数据驱动技术及其在聚丙烯生产过程中的应用研究
|
摘要
聚合过程监测与故障诊断、聚合物生产过程调优以及聚合物重要质量指标预测等研究，交叉了化学反应工程和过程系统工程等学科，无论在理论研究还是实际应用方面都是目前急待解决的前沿研究课题。聚合过程具有高度复杂性、耦合性和强非线性等特点，导致聚合过程研究存在较大困难。随着计算机技术和信息化水平的不断提高，聚合物生产企业可以轻易获得大量生产数据，采用数据驱动技术可以从大量生产数据中获取信息并加以利用，将数据驱动技术引入到聚合过程研究中得到了越来越广泛的关注。
针对常规数据驱动方法所存在的不足，提出了组合神经网络建模方法、改进多尺度主元分析方法和改进混沌粒子群优化算法等改进数据驱动方法，并将其应用于聚丙烯生产过程研究中，开展了聚丙烯熔融指数软测量、丙烯聚合过程监测与故障诊断和聚丙烯最优牌号切换等研究工作，取得了以下研究成果：
（1）针对单一神经网络建模存在预测精度较低的缺点，提出了一种组合神经网络建模方法。通过将多个单一神经网络模型合理组合在一起，可显著提高模型的预测精度。由于选择合适的组合权重系数对取得良好模型预测性能是至关重要的，因此提出了两种求取组合权重系数的方法，分别是最小化最大绝对预测误差方法和岭回归分析方法。熔融指数是决定聚丙烯牌号的最重要参数，但由于缺乏合适的在线测量仪表，导致熔融指数测量存在较大时滞，无法有效实现聚丙烯质量控制，因此将组合神经网络应用于聚丙烯熔融指数软测量研究中。研究结果表明，与单一神经网络模型相比，基于组合神经网络的聚丙烯熔融指数软测量模型具有更佳的预测精度；
（2）针对常规多尺度主元分析在数据去噪方面所存在的不足，提出了一种改进多尺度主元分析方法。针对过程数据所具有的随机性、非平稳性及含有大量噪声等特点，首先采用改进小波阈值去噪方法去除过程数据中的大部分高频随机噪声，提高数据的置信度，然后运用小波多尺度分解方法，将每个变量依次分解成逼近系数和细节系数，建立各个尺度矩阵相对应的主元分析模型，最后对小波系数进行重构得到了综合尺度主元分析模型。为了提高聚丙烯生产过程的稳定性，将改进多尺度主元分析方法应用于丙烯聚合过程监测与故障诊断研究中。研究结果表明，与常规多尺度主元分析相比，改进多尺度主元分析减少了丙烯聚合过程的故障误报率和漏报率，大幅度提高了丙烯聚合过程监测与故障诊断的精度；
（3）针对常规粒子群优化算法存在的收敛速度慢、易陷入局部极值等缺点，提出了一种改进混沌粒子群优化算法。通过算法混合，将混沌搜索融入到粒子群优化算法中，建立了早熟收敛判断和处理机制，以帮助种群摆脱局部最优，提高了优化算法的搜索效率和全局性能。将改进混沌粒子群优化算法应用于聚丙烯最优牌号切换研究中。首先建立了聚丙烯最优牌号切换模型，然后采用改进混沌粒子群优化算法求解该最优牌号切换模型。优化结果表明，与常规混沌粒子群优化算法相比，改进混沌粒子群优化算法具有更佳的优化效率和全局性能；
（4）最后，对全文研究内容进行总结与归纳，并结合研究工作所存在的问题，指出了今后值得关注和进一步深入研究的方向。
Process monitoring and fault diagnosis of polymerization process, optimization ofpolymerization process and prediction of polymer important quality index are theinterdisciplinary field with many unsolved and challenging fundamental research topics andpractical applications, relating to chemical reaction engineering and process system engineering.But polymerization process is high complexity, high coupled and high nonlinearity, this lead theresearch for polymerization process very difficult. Because of the rapid improvement ofcomputer techniques, massive data can be obtained by database, so the so-called ‘data-driven’technique how to drawing information from data has been getting more and more attention.
