基于核慢特征分析和时滞估计的GPR建模

doi:10.16182/j.issn1004731x.joss.17-0389

系统仿真学报 ›› 2019, Vol. 31 ›› Issue (8): 1562-1571.doi: 10.16182/j.issn1004731x.joss.17-0389

基于核慢特征分析和时滞估计的GPR建模

彭慧来¹, 熊伟丽^1,2

1. 江南大学物联网工程学院自动化研究所,无锡 214122;
2. 江南大学轻工过程先进控制教育部重点实验室,无锡 214122

收稿日期:2017-08-09 修回日期:2017-11-04 发布日期:2019-12-12
作者简介:彭慧来（1991-）,男,江西上饶,硕士生,研究方向为工业过程建模;熊伟丽（1978-）,女,江苏无锡,博士,教授,博导,研究方向为复杂工业过程建模及优化、软测量技术。
基金资助:
国家自然科学基金(61773182),国家重点研发计划子课题(2018YFC1603705-03)

GPR Modeling Method Based on Kernel Slow Feature Analysis and Time Delay Estimation

Peng Huilai¹, Xiong Weili^1,2

1. College of Internet of Things Engineering, Jiangnan University, Wuxi 214122, China;
2. Key Laboratory of Advanced Process Control for Light Industry (Ministry of Education), Jiangnan University, Wuxi 214122, China

Received:2017-08-09 Revised:2017-11-04 Published:2019-12-12

摘要/Abstract

摘要： 针对工业过程中存在的时滞和非线性特性,提出了一种基于核慢特征分析的高斯过程回归建模方法。该方法通过模糊曲线分析充分挖掘工业数据中的时滞信息,求出数据中的最优时滞,并进行建模数据的重构;通过核慢特征分析方法对重构数据进行非线性的特征提取;基于提取后的特征建立高斯过程回归(GPR)模型。通过对脱丁烷塔塔底丁烷浓度软测量的仿真实验,验证了方法的有效性与性能。

关键词: 时延, 核慢特征分析, 模糊曲线分析, 高斯过程回归

Abstract: A Gaussian process regression modeling method based on kernel slow feature analysis is proposed to deal with the time delay and nonlinear characteristics in industrial processes. The time delay in the industrial data is effectively extracted by fuzzy curve analysis and the optimal time-delay in the data is obtained. The model data is reconstructed based on the optimal delay. The method of kernel slow feature analysis is used to extract the nonlinear features of the reconstructed data. The Gaussian process regression model is established based on the extracted features. The effectiveness and performance of the method are verified by the simulation experiment of the soft measurement of butane concentration at the bottom of the de-butane tower.

Key words: time delay, kernel slow feature analysis, fuzzy curve analysis, Gaussian process regression

中图分类号:

TP273

彭慧来, 熊伟丽. 基于核慢特征分析和时滞估计的GPR建模[J]. 系统仿真学报, 2019, 31(8): 1562-1571.

Peng Huilai, Xiong Weili. GPR Modeling Method Based on Kernel Slow Feature Analysis and Time Delay Estimation[J]. Journal of System Simulation, 2019, 31(8): 1562-1571.

