丁苯橡胶聚合转化率软测量方法研究
|
摘要
丁苯橡胶是一种人工合成橡胶，因其良好的性能可在许多方面替代天然橡胶，已被广泛应用于人们日常生活及军工领域。丁苯橡胶生产的难点在于对聚合转化率的时实监控。目前，国内丁苯橡胶生产企业多采用实验室人工方法对其分析控制，该方法因存有严重的时滞问题，不仅控制效果不佳还会带来人力，物力等资源的浪费。软测量技术的发展为上述问题的解决提供了一条良好途径。
然而，目前采用该技术对聚合转化率进行研究的文章还不多见。就丁苯橡胶生产而言，虽工艺复杂，但工作点相对稳定，辅助变量间存有较强的非线性，企业对聚合转化率预测也有较高精度要求。因此，针对该具体对象，本文以核函数思想为基础，提出了3种丁苯橡胶聚合转化率预测的软测量建模方法。
其一是基于KPCA–LSSVM的丁苯橡胶聚合转化率软测量方法。考虑实际工况复杂性和企业对预测精度的要求，首先采用具有较强非线性特征提取能力的核主元分析（KPCA）对数据进行前期处理，并将其结果作为具有小样本、良好泛化能力最小二乘支持向量机（LSSVM）的输入，建立了丁苯橡胶聚合转化率软测量模型。
其二是基于多核技术对丁苯橡胶聚合转化率软测量方法。在采用KPCA对数据进行前期处理的基础上，以此为输入建立了丁苯橡胶聚合转化率在线预测的径向基函数（RBF）神经网络模型。考虑RBF存在单一核难以全面精确描述复杂问题的缺陷，将具有时频突出局部表征能力的小波核引入，构造了高斯核及小波核的混合核预以弥补。
其三是基于核函数的PLS丁苯橡胶聚合转化率软测量方法。考虑偏最小二乘（PLS）算法非线性处理能力的不足，核函数的引入可以提高其非线性处理能力，分别建立了单核和混合核函数的丁苯橡胶聚合转化率PLS预测模型。
上述三种软测量建模方法经工业数据仿真研究，结果表明均能满足企业对丁苯橡胶聚合转化率预测指标的要求。KPCA用于复杂数据的前期处理可以为后续模型提供更为精准的数据信息；小波核具有更好的局部表征能力；而基于混合核函数的PLS建模方法，不但可以较好的描述对象的复杂特征，去除噪声，而且同时考虑输入输出间相关性，在对同批数据仿真结果中，基于核函数的PLS的预测精度和效果最好，更适宜建模预测。
As a kind of synthetic rubber, Styrene-butadiene Rubber(SBR) has been widelyused in our daily life and military areas, because its good performances can be insteadof natural rubber in many ways. The real-time monitoring of polymerizationconversion rate is a difficulty for SBR production. At present, laboratory analysismethods are used to control the index in many domestic enterprise. As its serious delayproblem, not only the effects don’t very well but also lots of resources are waste. Thedevelopment of the soft measurement technology provides a good approach for thesolution the problems.
Articles are not much on polymerization conversion rate by the technology. Justthe production of SBR, although the process is complex, its working points are relativestability, there are strong relevance between the secondary variables and the precisionprediction for polymerization conversion rate. According to the object, this paper putforward three kinds of soft sensor modeling method based on nuclear functionthoughts.
First of all, a soft sensor for SBR Polymerization Conversion Rate Based onKPCA-LSSVM Model. Considering the complexity of actual working condition andthe prediction accuracy requirement of enterprise, it used kernel principal componentanalysis (KPCA), which has a strong ability of nonlinear feature extraction, to processthe data firstly, took the result as input of the least squares support vector machines(LSSVM) model, which has these characteristics, such as small sample and goodgeneralization ability, etc, and established the model for SBR polymerization conver-sion rate.
Secondly, these soft-sensing issues studied based on more kernels technology forpolymerization conversion rate of SBR. The KPCA was used to process the data firstly.And the results were taken as input of the radial basis function (RBF) model. Conside-ring the defect that RBF having a single kernel was difficult to describe complexproblems accurately, the Wavelet kernel having a ability of time-frequency and localcharacterization was introduced, and a mixture kernel (MK) was constructed byGaussian and Wavelet kernels to remedy for the defect.
Finally, this paper provides soft-sensing for polymerization conversion rate ofSBR based on kernel function PLS models. considering the complexity of actualworking condition and the disadvantages of partial least squares (PLS) algorithm for its nonlinear processing power, and the kernel function introduced could increase itsnonlinear processing power. Then PLS models with single or mixed kernel functionwere created separately and used to forecast the SBR polymerization conversion rate.
The simulation results showed that the three kinds of models could all meet theenterprise requirements. KPCA used for complex data processing can provide moreaccurate data information for the subsequent model. Wavelet nuclear has a good localcharacterization ability. The PLS modeling method based on hybrid kernel function,not only can better description of the complex object characteristics, remove noises,but also consider the correlation between the input and output at feature extraction. Inthe same group of data simulation results, models based on the kernel function havehighest prediction accuracy and best effects. Show that these kinds of models are moresuitable for the object modeling.
