基于粗差判别的参数优化自适应加权最小二乘支持向量机在PX氧化过程参数估计中的应用

doi:10.3969/j.issn.0438-1157.2012.12.029

化工学报 ›› 2012, Vol. 63 ›› Issue (12): 3943-3950.DOI: 10.3969/j.issn.0438-1157.2012.12.029

基于粗差判别的参数优化自适应加权最小二乘支持向量机在PX氧化过程参数估计中的应用

陶莉莉¹, 钟伟民¹, 罗娜², 钱锋¹

1. 华东理工大学化工过程先进控制和优化技术教育部重点实验室, 上海 200237;
2. 华东理工大学自动化系, 上海 200237

收稿日期:2012-08-02 修回日期:2012-08-12 出版日期:2012-12-05 发布日期:2012-08-29
通讯作者: 钟伟民,钱锋
作者简介:陶莉莉(1983-),女,博士研究生。
基金资助:
国家重点基础研究发展计划项目(2012CB720500);国家自然科学基金项目重点基金(U1162202);国家自然科学基金项目(61174118);上海市重点学科建设项目(B504)。

Adaptive weighted least square support vector machine regression with gross error detection and its application to estimate kinetic parameters for industrial oxidation of p-xylene

TAO Lili¹, ZHONG Weimin¹, LUO Na², QIAN Feng¹

1. Key Laboratory of Advanced Control and Optimization for Chemical Processes, Ministry of Education, East China University of Science and Technology, Shanghai 200237, China;
2. Automation Institute, East China University of Science and Technology, Shanghai 200237, China

Received:2012-08-02 Revised:2012-08-12 Online:2012-12-05 Published:2012-08-29
Supported by:
supported by the National Basic Research Program of China(2012CB720500),the Key Program of the National Natural Science Foundation of China(U1162202),the National Natural Science Foundation of China(61174118)and the Leading Academic Discipline Project of Shanghai(B504).

摘要/Abstract

摘要： 针对软测量建模过程中数据可能存在粗大误差以及粗差数据对模型的性能产生的影响,提出了一种基于粗差判别的自适应加权最小二乘支持向量机回归方法(WLS-SVM)。该方法首先根据3δ法则检测出样本中的显著误差并加以剔除,然后根据样本误差的大小自适应地调整权值,使得非显著误差对模型性能的影响大大降低。另外,由于最小二乘支持向量机的正则化参数和核宽度参数对模型的拟合精度和泛化能力有较大的影响,一般依靠经验和试算的方法进行估计,耗时且不准确,本文将模型的参数作为进化算法的优化问题,应用自适应免疫算法(AIGA)对参数进行优化选择。仿真实验表明,该方法对非线性系统的建模具有很好的效果。同时,将该方法应用于工业PX氧化建模过程中动力学参数的估计中,结果表明,基于粗差判别的参数优化自适应最小二乘支持向量机预测精度高,取得了较好的效果。

关键词: 粗差, 加权最小二乘支持向量机, 免疫算法, PX氧化过程建模

Abstract: The presence of gross errors can corrupt a model’s performance,giving undesirable results.A novel weighted least square support vector machine regression(WLS-SVM)is proposed,which combines gross error detection and adaptive weight value for the training sample.First,the 3δ principle is applied to detect the gross error.Second,the initial weight is obtained according to the fitting error of each sample.Then,an adaptive immune algorithm(AIGA)is applied to obtain the optimal parameters of the WLS-SVM.To illustrate the performance of the WLS-SVM,simulation experiment is designed to produce the training sample.The results showed that the predicting performance of AIGA-WLS-SVM is the best. Furthermore,the AIGA-WLS-SVM method was applied to estimate the rate constants of an industrial p-xylene oxidation model,and the satisfactory result was obtained.