参考文献

[1] Yuan X, Ge Z, Song Z.Locally weighted kernel principal component regression model for soft sensing of nonlinear time-variant processes[J]. Industrial & Engineering Chemistry Research (S0888-5885), 2014, 53(35): 13736-13749.
[2] Petr K, Ratko G, Bogdan G.Review of adaptation mechanisms for data-driven soft sensors[J]. Computers and Chemical Engineering (S0098-1354), 2011, 35(1): 1-24.
[3] Facco P, Doplicher F, Bezzo F, et al.Moving average PLS soft sensor for online product quality estimation in an industrial batch polymerization process[J]. Journal of Process Control (S0959-1524), 2009, 19(3): 520-529.
[4] Sun K, Liu J, Kang J L, et al.Development of a variable selection method for soft sensor using artificial neural network and nonnegative garrote[J]. Journal of Process Control (S0959-1524), 2014, 24(7): 1068-1075.
[5] 刘毅, 王海清. 采用最小二乘支持向量机的青霉素发酵过程建摸研究[J]. 生物工程学报, 2006, 22(1): 144-149.
Liu Yi, Wang Haiqing.Modeling optimization of SVM for a fermentation soft sensor[J]. Chinese Journal of Biotechnology, 2006, 22(1): 144-149.
[6] 何志昆, 刘光斌, 赵曦晶, 等. 高斯过程回归方法综述[J]. 控制与决策, 2013, 28(8): 1121-1129.
He Zhikun, Liu Guangbin, Zhao Xijing, et al.Overview of Gaussian process regression[J]. Control and Decision, 2013, 28(8): 1121-1129.
[7] Zhang W, Li Y, Xiong W, et al.Adaptive soft sensor for online prediction based on enhanced moving window GPR[C]//Control, Automation and Information Sciences (ICCAIS), 2015 International Conference on. Changshu China: IEEE, 2015: 291-296.
[8] 范雪莉, 冯海泓, 原猛. 基于互信息的主成分分析特征选择算法[J]. 控制与决策, 2013, 28(6): 915-919.
Fan Xueli, Feng Haihong, Yuan Meng, et al.PCA based on mutual information for feature selection[J]. Control and Decision, 2013, 28(6): 915-919.
[9] Lee J M, Yoo C K, Lee I B.Statistical process monitoring with independent component analysis[J]. Journal of Process Control (S0959-1524), 2004, 14(5): 467-485.
[10] Zhang Y.Enhanced statistical analysis of nonlinear processes using KPCA, KICA and SVM[J]. Chemical Engineering Science (S0009-2509), 2009, 64(5): 801-811.
[11] Wiskott L, Sejnowski T J.Slow feature analysis: Unsupervised learning of invariances[J]. Neural computation (S0899-7667), 2002, 14(4): 715-770.
[12] Blaschke T, Zito T, Wiskott L.Independent slow feature analysis and nonlinear blind source separation[J]. Neural computation (S0899-7667), 2007, 19(4): 994-1021.
[13] Shang C, Yang F, Gao X, et al.Extracting latent dynamics from process data for quality prediction and performance assessment via slow feature regression[C]//American Control Conference (ACC), 2015. Chicago, IL, USA: IEEE, 2015: 912-917.
[14] Shang C, Huang B, Yang F, et al.Probabilistic slow feature analysis‐based representation learning from massive process data for soft sensor modeling[J]. AIChE Journal (S0001-1541), 2015, 61(12): 4126-4139.
[15] Zhang Z, Tao D.Slow feature analysis for human action recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence (S0162-8828), 2012, 34(3): 436-450.
[16] Shang C, Huang B, Yang F, et al.Slow feature analysis for monitoring and diagnosis of control performance[J]. Journal of Process Control (S0959-1524), 2016, 39: 21-34.
[17] Zhang L, Wu C, Du B.Automatic radiometric normalization for multitemporal remote sensing imagery with iterative slow feature analysis[J]. IEEE Transactions on Geoscience and Remote Sensing (S0196-2892), 2014, 52(10): 6141-6155.
[18] Böhmer W, Grünewälder S, Nickisch H, et al.Regularized sparse kernel slow feature analysis[J]. Machine Learning and Knowledge Discovery in Databases (S0302-9743), 2011: 235-248.
[19] Böhmer W, Grünewälder S, Nickisch H, et al.Generating feature spaces for linear algorithms with regularized sparse kernel slow feature analysis[J]. Machine Learning (S0885-6125), 2012, 89(1/2): 67-86.
[20] Fortuna L, Graziani S, Rizzo A, et al.Soft sensors for monitoring and control of industrial processes[M]. London: Springer Science & Business Media, 2007.
[21] Zhang J, Jin Q, Xu Y.Inferential estimation of polymer melt index using sequentially trained bootstrap aggregated neural networks[J]. Chemical engineering & technology (S0930-7516), 2006, 29(4): 442-448.
[22] 阮宏镁, 田学民, 王平. 基于联合互信息的动态软测量方法[J]. 化工学报, 2014, 65(11): 4497-4502.
Ruan Hongmei, Tian Xuemin, Wang Ping.Dynamic soft sensor method based on joint mutual information[J]. CIESC Journal, 2014, 65(11): 4497-4502.
[23] Lin Y, Cunningham G A.A new approach to fuzzy-neural system modeling[J]. IEEE Transactions on Fuzzy systems (S1063-6706), 1995, 3(2): 190-198.
[24] 蒋昕祎, 杜红彬, 李绍军. 基于核慢特征回归与互信息的常压塔软测量建模[J]. 化工学报, 2017, 68(5): 1977-1986.
Jiang Xinyi, Du Hongbin, Li Shaojun.Atmospheric tower soft sensor based on kernel slow feature regression and mutual information[J]. CIESC Journal, 2017, 68(5): 1977-1986.
[25] 刘瑞兰. 软测量技术若干问题的研究及工业应用[D]. 杭州: 浙江大学, 2004.
Liu Ruilan.Some Studies on Soft Sensor Technology and Their Applications to Industry Process[D]. Hangzhou: Zhejiang University, 2004.
[26] Fortuna L, Graziani S, Xibilia M G.Soft sensors for product quality monitoring in debutanizer distillation columns[J]. Control Engineering Practice (S0967-0661), 2005, 13(4): 499-508.

基于核慢特征分析和时滞估计的GPR建模

GPR Modeling Method Based on Kernel Slow Feature Analysis and Time Delay Estimation

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 8

编辑推荐

Metrics

本文评价

[1]	周博文, 熊伟丽. 基于协方差矩阵的主动学习方法及应用研究[J]. 系统仿真学报, 2022, 34(3): 452-460.
[2]	祁成, 熊伟丽. 基于BGMM的即时学习软测量建模方法[J]. 系统仿真学报, 2019, 31(8): 1555-1561.
[3]	王琳, 仲崇权. 交流伺服系统网络控制的仿真研究[J]. 系统仿真学报, 2019, 31(8): 1674-1682.
[4]	赵帅, 史旭东, 熊伟丽. 一种分层集成的高斯过程回归软测量建模方法[J]. 系统仿真学报, 2019, 31(10): 2042-2051.
[5]	李世峰, 邱占芝, 樊丽萍, 赵丽娜. 基于改进BP网络的网络控制系统建模与仿真[J]. 系统仿真学报, 2018, 30(6): 2279-2287.
[6]	邱占芝, 李世峰. 基于神经网络的PID网络化控制系统建模与仿真[J]. 系统仿真学报, 2018, 30(4): 1423-1432.
[7]	刘义才, 刘斌, 石安伟. 区间化随机时延的网络控制系统建模与控制[J]. 系统仿真学报, 2018, 30(2): 654-663.
[8]	李春胜, 张雪. 一类具有双向时延的脑皮层电模型的分支研究[J]. 系统仿真学报, 2015, 27(12): 3063-3069.