然而，目前采用该技术对聚合转化率进行研究的文章还不多见。就丁苯橡胶生产而言，虽工艺复杂，但工作点相对稳定，辅助变量间存有较强的非线性，企业对聚合转化率预测也有较高精度要求。因此，针对该具体对象，本文以核函数思想为基础，提出了3种丁苯橡胶聚合转化率预测的软测量建模方法。
其一是基于KPCA–LSSVM的丁苯橡胶聚合转化率软测量方法。考虑实际工况复杂性和企业对预测精度的要求，首先采用具有较强非线性特征提取能力的核主元分析（KPCA）对数据进行前期处理，并将其结果作为具有小样本、良好泛化能力最小二乘支持向量机（LSSVM）的输入，建立了丁苯橡胶聚合转化率软测量模型。
其二是基于多核技术对丁苯橡胶聚合转化率软测量方法。在采用KPCA对数据进行前期处理的基础上，以此为输入建立了丁苯橡胶聚合转化率在线预测的径向基函数（RBF）神经网络模型。考虑RBF存在单一核难以全面精确描述复杂问题的缺陷，将具有时频突出局部表征能力的小波核引入，构造了高斯核及小波核的混合核预以弥补。
其三是基于核函数的PLS丁苯橡胶聚合转化率软测量方法。考虑偏最小二乘（PLS）算法非线性处理能力的不足，核函数的引入可以提高其非线性处理能力，分别建立了单核和混合核函数的丁苯橡胶聚合转化率PLS预测模型。
上述三种软测量建模方法经工业数据仿真研究，结果表明均能满足企业对丁苯橡胶聚合转化率预测指标的要求。KPCA用于复杂数据的前期处理可以为后续模型提供更为精准的数据信息；小波核具有更好的局部表征能力；而基于混合核函数的PLS建模方法，不但可以较好的描述对象的复杂特征，去除噪声，而且同时考虑输入输出间相关性，在对同批数据仿真结果中，基于核函数的PLS的预测精度和效果最好，更适宜建模预测。
As a kind of synthetic rubber, Styrene-butadiene Rubber(SBR) has been widelyused in our daily life and military areas, because its good performances can be insteadof natural rubber in many ways. The real-time monitoring of polymerizationconversion rate is a difficulty for SBR production. At present, laboratory analysismethods are used to control the index in many domestic enterprise. As its serious delayproblem, not only the effects don’t very well but also lots of resources are waste. Thedevelopment of the soft measurement technology provides a good approach for thesolution the problems.
Articles are not much on polymerization conversion rate by the technology. Justthe production of SBR, although the process is complex, its working points are relativestability, there are strong relevance between the secondary variables and the precisionprediction for polymerization conversion rate. According to the object, this paper putforward three kinds of soft sensor modeling method based on nuclear functionthoughts.
First of all, a soft sensor for SBR Polymerization Conversion Rate Based onKPCA-LSSVM Model. Considering the complexity of actual working condition andthe prediction accuracy requirement of enterprise, it used kernel principal componentanalysis (KPCA), which has a strong ability of nonlinear feature extraction, to processthe data firstly, took the result as input of the least squares support vector machines(LSSVM) model, which has these characteristics, such as small sample and goodgeneralization ability, etc, and established the model for SBR polymerization conver-sion rate.
Secondly, these soft-sensing issues studied based on more kernels technology forpolymerization conversion rate of SBR. The KPCA was used to process the data firstly.And the results were taken as input of the radial basis function (RBF) model. Conside-ring the defect that RBF having a single kernel was difficult to describe complexproblems accurately, the Wavelet kernel having a ability of time-frequency and localcharacterization was introduced, and a mixture kernel (MK) was constructed byGaussian and Wavelet kernels to remedy for the defect.
Finally, this paper provides soft-sensing for polymerization conversion rate ofSBR based on kernel function PLS models. considering the complexity of actualworking condition and the disadvantages of partial least squares (PLS) algorithm for its nonlinear processing power, and the kernel function introduced could increase itsnonlinear processing power. Then PLS models with single or mixed kernel functionwere created separately and used to forecast the SBR polymerization conversion rate.
The simulation results showed that the three kinds of models could all meet theenterprise requirements. KPCA used for complex data processing can provide moreaccurate data information for the subsequent model. Wavelet nuclear has a good localcharacterization ability. The PLS modeling method based on hybrid kernel function,not only can better description of the complex object characteristics, remove noises,but also consider the correlation between the input and output at feature extraction. Inthe same group of data simulation results, models based on the kernel function havehighest prediction accuracy and best effects. Show that these kinds of models are moresuitable for the object modeling.