Aiming to the shortage of traditional data-driven techniques, some improved data-driventechniques are proposed, such as stacked neural networks, improved multi-scale principalcomponent analysis and improved chaotic particle swarm optimization. And the improveddata-driven techniques are applied in polypropylene production process in this paper, such asprediction of polypropylene melt index, process monitoring and fault diagnosis of propylenepolymerization process and optimal grade transition of polypropylene production process. Theresults obtained in this work are summarized as follows:
(1) A modeling approach based on stacked neural networks is proposed. Single neuralnetwork model generalization capability can be significantly improved by using stacked neuralnetworks model. Proper determination of the stacking weights is essential for good stackedneural networks model performance, so two methods about determination of appropriate weightsfor combining individual networks are proposed. Melt index is the most important parameter indetermining the polypropylene grade. Since the lack of proper on-line instruments, itsmeasurement interval and delay are both very long. This makes the quality control quite difficult.Application to real industrial data demonstrates that the polypropylene melt index can besuccessfully estimated using stacked neural networks. The results obtained demonstratesignificant improvements in model accuracy, as a result of using stacked neural networks model,compared to using single neural network model.
(2) For enhancing preformance of process monitoring and fault diagnosis, an improvedmulti-scale principal component analysis (MSPCA) is proposed. Considering the non-stationaryand random nature of data in the process industry it contains different noises inevitably. Based onthe characteristics of wavelet analysis, this paper proposed an improved method which combinesmultiple wavelet transform with a new threshold function. The data collected from the industrycondition are processed by means of the improved wavelet threshold denoising method. Usingwavelets, each variable is decomposed into approximations and details at different scales.Contributions from each scale are collected in separate matrices, and a principal componentanalysis model is then constructed to extract correlation at each scale. According to thesimulation of propylene polymerization, and comparing the improved MSPCA with traditionalMSPCA, it shows that the improved MSPCA has enhanced the accuracy of process monitoringand fault diagnosis.
(3) Aiming to improve the performance of standard particle swarm optimization algorithm,an improved chaotic particle swarm optimization algorithm is introduced. Chaotic searching isintegrated into particle swarm optimization algorithm. Judgment and handling mechanism oflocal convergence is developed. It greatly enhances the local searching efficiency and globalsearching performance of algorithm. A model of grade transition for the industrial multi-reactorpolypropylene production process is conducted according to the Spheripol technique. The modelof grade transition is solved by using both improved chaotic particle swarm optimizationalgorithm and traditional chaotic particle swarm optimization algorithm. The results show thatthe proposed improved chaotic particle swarm optimization algorithm is superior to thetraditional chaotic particle swarm optimization algorithm one in the optimization efficiency andglobal performance.
(4) Finally, the research finding is concluded, and pointed out some future research areas.
针对常规数据驱动方法所存在的不足，提出了组合神经网络建模方法、改进多尺度主元分析方法和改进混沌粒子群优化算法等改进数据驱动方法，并将其应用于聚丙烯生产过程研究中，开展了聚丙烯熔融指数软测量、丙烯聚合过程监测与故障诊断和聚丙烯最优牌号切换等研究工作，取得了以下研究成果：
（1）针对单一神经网络建模存在预测精度较低的缺点，提出了一种组合神经网络建模方法。通过将多个单一神经网络模型合理组合在一起，可显著提高模型的预测精度。由于选择合适的组合权重系数对取得良好模型预测性能是至关重要的，因此提出了两种求取组合权重系数的方法，分别是最小化最大绝对预测误差方法和岭回归分析方法。熔融指数是决定聚丙烯牌号的最重要参数，但由于缺乏合适的在线测量仪表，导致熔融指数测量存在较大时滞，无法有效实现聚丙烯质量控制，因此将组合神经网络应用于聚丙烯熔融指数软测量研究中。研究结果表明，与单一神经网络模型相比，基于组合神经网络的聚丙烯熔融指数软测量模型具有更佳的预测精度；
（2）针对常规多尺度主元分析在数据去噪方面所存在的不足，提出了一种改进多尺度主元分析方法。针对过程数据所具有的随机性、非平稳性及含有大量噪声等特点，首先采用改进小波阈值去噪方法去除过程数据中的大部分高频随机噪声，提高数据的置信度，然后运用小波多尺度分解方法，将每个变量依次分解成逼近系数和细节系数，建立各个尺度矩阵相对应的主元分析模型，最后对小波系数进行重构得到了综合尺度主元分析模型。为了提高聚丙烯生产过程的稳定性，将改进多尺度主元分析方法应用于丙烯聚合过程监测与故障诊断研究中。研究结果表明，与常规多尺度主元分析相比，改进多尺度主元分析减少了丙烯聚合过程的故障误报率和漏报率，大幅度提高了丙烯聚合过程监测与故障诊断的精度；
