering and K-means based KPLS regression method. First of all, it divides the original data into several clusters by entropy clustering method and obtains the initial clustering centers. Secondly, K-means algorithm is applied on these initial clustering centers to get the final central nodes of the kernel functions. Finally, the width of kernel function is determined according to the Euclidean distance between the central node of the kernel function and its adjacent central node. The proposed method is verified based on the process data from a chemical enterprise. The experiment results show that the proposed algorithm greatly reduces the measurement error compared to the traditional KPLS regression method.
Kernel partial least squares(KPLS) is widely adopted for soft-sensing in nonlinear industrial process. For KPLS method, the determination of central nodes and kernel width in the kernel function will affects generalization ability and predictiability. This paper proposes an entropy-clustering and K-means based KPLS regression method. First of all, it divides the original data into several clusters by entropy clustering method and obtains the initial clustering centers. Secondly, K-means algorithm is applied on these initial clustering centers to get the final central nodes of the kernel functions. Finally, the width of kernel function is determined according to the Euclidean distance between the central node of the kernel function and its adjacent central node. The proposed method is verified based on the process data from a chemical enterprise. The experiment results show that the proposed algorithm greatly reduces the measurement error compared to the traditional KPLS regression method.
Kernel partial least squares(KPLS) is widely adopted for soft-sensing in nonlinear industrial process. For KPLS method, the determination of central nodes and kernel width in the kernel function will affects generalization ability and predictiability. This paper proposes an entropy-clustering and K-means based KPLS regression method. First of all, it divides the original data into several clusters by entropy clustering method and obtains the initial clustering centers. Secondly, K-means algorithm is applied on these initial clustering centers to get the final central nodes of the kernel functions. Finally, the width of kernel function is determined according to the Euclidean distance between the central node of the kernel function and its adjacent central node. The proposed method is verified based on the process data from a chemical enterprise. The experiment results show that the proposed algorithm greatly reduces the measurement error compared to the traditional KPLS regression method.
引文
[1]Abdi H.Partial least square regression(PLS regression)[J].Encyclopedia for research methods for the social sciences,2003:792-795.
[2]Rosipal R,Trejo L J.Kernel partial least squares regression in reproducing kernel hilbert space[J].Journal of machine learning research,2001,2(Dec):97-123.
[3]Momma M,Bennett K P.Sparse kernel partial least squares regression[M]//Learning Theory and Kernel Machines.Springer Berlin Heidelberg,2003:216-230.
[4]Li B,Morris A J,Martin E B.Generalized partial least squares regression based on the penalized minimum norm projection[J].Chemometrics and Intelligent Laboratory Systems,2004,72(1):21-26.
[5]Ding Tao(Sch.of Commun.&Control Eng.,Jiangnan Univ.,Wuxi,China);Yang Hui-zhong.On improvement of generalization ability of partial least-squares regression model.Control Engineering China,v 15,n 2,p 150-3,20 March 2008
[6]Zuo X B,Fang S,Liang X L.Synergy interval partial least square(si PLS)with potentiometric titration multivariate calibration for the simultaneous determination of amino acids in mixtures[J].Advance Journal of Food Science and Technology,2014,6(11):1209-1218.
[7]Yong Zhao(Coll.of Inf.Sci.&Eng.,Northeastern Univ.,Shenyang,China);Shenghao Wang;Zhi Li;Zhenying Pei;Fuyi Cao.An improved changeable size moving window partial least square applied for molecular spectroscopy.Chemometrics and Intelligent Laboratory Systems,v 152,p118-24,15 March 2016
[8]NIU D,Chun-xiang L,MENG M.Annual load forecasting based on grey model and Partial Least-Square method[J].East China Electric Power,2009,6:030.
[9]Liu Y,Zhou S.Optimizing Neuron Function Based on Entropy Clustering in Functional Networks[C]//Computational Intelligence and Security,2008.CIS'08.International Conference on.IEEE,2008,2:152-156.
[10]Benoudjit N,Verleysen M.On the kernel widths in radial-basis function networks[J].Neural Processing Letters,2003,18(2):139-154.