引文
[1]俞金寿.软测量技术及其应用[J].自动化仪表,2008,29(1):1-7
[2] MCAVOY T J. Contemplative Stance for Chemical Process Control[J].Auto-matica,1992,28(2):441-442
[3]李炜,徐鸿飞,倪源.基于核函数的PLS丁苯橡胶聚合转化率软测量[J].传感器与微系统,2012,31(3):143-146
[4] LI Lijuan,SU Hongye, CHU Jian. Modeling of Isomerization of C8Aromatics byOnline Least Squares Support Vector Machine [J].Chinese Journal of ChemicalEngineering,2009,17(3):437-444
[5] Luigi Fortuna,Salvatore Graziani,Maria Gabriella Xibilia. Comparison of Soft-Sensor Design Methods for Industrial Plants Using Small Data Sets[J]. IEEETransactions On Instrumentation And Measurement,2009,58(8):2444-2451
[6]刘毅,王海清,李平.用于发酵过程在线建模的自适应局部最小二乘支持向量机回归方法[J].化工学报,2008,59(8):2052-2057
[7]徐海霞.多模型软测量建模方法研究及其在生物发酵过程中的应用[D].江苏大学工学硕士论文,2009
[8]唐春霞,阳春华,桂卫华,等.基于KPCA_LSSVM的硅锰合金熔炼过程炉渣碱度预测研究[J].仪器仪表学报,2010,31(3):689-693
[9]张炎欣,张航,王伟.基于即时学习策略的火电厂烟气含氧量软测量[J].装备制造技术,2010,4:47-51
[10]黄道平,刘乙奇,李艳.软测量在污水处理过程中的研究与应用[J].化工学报,2011,62(1):1-9
[11]俞金寿,刘爱伦,张克进.软测量技术及其在石油化工中的应用[M].北京:化工出版社,2000
[12]李海青.软测量技术原理及应用[M].北京:化工出版社,2000
[13] Cleveland W. S.,Devlin S. J. Locally Weighted Regression Model[J].Chemo-metrics of Intelligent Laboratory Systems,1988,3(1):301-313
[14] Sungyong Park,Han Chenghun. A Nonlinear Soft Sensor Based on MultivariateSmoothing Procedure for Quality Estimation in Distillation Columns[J].Computers and Chemical Eng,2000,24(2-7):871-877
[15]傅永峰.软测量建模方法研究及其工业应用[D].浙江大学博士学位论文,2007.
[16]郭彦,邱彤,何小荣,等.气体分馏装置软测量模型在线修正策略的研究[J].计算机与应用化学,2004,21(3):407-410
[17]朱波.氧化铝生产苛性比值软测量神经网络模型的在线修正方法研究[D].中南大学,2005(5)
[18]徐欧官,陈祥华.移动窗递推PLS软测量建模及其工业应用[J].高校化学工程学报,2009,23(6):1044-1050
[19]潘立登,李大字,马俊英.软测量技术原理与应用[M].北京:中国电力出版社,2009
[20]张彬,杨为民,钱锋,等.乙苯催化脱氢制苯乙烯生产过程的机理建模和优化[J].控制理论与应用,2011,27(7):903-908
[21]李琦,邵诚.气体分馏装置精馏过程的动态建模与仿真[J].系统仿真学报,2009,21(14):4492-4496
[22]侯卫锋.催化重整流程模拟与优化技术及其应用研究[D].浙江大学博士学位论文,2006
[23]陈彦桥,郭一,刘建民,等.一种改进烟气含氧量软测量建模方法与仿真[J].动力工程学报,2011.31(1):12-16
[24]席峰,刘中.基于状态预测自适应一致滤波器的分布式估计融合算法[J].信息与控制,2010,39(1):59-64
[25]孙文爽,陈兰祥.多元统计分析[M].北京:高等教育出版社,1994
[26]张勇,王介生.基于PCA-RBF神经网络的浮选过程软测量建模[J].南京航空航天大学学报,2006,38,Suppl.:116-119
[27]杨尔辅,周强,胡益锋,等.基于PCA-RBF神经网络的工业裂解炉收率在线预测软测量方法[J].系统仿真学报,2001,13(S1):194-197
[28]常玉清,王小刚,王福利. PCA-DRBFN模型在精馏塔精苯干点估计中的应用[J].东北大学学报(自然科学版),2004,25(2):103-105
[29] Shijian Zhao,Yongmao Xu. Multivariate statistical process monitoring usingrobust nonlinear principal component analysis[J].Tsinghua Science&Techno-logy,2005,10(5):582-586
[30] William W.Hsieh. Nonlinear principal component analysis of noisy data[J].Neural Networks,2007,20(4):434-443
[31] Marilie Linting, Jacqueline J.Meulman, Patrick J.F.Groenen, et al. Nonlinearprincipal components analysis: Introduction and application[J].PsychologicalMethods,2007,12(3):336-358
[32] Mei Liu,Daoping Huang,Zonghai Sun,Zhengshi Chen. Combining KPCA withLSSVM for the Mooney-viscosity Forecasting[C].Second International Confe-rence on Genetic and Evolutionary Computing
[33] Karim Salahshoor,Mojtaba Kordestani,Majid.S.Khoshro. Design of Online SoftSensors based on Combined Adaptive PCA and RBF Neural Networks[C].Computational Intelligence in Control and Automation,2009(CICA2009):89-94
[34] Fengqi Si,Carlos E. Romero,Zheng Yao, et al. Inferential sensor for on-linemonitoring of ammonium bisulfate formation temperature in coal-fired powerplants[J].Fuel Processing Technology,2009,90:56-66
[35]黄国宏,邵惠鹤.核主元分析及其在人脸识别中的应用[J].计算机工程,2004,30(13):13-14
[36]林材宽,乔建忠,王国仁,等.基于核方法的非线性时间序列预测建模[J].计算机工程,2007,33(17):23-25