（3）针对常规粒子群优化算法存在的收敛速度慢、易陷入局部极值等缺点，提出了一种改进混沌粒子群优化算法。通过算法混合，将混沌搜索融入到粒子群优化算法中，建立了早熟收敛判断和处理机制，以帮助种群摆脱局部最优，提高了优化算法的搜索效率和全局性能。将改进混沌粒子群优化算法应用于聚丙烯最优牌号切换研究中。首先建立了聚丙烯最优牌号切换模型，然后采用改进混沌粒子群优化算法求解该最优牌号切换模型。优化结果表明，与常规混沌粒子群优化算法相比，改进混沌粒子群优化算法具有更佳的优化效率和全局性能；
（4）最后，对全文研究内容进行总结与归纳，并结合研究工作所存在的问题，指出了今后值得关注和进一步深入研究的方向。
Process monitoring and fault diagnosis of polymerization process, optimization ofpolymerization process and prediction of polymer important quality index are theinterdisciplinary field with many unsolved and challenging fundamental research topics andpractical applications, relating to chemical reaction engineering and process system engineering.But polymerization process is high complexity, high coupled and high nonlinearity, this lead theresearch for polymerization process very difficult. Because of the rapid improvement ofcomputer techniques, massive data can be obtained by database, so the so-called ‘data-driven’technique how to drawing information from data has been getting more and more attention.
Aiming to the shortage of traditional data-driven techniques, some improved data-driventechniques are proposed, such as stacked neural networks, improved multi-scale principalcomponent analysis and improved chaotic particle swarm optimization. And the improveddata-driven techniques are applied in polypropylene production process in this paper, such asprediction of polypropylene melt index, process monitoring and fault diagnosis of propylenepolymerization process and optimal grade transition of polypropylene production process. Theresults obtained in this work are summarized as follows:
(1) A modeling approach based on stacked neural networks is proposed. Single neuralnetwork model generalization capability can be significantly improved by using stacked neuralnetworks model. Proper determination of the stacking weights is essential for good stackedneural networks model performance, so two methods about determination of appropriate weightsfor combining individual networks are proposed. Melt index is the most important parameter indetermining the polypropylene grade. Since the lack of proper on-line instruments, itsmeasurement interval and delay are both very long. This makes the quality control quite difficult.Application to real industrial data demonstrates that the polypropylene melt index can besuccessfully estimated using stacked neural networks. The results obtained demonstratesignificant improvements in model accuracy, as a result of using stacked neural networks model,compared to using single neural network model.
(2) For enhancing preformance of process monitoring and fault diagnosis, an improvedmulti-scale principal component analysis (MSPCA) is proposed. Considering the non-stationaryand random nature of data in the process industry it contains different noises inevitably. Based onthe characteristics of wavelet analysis, this paper proposed an improved method which combinesmultiple wavelet transform with a new threshold function. The data collected from the industrycondition are processed by means of the improved wavelet threshold denoising method. Usingwavelets, each variable is decomposed into approximations and details at different scales.Contributions from each scale are collected in separate matrices, and a principal componentanalysis model is then constructed to extract correlation at each scale. According to thesimulation of propylene polymerization, and comparing the improved MSPCA with traditionalMSPCA, it shows that the improved MSPCA has enhanced the accuracy of process monitoringand fault diagnosis.
(3) Aiming to improve the performance of standard particle swarm optimization algorithm,an improved chaotic particle swarm optimization algorithm is introduced. Chaotic searching isintegrated into particle swarm optimization algorithm. Judgment and handling mechanism oflocal convergence is developed. It greatly enhances the local searching efficiency and globalsearching performance of algorithm. A model of grade transition for the industrial multi-reactorpolypropylene production process is conducted according to the Spheripol technique. The modelof grade transition is solved by using both improved chaotic particle swarm optimizationalgorithm and traditional chaotic particle swarm optimization algorithm. The results show thatthe proposed improved chaotic particle swarm optimization algorithm is superior to thetraditional chaotic particle swarm optimization algorithm one in the optimization efficiency andglobal performance.