Key words: gross error, weighted least square support vector machine, immune algorithm, p-xylene oxidation model

中图分类号:

TP181

陶莉莉, 钟伟民, 罗娜, 钱锋. 基于粗差判别的参数优化自适应加权最小二乘支持向量机在PX氧化过程参数估计中的应用[J]. 化工学报, 2012, 63(12): 3943-3950.

TAO Lili, ZHONG Weimin, LUO Na, QIAN Feng. Adaptive weighted least square support vector machine regression with gross error detection and its application to estimate kinetic parameters for industrial oxidation of p-xylene[J]. CIESC Journal, 2012, 63(12): 3943-3950.

参考文献 13

[1]	Cristianini Nello,Shawe-Taylor John.An Introduction to Support Vector Machines and Other Kernel-based Learning Methods[M].Cambridge:Cambridge University Press,2000
[2]	Lee Dong Eon,Song Ji-Ho,Song Sang-Oak,Yoon En Sup. Weighted support vector machine for quality estimation in the polymerization process[J].Process Design and Control,2005,44:2101-2105
[3]	Zhang X.Using class-center vectors to build support vector machines[J].IEEE Workshop on Neural Networks for Signal Processing,1999,31:3-11
[4]	Lu Zengxiang(卢增祥),Li Yanda(李衍达).Interactive support vector machine learning algorithm and its application[J].Journal of Tsinghua University:Science and Technology(清华大学学报:自然科学版),1999,39(7):93-97
[5]	Suykens J A K,de Brabanter J,Lukas L,Vandewalle J. Weighted least squares support vector machines:robustness and sparse approximation[J].Neurocomputing,2002,48:85-105
[6]	Li Xin(李昕),Yan Xuefeng(颜学峰).Steady state identification method containing outliers detection and its application[J].Journal of East China University of Science and Technology:Natural Science Edition(华东理工大学学报:自然科学版),2009,35(1):144-148
[7]	Liu Ruilan(刘瑞兰),Mu Shengjing(牟盛静),Chu Jian(褚健).Modeling soft sensor based on support vector machine and particle swarm optimization algorithms[J].Control Theory & Applications(控制理论与应用),2006,23(6):895-900
[8]	Lin Bihua(林碧华),Gu Xingsheng(顾幸生).Soft sensor modeling based on DE-LSSVM[J].Journal of Chemical Industry and Engineering(China)(化工学报),2008,59(7):1681-1685
[9]	Cao G,Pisu M,Morbidelli M.A lumped kinetic model for liquid-phase catalytic oxidation of p-xylene to terephthalic acid[J].Chem. Eng. Sci.,1994,49:5775-5788
[10]	Sun W,Pan Y,Zhao L,Zhou X.Simplified free-radical reaction kinetics for p-xylene oxidation to terephthalic acid[J].Chem. Eng. Tech.,2008,31(10):1402-1409
[11]	Suykens J A K,van Gestel T,de Brabanter J,de Moor B,Vandewalle J.Least Squares Support Vector Machines[M].Singapore:World Scientific,2002
[12]	Farmer J,Packard N,Perelson A.The immune system,adaptation and machine learning[J].Physica D,1986,22:187-204
[13]	Qian F,Tao L,Sun W,Du W.Development of a free radical kinetic model for industrial oxidation of p-xylene based on artificial network and adaptive immune genetic algorithm[J].Ind. Eng. Chem. Res.,2011,51(8):3229-3237

基于粗差判别的参数优化自适应加权最小二乘支持向量机在PX氧化过程参数估计中的应用

Adaptive weighted least square support vector machine regression with gross error detection and its application to estimate kinetic parameters for industrial oxidation of p-xylene

PDF (PC)

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献 13

相关文章 2

编辑推荐

Metrics

本文评价

[1]	朱建勇, 张旭乾, 杨辉, 陆荣秀. 单光照条件变化的镨/钕元素组分含量软测量[J]. 化工学报, 2019, 70(2): 780-788.
[2]	林可鸿, 陈德钊. 混合免疫算法及在相平衡计算中的应用 [J]. 化工学报, 2007, 58(6): 1348-1352.