[37]李岳,温熙森,吕克洪.基于核主成分分析的铁谱磨粒特征提取方法研究[J].国防科技大学学报,2007,29(2):113-116
[38] Scholkopf B,Smola A. Nonlinear Component Analysis as a Kernel EigenvalueProblem[J]. Neural Computation,1998,10(5):1299
[39] Zhe Li,Xuemin Tian. Study of Soft Sensor Modeling Method Based onKPCA-SVM[C]. IEEE Sixth World Congress on Intelligent Control and Automa-tion,2006:4876-4880
[40] Chong Zhang,Chongxun Zheng,Xiaolin Yu,et al. Estimating VDT Mental FatigueUsing Multichannel Linear Descriptors and KPCA-HMM[J]. EURASIP Journalon Advances in Signal Processing,2008,2008(19):1687-6172
[41]胡彦蓉,吴冲,刘洪久.基于KPCA-SVM的公路客运量预测研究[J].技术经济与管理研究,2012,(1):8-12
[42]俞金寿.工业过程先进控制技术[M].上海：华东理工大学出版社,2008,8:83
[43] Frank B.E.,Friedman J.H.. A statistical view of sonic chemo metrics regressiontools[J].Technometrics,1993,35(2):109-135
[44] Lindgren F.,Geladi P.,Wold S.. The kernel algorithm for PLS[J].Journal ofChemometrics,1993,7:45-59
[45] Ivakhnenko A G. Polynomial theory of complex systems[J].IEEE Transaction onSystems, Man, and Cybernetics,1971,1(4):364-378
[46] Dong Hyuk Lim,Sung Han Lee,Man Gyun Na. Smart Soft-Sensing for the Feed-water Flowrate at PWRs Using a GMDH Algorithm[J]. IEEE Transactions onNuclear Science,2010,57(1):340-347
[47]王学武,顾幸生,刘卓倩,等.基于PSO-BP神经网络的催化裂化C3含量软测量模型[J].系统仿真学报,2009,21(4):973-980.
[48]樊绍胜,王耀南.基于特征选取与多RBF神经网络的冷凝器污垢软测量[J].仪器仪表学报,2008,29(4):723-728
[49] Huang G B,Zhu Q Y,Siew C K. Extreme learning machine: a new learningscheme of feed forward neural networks[C].Proceedings of International JointConference on Neural Networks (IJCNN2004).Piscataway: Institute of Electricaland Electronics Engineers Inc,2004:985-990
[50] Huang G B,Siew C K. Extreme Learning Machine with Randomly Assigned RBFKernels[J].International Journal of Technology,2005,11(1):16-24
[51] Nan-Ying Liang,Guang-Bin Huang,P. Saratchandran. A Fast and Accurate On-line Sequential Learning Algorithm for Feed forward Networks[J].IEEE Trans-actions on Neural Networks,2006,17(6):1411-1423
[52]吴德会.基于LS-SVM的特征提取及在凝点软测量中的应用[J].系统仿真学报,2008,20(4):917-920
[53]孙玉坤,王博,黄永红,等.基于聚类动态LS-SVM的L-赖氨酸发酵过程软测量方法[J].仪器仪表学报,2010,31(2):404-409
[54]朱红求,阳春华,桂卫华.基于KPLS和LS-SVM的过程参数预测及其应用[J].控制工程,2010,17(2):219-218
[55] Emre Comak,Kemal Polat,Salih Guenes, et al. A new medical decision makingsystem: Least square support vector machine(LSSVM) with Fuzzy WeightingPre-processing[J]. Expert Systems with Applications,2007,32(26):409-414
[56] Lu Chunyan,Wei Li,Weirong Liu. Application of Soft Sensor based on LS-SVMon Estimation of Alumina Powder Flow[C].IEEE International Conference onMeasuring Technology and Mechatronics Automation,2009,1:281-284
[57] Hong-Qiao Wang,Fu-Chun Sun,Yan-Ning Cai,et al. An unbiased LSSVM modelfor classifycation and regression[J].Soft Computing-A Fusion of Foundations,Methodologies and Applications,2010,14(2):171-180
[58]李明迅,孟相如,袁荣坤,等.融合提升小波降噪和LSSVM的网络流量在线预测[J].计算机应用,2012,32(2):340-342
[59]徐晔,杜文莉,钱锋.基于核主元分析和最小二乘支持向量机的软测量建模[J].系统仿真学报,2007,19(17):3873-3875
[60]汤健,赵立杰,岳恒,等.基于多源数据特征融合的球磨机负荷软测量[J].浙江大学学报(工学版),2010,44(7):1406-1413
[61]王强,田学民.基于KPCA-LSSVM的软测量建模方法[J].化工学报,2011,62(10):2813-2817
[62] Holland J H. Adaptation in Natual and Artificial System[M].Cambrigde:MITPress.1975
[63] Kitano H. Designing neural network using genetic algorithm with graph gener-ation system. Complex System[J].1990,4(4):461-476
[64] Vaario J,Ohsuga S. Adaptive Neural Architectures Through Growth Control [M].Intelligent Engineering Systems Through Artifical Neural Networks. New York:ASM Press,1991:11-16