(4) Finally, the research finding is concluded, and pointed out some future research areas.
引文
[1]蒋慰孙,蒋敏伟.过程控制21世纪展望[J].世界仪表和自动化.1999,3(4):10-14.
[2]顾海杰,荣冈.数据驱动方法在流程工业中的应用命题综述[J].化工自动化及仪表,2009,36(5):1-6.
[3]古勇,苏宏业,王朝晖等. FCCU软测量综述[J].化工自动化及仪表.1998,增刊:6-10.
[4]徐敏,俞金寿.软测量技术[J],石油化工自动化.1998,6(2):1-3.
[5]荣冈,金晓明,王树青.先进控制技术及应用[J].化工自动化及仪表.1999,26(4):70-72.
[6]孙欣,王金春,何声亮.过程软测量[J].自动化仪表.1995,16(8):1-5.
[7] Riviere J M, Bayart M, Thiriet J M. Intelligent instruments: some modeling approaches [J].Measurement and Control,1996,29(4):179-186.
[8] Van den Bos A, Eykboff P. Model building and parameter estimation as means for intelligentmeasurement [J]. Measurement,1988,6(1):25-32.
[9]蒙建波,朱林章.检测理论的现状及发展展望[J].化工自动化及仪表.1994,31(3):40-41,45.
[10]赵新民,陈海军.模型化测量的发展现状与展望[J].计量学报.1992,13(4):314-319.
[11]李锡钢,魏传,曹先明.基于MSR的催化裂化装置柴油闪点在线测量[J].石油化工自动化,2007(5):51–52.
[12]王华忠,俞金寿.基于混合核函数PCR方法的工业过程软测量建模[J].化工自动化及仪表,2005,32(2):23-25.
[13] Weber R, Brosilow C. The use of secondary measurements to improve control [J]. AIChE J.1972,18(3):614-623.
[14]梁军.工业流化床乙烯气相聚合中的在线测量与模型分析(I)[J].传感技术学报,2003,3:242-248.
[15]梁军.工业流化床乙烯气相聚合中的在线测量与模型分析(II)[J].传感技术学报,2003,3:250-255.
[16]李春富, ZHANG Jie,王桂增.基于PLS模型的自适应间歇过程质量预测[J].清华大学学报(自然科学版),2004,44(10):1360-1363.
[17]成忠,陈德钊.基于极小极大估计器的偏最小二乘方法及应用[J].化学工程,2007,35(9):29-32.
[18] Shaw A.M., Doyle III F.J. Multivariable nonlinear control applications for a high purity distillationcolumn using a recurrent dynamic neuron model [J]. Journal of Process Control.1997,7(4):255-262.
[19] Joseph B and Brosilow C B. Inferential control of processes: Part I Steady state analysis and design[J].AIChE J.1978,15(2):150-159.
[20] Hinchliffe M, Montague G, Willis M, et al. Hybrid Approach to Modeling an Industrial PolyethyleneProcess [J]. AIChE J,2003,49(12):3127-3137.
[21]吴海燕,曹柳林,王晶等.利用离散正交多项式组合神经网络建立聚合物分子量分布灰箱模型[J].化工学报,2009,60(11):2833-2837.
[22]张宇,李柠,黄道.基于多神经网络模型的酯化反应软测量[J].华东理工大学学报(自然科学版),2005,31(2):208-211,226.
[23]陈国初,徐余法,俞金寿.基于粒子群模糊神经网络的丙烯腈收率软测量建模[J].系统仿真学报,2007,19(23):5370-5372.
[24] Keerthi S S, Shevade S K, Bhattacharyya C, et al. A fast iterative nearest point algorithm for supportvector machine classifier design [J]. IEEE Transactions on Neural Networks,2000,11(1):124-136.
[25] Lin Chun-Fu, Wang Sheng-De. Fuzzy support vector machines [J]. IEEE Transactions on NeuralNetworks,2002,13(2):464-471.
[26]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):3242.
[27]马勇,黄德先,金以慧.基于支持向量机的软测量建模方法[J].信息与控制,2004,33(4):120-125.