[11]Mevik B H,Wehrens R.The pls package:principal component and partial least squares regression in R[J].Journal of Statistical software,2007,18(2):1-24.
[12]Huang S C.Forecasting stock indices with wavelet domain kernel partial least square regressions[J].Applied Soft Computing,2011,11(8):5433-5443.
[13]S.Chiu.Fuzzy model identification based on cluster estimation[J].Journal of Intelligent and Fuzzy Systems,1994,2(3):267-278.
[14]J.N.Kapur,H.K.Kasdvan.Entropy optimization principles with applications[M].New York:Academic Press,1992.
[15]Huang S C.Integrating spectral clustering with wavelet based kernel partial least square regressions for financial modeling and forecasting[J].Applied Mathematics and Computation,2011,217(15):6755-6764.
[16]Y.X.Huang,C.G.Zhou,G.H.Yang and S.X.Zou,“Identification of Taste Signals Based on an Entropy-based Clustering Fuzzy Neural Network,”Computer Research&Development,Mar.2004vol.41,no.3,pp.414-419.
[2]Rosipal R,Trejo L J.Kernel partial least squares regression in reproducing kernel hilbert space[J].Journal of machine learning research,2001,2(Dec):97-123.
[3]Momma M,Bennett K P.Sparse kernel partial least squares regression[M]//Learning Theory and Kernel Machines.Springer Berlin Heidelberg,2003:216-230.
[4]Li B,Morris A J,Martin E B.Generalized partial least squares regression based on the penalized minimum norm projection[J].Chemometrics and Intelligent Laboratory Systems,2004,72(1):21-26.
[5]Ding Tao(Sch.of Commun.&Control Eng.,Jiangnan Univ.,Wuxi,China);Yang Hui-zhong.On improvement of generalization ability of partial least-squares regression model.Control Engineering China,v 15,n 2,p 150-3,20 March 2008
[6]Zuo X B,Fang S,Liang X L.Synergy interval partial least square(si PLS)with potentiometric titration multivariate calibration for the simultaneous determination of amino acids in mixtures[J].Advance Journal of Food Science and Technology,2014,6(11):1209-1218.
[7]Yong Zhao(Coll.of Inf.Sci.&Eng.,Northeastern Univ.,Shenyang,China);Shenghao Wang;Zhi Li;Zhenying Pei;Fuyi Cao.An improved changeable size moving window partial least square applied for molecular spectroscopy.Chemometrics and Intelligent Laboratory Systems,v 152,p118-24,15 March 2016
[8]NIU D,Chun-xiang L,MENG M.Annual load forecasting based on grey model and Partial Least-Square method[J].East China Electric Power,2009,6:030.
[9]Liu Y,Zhou S.Optimizing Neuron Function Based on Entropy Clustering in Functional Networks[C]//Computational Intelligence and Security,2008.CIS'08.International Conference on.IEEE,2008,2:152-156.
[10]Benoudjit N,Verleysen M.On the kernel widths in radial-basis function networks[J].Neural Processing Letters,2003,18(2):139-154.
[11]Mevik B H,Wehrens R.The pls package:principal component and partial least squares regression in R[J].Journal of Statistical software,2007,18(2):1-24.
[12]Huang S C.Forecasting stock indices with wavelet domain kernel partial least square regressions[J].Applied Soft Computing,2011,11(8):5433-5443.
[13]S.Chiu.Fuzzy model identification based on cluster estimation[J].Journal of Intelligent and Fuzzy Systems,1994,2(3):267-278.
[14]J.N.Kapur,H.K.Kasdvan.Entropy optimization principles with applications[M].New York:Academic Press,1992.
[15]Huang S C.Integrating spectral clustering with wavelet based kernel partial least square regressions for financial modeling and forecasting[J].Applied Mathematics and Computation,2011,217(15):6755-6764.
[16]Y.X.Huang,C.G.Zhou,G.H.Yang and S.X.Zou,“Identification of Taste Signals Based on an Entropy-based Clustering Fuzzy Neural Network,”Computer Research&Development,Mar.2004vol.41,no.3,pp.414-419.