[65]李惠峰,王健.基于遗传算法的高速飞行器模糊飞控一体化[J].北京航空航天大学学报,2009,35(1):44-47
[66]范锐锐,王明海.遗传算法解算弹道导弹再入弹道[J].弹箭与制导学报,2010,30(5):156-158
[67] A Colorni Ant System for Job-shop Scheduling[J].JORBEL,1994,34(1):39-53
[68] M Dorigo,L M Gambardella. Ant Colonies for the Traveling Salesman Problem[J]. BioSystems,1997,43(2):73-81
[69] M Dorigo,G Di Caro. Ant Algorithms for Discrete Opetimization[J]. ArtificialLife,1999,5(3):137-172
[70]温如春,汤青波,杨国亮.基于改进蚁群算法的移动机器人路径规划[J].兵工自动化,2010,29(8):69-70
[71] Kennedy J,Eberhart R C. Particle swarm optimization[C].IEEE InternationalConference on Neural Networks,1995:1942-1948
[72] Clerc M. Particle Swarm Optimization[M]. London: ISTE Publishing Company,2006
[73] Poli R. Analysis of the publications on the applications of particle swarm opti-mazation[J]. Journal of Artificial Evolution and Applications,2008,8(2):4
[74]周云龙,何小斌.基于PSO-LSSVR的环状流截面含气率软测量方法[J].化工自动化及仪表,2010,37(1):34-38
[75]史志富,刘海燕.基于PSO与LS-SVM的飞机总体设计的综合论证[J].计算机仿真,2010,27(5):62-65
[76] Zhengrong Xiang,Songqing Liu. Component Content Soft-sensor in Rare-EarthExtraction Based on PSO and LS-SVM[C]. IEEE Fourth International Confernceon Natural Computation.2008,6:392-395
[77]吴洲,田鹏,潘丰.基于APSO-LSSVM的软测量建模研究[J].自动化技术与应用,2009,28(1):6-9
[78]阳春华,钱晓山.基于变异CPSO算法LSSVM蒸发过程软测量[J].化工进展,2010,29(3):440-448
[79] Shi Y,Eberhart R. Empirical study of particle swarm optimization[C].Proceed-ings of the1999Congress on Evolutionary Computation,1999:1945-1950
[80] Zheng Y L,Ma L H,Zhang L Y,et al. On the convergence analysis and parameterselection in particle swarm optimization[C].Proceedings of the Second Inter-national Conference on Machine Learning and Cybernetics, Xi’an,2003:1802-1807
[81] Chatterjee A,Siarry P. Nonlinear inertia weight variation for dynamic adaptationin particle swarm optimization[J]. Computers and Operations Research,2006,33(3):859-871
[82] Karim H. Youssef,Hasan A. Yousef,Omar A. Sebakhy,et al. Adaptive fuzzyAPSO based inverse tracking-controller with an application to DC motors[J].Expert Systems with Applications,2009,36(2):3454-3458
[83]耿立艳,梁毅刚.基于灰色自适应粒子群LSSVM的铁路货运量预测[J].西南交通大学学报,2012,47(1):144-150
[84] ZADEH L A. Roles of soft computing and fuzzy logic in the conception, designand deployment of information/intelligent systems[J].Computational Intelligen-ce:Soft Computing and Fuzzy-Neuro Integration with Applications,1998:1-9
[85] YAN S,MASAHARU M. A new approach of neurofuzzy learning algorithm fortuning fuzzy rules[J].Fuzzy sets and systems,2000,112:99-116
[86] MAUR ICIO F,FERNANDO G. Design of fuzzy system using neuro-fuzzynetworks[J].IEEE Trans. Neural Networks,1999,10(4):815-827
[87] DONG F,J IANG Z X,QIAO X T,et al. Application of electrical resistancetomography to two-phase pipe flow parameters measurement[J].FlowMeas.Instrum,2003,14(1):183-192
[88]刘美,黄道平,李桢.丁苯橡胶门尼粘度软测量[J].自动化仪表,2008,29(12):33-34
[89]刘美,黄道平,孙宗海.基于PCA和LS-SVM的丁苯橡胶的门尼粘度预测[J].计算机与应用化学,2008,15
[90]李桢,肖迳,刘美.基于NNPLS的丁苯橡胶门尼粘度预测[J].微计算机信息,2008,24(8):247-249
[91]李炜,章寅,倪源.基于集成修剪的丁苯橡胶聚合转化率软测量[J].仪器仪表学报,2011,32(1):212-217
[92]徐文艳.基于软测量技术的丁苯橡胶聚合率在线监测[J].化工自动化及仪表,2010,37(10):124-126
[93] Scholkopf B,Sung K,Burges C,Girosi F,et a1. Comparing support vector machineswith Gaussian kernels to radial basis function classifier[J].IEEE Transactions onSignal Processing,1997,45(11):2758-2765
[94] Keerthi S S K,Lin C J. Asymptotic behaviors of support vector machines withGaussian kernel[J].Neural Computation,2003,15(7):l667-1689