[28]阎威武,朱宏栋,邵惠鹤.基于最小二乘支持向量机的软测量建模[J].系统仿真学报,2003,15(10):1494-1496.
[29]王旭东,邵惠鹤.基于神经网络的通用软测量技术[J].自动化学报.1998,24(5):702-706.
[30]邵信光,杨慧中.石晨曦. ε不敏感支持向量回归在化工数据建模中的应用[J].东南大学学报(自然科学版),2004,34(增):215-218.
[31]杨希,钱锋,张兵.基于核函数主元分析的SVM建模方法及应用[J].华东理工大学学报(自然科学版),2007,33(2):259-262,270.
[32] Frank P M. Fault diagnosis in dynamic systems using analytical and knowledge-based redundant-asurvey and some new results [J]. Automatica,1990,26(3):459-474.
[33]郭明.基于数据驱动的流程工业性能监控与故障诊断研究[D].杭州:浙江大学,2004.
[34]张杰,阳宪惠.多变量统计过程控制[M].北京:化学工业出版社,2000.
[35] Ku W, Storey R H, Georgakis C. Disturbance detection and isolation by dynamic principal componentanalysis[J]. Chemometrics and Intelligent Laboratory Systems,1995,30(1):179-196.
[36] Li R Y, Rong G. Fault isolation by partial dynamic principal component analysis in dynamic process [J].Chinese Journal of Chemical Engineering,2006,14(4):486-493.
[37] Kampjarvi P, Sourander M, Komulainen T, et al. Fault detection and isolation of an on-line analyzer foran ethylene cracking process [J]. Control Engineering Practice,2008,16(1):1-13.
[38]杨文婧,曹柳林.改进的非线性PCA方法及其在过程监控中的应用[J].北京化工大学学报,2008,35(4):95-99.
[39] Nomikos P, MacGregor J F. Monitoring batch processes using multiway principal component analysis [J].AICHE J,1994,40(8):1361-1375.
[40] Zhao X, Wen X J, Shao H H. Combination method of principal component analysis and support vectormachine for on-line process monitoring and fault diagnosis [J]. Journal of Donghua University (EnglishEdition),2006,23(1):53-58.
[41]赵立杰,王纲,李元.非线性主元分析故障检测和诊断方法及应用[J].信息与控制,2001,30(4):359-364.
[42]李元,谢植,周东华.基于CMPCA的化工生产过程性能检测与故障诊断[J].信息与控制,2004,33:253-256.
[43]李元,王纲,曹锐.基于MPCA与DTW的间歇反应过程的性能监视及故障诊断[J].沈阳化工学院学报,2003,17(4):285-289.
[44] Lee G, Han C, Yoon E S. Multiple-fault diagnosis of the Tennessee Eastman process based on systemdecomposition and dynamic PLS [J]. Industrial and Engineering Chemistry Research,2004,43(25):8037-8048.
[45] Kourti T, Nomikos P, MacGregor J F. Analysis, monitoring and fault diagnosis of batch processes usingmultiblock and multiway PLS [J]. Journal of Process Control,1995,5(4):277-284.
[46]蒋丽英,王树青.基于MPCA-MDPLS的间歇过程的故障诊断[J].化工学报,2005,56(3):482-486.
[47] Li Y F, Wang Z F, Yuan J Q. On-line Fault Detection Using SVM-based Dynamic MPLS for BatchProcesses[J]. Chinese Journal of Chemical Engineering,2006,14(6):754-758.
[48] Comon P. Independent component analysis–a new concept [J]. Signal Processing,1991,36(3):287-314.
[49] Kano M, Tanaka S, Hasebe S. Monitoring independent components for fault detection [J]. AIChE J,2003,49(4):969-976.
[50] Albazzaz H, Wang X Z. Introduction of dynamics to approach for batch process monitoring usingindependent component analysis [J]. Chemical Engineering Communications,2007,194(2):218-233.
[51] Ge Z Q, Song Z H. Process monitoring based on independent Component Analysis-Principal ComponentAnalysis (ICA-PCA) and similarity factors [J]. Industrial and Engineering Chemistry Research,2007,46(7):2054-2063.
[52] Chen G J, Liang J, Qian J X. ICA and its Application to Chemical Process Monitoring and FaultDiagnosis [J]. Journal of Chemical Industry and Engineering (China),2003,54(10):1474-1477.