[2] MCAVOY T J. Contemplative Stance for Chemical Process Control[J].Auto-matica,1992,28(2):441-442
[3]李炜,徐鸿飞,倪源.基于核函数的PLS丁苯橡胶聚合转化率软测量[J].传感器与微系统,2012,31(3):143-146
[4] LI Lijuan,SU Hongye, CHU Jian. Modeling of Isomerization of C8Aromatics byOnline Least Squares Support Vector Machine [J].Chinese Journal of ChemicalEngineering,2009,17(3):437-444
[5] Luigi Fortuna,Salvatore Graziani,Maria Gabriella Xibilia. Comparison of Soft-Sensor Design Methods for Industrial Plants Using Small Data Sets[J]. IEEETransactions On Instrumentation And Measurement,2009,58(8):2444-2451
[6]刘毅,王海清,李平.用于发酵过程在线建模的自适应局部最小二乘支持向量机回归方法[J].化工学报,2008,59(8):2052-2057
[7]徐海霞.多模型软测量建模方法研究及其在生物发酵过程中的应用[D].江苏大学工学硕士论文,2009
[8]唐春霞,阳春华,桂卫华,等.基于KPCA_LSSVM的硅锰合金熔炼过程炉渣碱度预测研究[J].仪器仪表学报,2010,31(3):689-693
[9]张炎欣,张航,王伟.基于即时学习策略的火电厂烟气含氧量软测量[J].装备制造技术,2010,4:47-51
[10]黄道平,刘乙奇,李艳.软测量在污水处理过程中的研究与应用[J].化工学报,2011,62(1):1-9
[11]俞金寿,刘爱伦,张克进.软测量技术及其在石油化工中的应用[M].北京:化工出版社,2000
[12]李海青.软测量技术原理及应用[M].北京:化工出版社,2000
[13] Cleveland W. S.,Devlin S. J. Locally Weighted Regression Model[J].Chemo-metrics of Intelligent Laboratory Systems,1988,3(1):301-313
[14] Sungyong Park,Han Chenghun. A Nonlinear Soft Sensor Based on MultivariateSmoothing Procedure for Quality Estimation in Distillation Columns[J].Computers and Chemical Eng,2000,24(2-7):871-877
[15]傅永峰.软测量建模方法研究及其工业应用[D].浙江大学博士学位论文,2007.
[16]郭彦,邱彤,何小荣,等.气体分馏装置软测量模型在线修正策略的研究[J].计算机与应用化学,2004,21(3):407-410
[17]朱波.氧化铝生产苛性比值软测量神经网络模型的在线修正方法研究[D].中南大学,2005(5)
[18]徐欧官,陈祥华.移动窗递推PLS软测量建模及其工业应用[J].高校化学工程学报,2009,23(6):1044-1050
[19]潘立登,李大字,马俊英.软测量技术原理与应用[M].北京:中国电力出版社,2009
[20]张彬,杨为民,钱锋,等.乙苯催化脱氢制苯乙烯生产过程的机理建模和优化[J].控制理论与应用,2011,27(7):903-908
[21]李琦,邵诚.气体分馏装置精馏过程的动态建模与仿真[J].系统仿真学报,2009,21(14):4492-4496
[22]侯卫锋.催化重整流程模拟与优化技术及其应用研究[D].浙江大学博士学位论文,2006
[23]陈彦桥,郭一,刘建民,等.一种改进烟气含氧量软测量建模方法与仿真[J].动力工程学报,2011.31(1):12-16
[24]席峰,刘中.基于状态预测自适应一致滤波器的分布式估计融合算法[J].信息与控制,2010,39(1):59-64
[25]孙文爽,陈兰祥.多元统计分析[M].北京:高等教育出版社,1994
[26]张勇,王介生.基于PCA-RBF神经网络的浮选过程软测量建模[J].南京航空航天大学学报,2006,38,Suppl.:116-119
[27]杨尔辅,周强,胡益锋,等.基于PCA-RBF神经网络的工业裂解炉收率在线预测软测量方法[J].系统仿真学报,2001,13(S1):194-197
[28]常玉清,王小刚,王福利. PCA-DRBFN模型在精馏塔精苯干点估计中的应用[J].东北大学学报(自然科学版),2004,25(2):103-105
[29] Shijian Zhao,Yongmao Xu. Multivariate statistical process monitoring usingrobust nonlinear principal component analysis[J].Tsinghua Science&Techno-logy,2005,10(5):582-586
[30] William W.Hsieh. Nonlinear principal component analysis of noisy data[J].Neural Networks,2007,20(4):434-443
[31] Marilie Linting, Jacqueline J.Meulman, Patrick J.F.Groenen, et al. Nonlinearprincipal components analysis: Introduction and application[J].PsychologicalMethods,2007,12(3):336-358
[32] Mei Liu,Daoping Huang,Zonghai Sun,Zhengshi Chen. Combining KPCA withLSSVM for the Mooney-viscosity Forecasting[C].Second International Confe-rence on Genetic and Evolutionary Computing
[33] Karim Salahshoor,Mojtaba Kordestani,Majid.S.Khoshro. Design of Online SoftSensors based on Combined Adaptive PCA and RBF Neural Networks[C].Computational Intelligence in Control and Automation,2009(CICA2009):89-94
[34] Fengqi Si,Carlos E. Romero,Zheng Yao, et al. Inferential sensor for on-linemonitoring of ammonium bisulfate formation temperature in coal-fired powerplants[J].Fuel Processing Technology,2009,90:56-66
[35]黄国宏,邵惠鹤.核主元分析及其在人脸识别中的应用[J].计算机工程,2004,30(13):13-14
[36]林材宽,乔建忠,王国仁,等.基于核方法的非线性时间序列预测建模[J].计算机工程,2007,33(17):23-25
[37]李岳,温熙森,吕克洪.基于核主成分分析的铁谱磨粒特征提取方法研究[J].国防科技大学学报,2007,29(2):113-116