[53]杨慧中,高岩,张素贞等.独立成分分析和支持向量机混合方法在过程监控中的应用[J].计算机与应用化学,24(3):295-298.
[54]赵旭,阎威武,邵惠鹤.基于核Fisher判别分析方法的非线性统计过程监控与故障诊断[J].化工学报,2007,58(4):951-956.
[55] Jiang L Y, Xie L, Wang S Q. Fault diagnosis for batch processes by improved multi-model Fisherdiscriminant analysis [J]. Chinese Journal of Chemical Engineering,2006,14(3):343-348.
[56] Chiang L H, Kotanchek M E, Kordon A K. Fault diagnosis based on Fisher discriminant analysis andsupport vector machines [J]. Computers and Chemical Engineering,2004,28(8):1389-1401.
[57]何小荣.化工过程优化[M].北京:清华大学出版社,2003.
[58]邓正龙.化工中的优化方法[M].北京:化学工业出版社,1992.
[59] Holland J H. Adaption in Nature and Artificial System[M]. Boston: MIT Press,1991.
[60] Kirkpatrick S, Gelatt J C D, Vecchi M P. Optimization by simulated annealing [J]. Science,1983,220:671-680.
[61]王举,袁希钢,陈中州.用于多产品间歇化工过程排序的模拟退火算法[J].化工学报,2000,51(6):751-756.
[62]王浩平,项曙光.间歇精馏过程模拟优化研究进展[J].计算机仿真,2004,21(2):4-6.
[63]姚兆玲,袁希钢,王举.考虑环境影响的多产品间歇化工过程优化排序[J].化工学报,2002,53(2):167-171.
[64]岳金彩,郑世清,韩方煜.多目标遗传算法及在过程优化综合中的应[J].计算机与应用化学,2006,22(8):748-752.
[65] Nascimento C A O, Giudici R. Neural network based approach for optimization applied to an industrialnylon-6,6polymerisation process [J]. Computer and Chemical Engineering,1988,22(7):595-600.
[66]王举,陈中州,袁希钢.用于间歇化工过程最优设计的遗传算法.计算机与应用化学,1999,16(1):24-28.
[67]汤文生,合烨,陈小安.基于BP神经网络和遗传算法的硫化工艺参数优化[J].橡胶工业,2008,55(2):105-109.
[68]蒲黎明,俞欢军,陈德钊.连续多蚁群算法的构建及其在过程动态优化中的应用[J].高校化学工程学报,2008,22(5):871-876.
[69]郭相坤,王晓静,许德平等.间歇自由基聚合反应器的蚁群优化[J].计算机与应用化学,2009,26(6):732-736.
[70]邢杰,萧德云.混合粒子群优化算法及其应用[J].化工学报,2008,59(7):1707-1710.
[71]莫愿斌,陈德钊.多目标过程系统优化的粒子群算法求解[J].化工学报2008,22(1):94-98.
[72] Kohonen T. Self-Organizing Maps [M]. Berlin: Springer-Verlag,1997,50-75.
[73] Mhasker H N. Neural networks for optimal approximation of smooth and analytic functions[J]. NeuralComputation,1996,32(8):1731-1742.
[74] Poggio T, Girosi F. Regilarization algorithms for learning that are equivalent to multilayer networks[J].Science,1990,1247(32):978-982.
[75]史忠植.神经网络[M].北京:高等教育出版社,2009,48-63.
[76] Haykin S. Adaptive Filter Theory.3rd. Upper Saddle River [M]. NJ: Prentice Hall,1996,56-89.
[77]俞金寿.工业过程先进控制[M].北京:中国石化出版社,2002.
[78]傅庚,唐小我,曾勇.广义递归方差倒数组合预测方法研究[J].电子科技大学学报,1995,24(2):211-217.
[79]陈友华.基于预测有效度的组合预测模型研究[J].预测,1997,16(3):72-73.
[80]王应明,罗英.调和平均组合预测中的参数估计技术[J].系统工程与电子技术,1997,19(10):45-49.
[81]王应明.广义加权几何平均组合预测方法研究[J].厦门大学学报(自然科学版),1998,37(1):6-10.
[82]陈华友.组合预测方法有效性理论及其应用[M].北京:科学出版社,2008.
[83] Wolpert D H. Stacked generalization [J]. Neural Networks,1992,5(2):241-259.
[84] Sridhar D V, Bartlett E B, Seagrave R C. Information theoretic approach for combining neural networkprocess models[J]. Neural Networks,1999,12(6):915-926.
[85] Sridhar D V,Seagrave R C, Bartlett E B. Process modelling using stacked neural networks[J]. AIChE J,1996,42(9):2529-2539.
[86]夏陆岳.聚氯乙烯智能制造技术研究[D].杭州:浙江工业大学,2004.
[87]王新楼.小波去噪方法分析与MATLAB仿真[J].工业控制计算机,2008,21(6):55-60.
[88]杜浩藩,丛爽.基于MATLAB的小波去噪方法研究[J].计算机仿真,2003,20(7):119-221.
[89]杨慧中,钟豪,丁锋.基于多重小波变换的信号去噪及其在软测量中的应用[J].仪器仪表学报,2007,28(7):1245-1449.
[90]王海清,余世明.基于故障诊断性能优化的主元个数选取方法[J].化工学报,2004,55(2):214-219.
[91] Rudolph G. Convergence analysis of canonical genetic algorithms[J]. IEEE Transactions on NeuralNetworks1994,5(1):96-101.
[92] Kennedy J, Eberhart R C. Particle swarm optimization[C]. Proceedings of IEEE InternationalConference on Neural Networks, Perth: IEEE,1995,1942-1948.
[93] Wilson E O. Sociobiology: The new synthesis [M]. Cambridge, MA: Belknap Press.1975.
[94]曾建潮,介婧,崔志华.微粒群算法[M].北京:科学出版社,2004.
[95] Zhou C, Chen T. Chaotic annealing for optimization [J].Physical Review E,1997,55(3):2580-2587.
[96] Choi C, Lee J. Chaotic local search Algorithm [J]. Artificial Life and Robotics,1998,2(1):41-47.
[97]梁瑞鑫,郑德玲.基于区间套混沌搜索的混合优化方法[J].北京科技大学学报,2002,2(1):41-47.
[98]张彤,土宏伟,王子才.变尺度混沌优化方法及其应用[J].控制与决策,1999,14(3):285-288.
[99]王子才,张彤,土宏伟.基于混沌变量的模拟退火优化方法[J].控制与决策,1999,14(4):381-384.
[100]盛煜翔.基于混沌粒子群算法的闪蒸过程优化研究[D].浙江：浙江工业大学，2009.
[101]Victoire T A A, Jeyakumar A E. Hybrid PSO-SQP for economic dispatch with valve-point effect[J].Electric Power Systems Research,2004,71(1):51-59.
[102]骆晨钟,邵惠鹤.用混沌搜索求解非线性约束优化问题[J].系统工程理论与实践,2000,26(8):54-56.
[103]Kiparissides C. Polymerization reactor modeling: a review of recent development and future directions[J]. Chemical Engineering Science,1996,51(10):1637-1659.
[104]Scorah M J, Dhib R, Penlidis A. Modelling of free radical polymerization of styrene and methylmethacrylate by a tetrafunctional initiator [J]. Chemical Engineering Science,2006,61(15):4827-4859.
[105]Dong M K, Piet D I. Modeling of branching density and branching distribution in low-densitypolyethylene polymerization [J]. Chemical Engineering Science,2008,63(8):2035-2046.
[106]Mariano A, Adriana B. Modeling and optimization of a high-pressure ethylene polymerization reactorusing gPROMS [J]. Computers and Chemical Engineering,2008,32(3):396-408.
[107]Yi H S, Kin J H, Han C, et al. Plantwide optimal grade transition for an industrial high-densitypolyethylene plant [J]. Industrial and Engineering Chemistry Research,2003,42(1):91-98.
[108]Flores T A, Biegler L T, Saldivar G E. Optimal grade transitions in the high-impact polystyrenepolymerization process [J]. Industrial and Engineering Chemistry Research,2006,45(18):6175-6189.
[109]Sharmin R, Sundararaj U, Shah Sirish, et al. Inferential sensors for estimation of polymer qualityparameters: Industrial application of a PLS-based soft sensor for a LDPE plant [J]. ChemicalEngineering Science,2006,61(19):6372-6384.