[38] Scholkopf B,Smola A. Nonlinear Component Analysis as a Kernel EigenvalueProblem[J]. Neural Computation,1998,10(5):1299
[39] Zhe Li,Xuemin Tian. Study of Soft Sensor Modeling Method Based onKPCA-SVM[C]. IEEE Sixth World Congress on Intelligent Control and Automa-tion,2006:4876-4880
[40] Chong Zhang,Chongxun Zheng,Xiaolin Yu,et al. Estimating VDT Mental FatigueUsing Multichannel Linear Descriptors and KPCA-HMM[J]. EURASIP Journalon Advances in Signal Processing,2008,2008(19):1687-6172
[41]胡彦蓉,吴冲,刘洪久.基于KPCA-SVM的公路客运量预测研究[J].技术经济与管理研究,2012,(1):8-12
[42]俞金寿.工业过程先进控制技术[M].上海：华东理工大学出版社,2008,8:83
[43] Frank B.E.,Friedman J.H.. A statistical view of sonic chemo metrics regressiontools[J].Technometrics,1993,35(2):109-135
[44] Lindgren F.,Geladi P.,Wold S.. The kernel algorithm for PLS[J].Journal ofChemometrics,1993,7:45-59
[45] Ivakhnenko A G. Polynomial theory of complex systems[J].IEEE Transaction onSystems, Man, and Cybernetics,1971,1(4):364-378
[46] Dong Hyuk Lim,Sung Han Lee,Man Gyun Na. Smart Soft-Sensing for the Feed-water Flowrate at PWRs Using a GMDH Algorithm[J]. IEEE Transactions onNuclear Science,2010,57(1):340-347
[47]王学武,顾幸生,刘卓倩,等.基于PSO-BP神经网络的催化裂化C3含量软测量模型[J].系统仿真学报,2009,21(4):973-980.
[48]樊绍胜,王耀南.基于特征选取与多RBF神经网络的冷凝器污垢软测量[J].仪器仪表学报,2008,29(4):723-728
[49] Huang G B,Zhu Q Y,Siew C K. Extreme learning machine: a new learningscheme of feed forward neural networks[C].Proceedings of International JointConference on Neural Networks (IJCNN2004).Piscataway: Institute of Electricaland Electronics Engineers Inc,2004:985-990
[50] Huang G B,Siew C K. Extreme Learning Machine with Randomly Assigned RBFKernels[J].International Journal of Technology,2005,11(1):16-24
[51] Nan-Ying Liang,Guang-Bin Huang,P. Saratchandran. A Fast and Accurate On-line Sequential Learning Algorithm for Feed forward Networks[J].IEEE Trans-actions on Neural Networks,2006,17(6):1411-1423
[52]吴德会.基于LS-SVM的特征提取及在凝点软测量中的应用[J].系统仿真学报,2008,20(4):917-920
[53]孙玉坤,王博,黄永红,等.基于聚类动态LS-SVM的L-赖氨酸发酵过程软测量方法[J].仪器仪表学报,2010,31(2):404-409
[54]朱红求,阳春华,桂卫华.基于KPLS和LS-SVM的过程参数预测及其应用[J].控制工程,2010,17(2):219-218
[55] Emre Comak,Kemal Polat,Salih Guenes, et al. A new medical decision makingsystem: Least square support vector machine(LSSVM) with Fuzzy WeightingPre-processing[J]. Expert Systems with Applications,2007,32(26):409-414
[56] Lu Chunyan,Wei Li,Weirong Liu. Application of Soft Sensor based on LS-SVMon Estimation of Alumina Powder Flow[C].IEEE International Conference onMeasuring Technology and Mechatronics Automation,2009,1:281-284
[57] Hong-Qiao Wang,Fu-Chun Sun,Yan-Ning Cai,et al. An unbiased LSSVM modelfor classifycation and regression[J].Soft Computing-A Fusion of Foundations,Methodologies and Applications,2010,14(2):171-180
[58]李明迅,孟相如,袁荣坤,等.融合提升小波降噪和LSSVM的网络流量在线预测[J].计算机应用,2012,32(2):340-342
[59]徐晔,杜文莉,钱锋.基于核主元分析和最小二乘支持向量机的软测量建模[J].系统仿真学报,2007,19(17):3873-3875
[60]汤健,赵立杰,岳恒,等.基于多源数据特征融合的球磨机负荷软测量[J].浙江大学学报(工学版),2010,44(7):1406-1413
[61]王强,田学民.基于KPCA-LSSVM的软测量建模方法[J].化工学报,2011,62(10):2813-2817
[62] Holland J H. Adaptation in Natual and Artificial System[M].Cambrigde:MITPress.1975
[63] Kitano H. Designing neural network using genetic algorithm with graph gener-ation system. Complex System[J].1990,4(4):461-476
[64] Vaario J,Ohsuga S. Adaptive Neural Architectures Through Growth Control [M].Intelligent Engineering Systems Through Artifical Neural Networks. New York:ASM Press,1991:11-16
[65]李惠峰,王健.基于遗传算法的高速飞行器模糊飞控一体化[J].北京航空航天大学学报,2009,35(1):44-47
[66]范锐锐,王明海.遗传算法解算弹道导弹再入弹道[J].弹箭与制导学报,2010,30(5):156-158
[67] A Colorni Ant System for Job-shop Scheduling[J].JORBEL,1994,34(1):39-53
[68] M Dorigo,L M Gambardella. Ant Colonies for the Traveling Salesman Problem[J]. BioSystems,1997,43(2):73-81
[69] M Dorigo,G Di Caro. Ant Algorithms for Discrete Opetimization[J]. ArtificialLife,1999,5(3):137-172
[70]温如春,汤青波,杨国亮.基于改进蚁群算法的移动机器人路径规划[J].兵工自动化,2010,29(8):69-70
[71] Kennedy J,Eberhart R C. Particle swarm optimization[C].IEEE InternationalConference on Neural Networks,1995:1942-1948
[72] Clerc M. Particle Swarm Optimization[M]. London: ISTE Publishing Company,2006
[73] Poli R. Analysis of the publications on the applications of particle swarm opti-mazation[J]. Journal of Artificial Evolution and Applications,2008,8(2):4
[74]周云龙,何小斌.基于PSO-LSSVR的环状流截面含气率软测量方法[J].化工自动化及仪表,2010,37(1):34-38
[75]史志富,刘海燕.基于PSO与LS-SVM的飞机总体设计的综合论证[J].计算机仿真,2010,27(5):62-65
[76] Zhengrong Xiang,Songqing Liu. Component Content Soft-sensor in Rare-EarthExtraction Based on PSO and LS-SVM[C]. IEEE Fourth International Confernceon Natural Computation.2008,6:392-395
[77]吴洲,田鹏,潘丰.基于APSO-LSSVM的软测量建模研究[J].自动化技术与应用,2009,28(1):6-9
[78]阳春华,钱晓山.基于变异CPSO算法LSSVM蒸发过程软测量[J].化工进展,2010,29(3):440-448
[79] Shi Y,Eberhart R. Empirical study of particle swarm optimization[C].Proceed-ings of the1999Congress on Evolutionary Computation,1999:1945-1950
[80] Zheng Y L,Ma L H,Zhang L Y,et al. On the convergence analysis and parameterselection in particle swarm optimization[C].Proceedings of the Second Inter-national Conference on Machine Learning and Cybernetics, Xi’an,2003:1802-1807
[81] Chatterjee A,Siarry P. Nonlinear inertia weight variation for dynamic adaptationin particle swarm optimization[J]. Computers and Operations Research,2006,33(3):859-871
[82] Karim H. Youssef,Hasan A. Yousef,Omar A. Sebakhy,et al. Adaptive fuzzyAPSO based inverse tracking-controller with an application to DC motors[J].Expert Systems with Applications,2009,36(2):3454-3458
[83]耿立艳,梁毅刚.基于灰色自适应粒子群LSSVM的铁路货运量预测[J].西南交通大学学报,2012,47(1):144-150
[84] ZADEH L A. Roles of soft computing and fuzzy logic in the conception, designand deployment of information/intelligent systems[J].Computational Intelligen-ce:Soft Computing and Fuzzy-Neuro Integration with Applications,1998:1-9
[85] YAN S,MASAHARU M. A new approach of neurofuzzy learning algorithm fortuning fuzzy rules[J].Fuzzy sets and systems,2000,112:99-116
[86] MAUR ICIO F,FERNANDO G. Design of fuzzy system using neuro-fuzzynetworks[J].IEEE Trans. Neural Networks,1999,10(4):815-827
[87] DONG F,J IANG Z X,QIAO X T,et al. Application of electrical resistancetomography to two-phase pipe flow parameters measurement[J].FlowMeas.Instrum,2003,14(1):183-192
[88]刘美,黄道平,李桢.丁苯橡胶门尼粘度软测量[J].自动化仪表,2008,29(12):33-34
[89]刘美,黄道平,孙宗海.基于PCA和LS-SVM的丁苯橡胶的门尼粘度预测[J].计算机与应用化学,2008,15
[90]李桢,肖迳,刘美.基于NNPLS的丁苯橡胶门尼粘度预测[J].微计算机信息,2008,24(8):247-249
[91]李炜,章寅,倪源.基于集成修剪的丁苯橡胶聚合转化率软测量[J].仪器仪表学报,2011,32(1):212-217
[92]徐文艳.基于软测量技术的丁苯橡胶聚合率在线监测[J].化工自动化及仪表,2010,37(10):124-126
[93] Scholkopf B,Sung K,Burges C,Girosi F,et a1. Comparing support vector machineswith Gaussian kernels to radial basis function classifier[J].IEEE Transactions onSignal Processing,1997,45(11):2758-2765
[94] Keerthi S S K,Lin C J. Asymptotic behaviors of support vector machines withGaussian kernel[J].Neural Computation,2003,15(7):l667-1689