[110]Liu J. On-line soft sensor for polyethylene process with multiple production grades [J]. ControlEngineering Practice,2007,15(7):769-778.
[111]金学兰,袁璞,胡品慧等.丙烯聚合反应动态模拟[J].化工自动化及仪表,2005,32(3):7-10.
[112]邬正义,施健.基于鲁棒最小二乘支持向量机的聚丙烯熔融指数预报[J].华东理工大学学报(自然科学版),2006,32(7):889-892.
[113]Shi J, Liu X G, Sun Y X. Melt index prediction by neural networks based on independent componentanalysis and multi-scale analysis [J]. Neurocomputing,2006,70(1-3):280-287.
[114]Zhang J, Qi B X, Yong M. Inferential estimation of polymer melt index using sequentially trainedbootstrap aggregated neutral networks [J]. Chemical Engineering Technology,2006,29(4):442-448.
[115]McAuley K B, MacGregor J F. On-line inference of polymer polymer properties in an industrialpolyethylene reactor [J]. AIChE J,1991,37(6):825-835.
[116]Rallo R, Ferre-GineJ, Arenas A, et al. Neural virtual sensor for the inferential prediction of productquality from process variables [J]. Computer and Chemical Engineering,2002,26:1735-1754.
[117]Kano M, Hasebe S, Hashimoto I. A new multivariate statistical process monitoring method usingprincipal component analysis [J]. Computers and Chemical Engineering,2001,25(7-8):1103-1113,2001.
[118]Seongkyu Y, John F, MacGregor J F. Fault diagnosis with multivariate statistical models part I: usingsteady state fault signatures [J]. Journal of Process Control,2001,11(4):387-400.
[119]Lee J M, Yoo C K, Choi S W. Nonlinear process monitoring using kernel principal component analysis[J]. Chemical Engineering Science,2004,59(1):223-234.
[120]李玉贵.液相本体法聚丙烯生产及应用[M].北京:中国石化出版社,1992.
[121]McAuley K.B., MacGregor J.F. Optimal grade transitions in a gas-phase polyethylene reactor[J]. AIChEJournal,1992,38:1564-1576.
[122]金学兰,袁璞,胡品慧.丙烯聚合反应动态建模及在牌号切换中的应用[J].系统仿真学报,2006,18(8):2089-2091,2096.
[123]费正顺,胡斌,叶鲁彬等.带路径约束的聚烯烃牌号切换操作优化方法[J].化工学报,2010,61(4):893-900.
[124]王靖岱,陈纪忠,阳永荣.迭代动态规划在树脂牌号切换最优化模型中的应用[J].高校化学工程学报,2000,14(3):264-269.
[125]黄骅.聚丙烯生产过程建模及牌号切换策略研究[D].杭州:浙江工业大学，2009.
[126]MahadevanR,Doyle III F J.Control-relevantsc heduling of polymer grade transitions[J]. AIChE,2002,48(8):1754-1764.
[127]荚亮.聚丙烯连续生产过程的牌号切换研究[D].杭州:浙江大学,2006.
[128]Vassiliadis V S, Sargent R W H, Pantelides C C. Solution of a class of multistage dynamic optimizationproblems (I): problems without path constraints [J]. Industrial and Engineering Chemistry Research,1994,33(9):2111-2122.
[129]Biegler L T, Grossmann I E. Retrospective on optimization [J]. Computers and Chemical Engineering,2004,28(8):1169-1192.
[130]Schlegel M, Marquardt W. Detection and exploitation of the control switching structure in the solutionof dynamic optimization problems[J]. Journal of Process Control,2006,16(3):275-290.
[2]顾海杰,荣冈.数据驱动方法在流程工业中的应用命题综述[J].化工自动化及仪表,2009,36(5):1-6.
[3]古勇,苏宏业,王朝晖等. FCCU软测量综述[J].化工自动化及仪表.1998,增刊:6-10.
[4]徐敏,俞金寿.软测量技术[J],石油化工自动化.1998,6(2):1-3.
[5]荣冈,金晓明,王树青.先进控制技术及应用[J].化工自动化及仪表.1999,26(4):70-72.
[6]孙欣,王金春,何声亮.过程软测量[J].自动化仪表.1995,16(8):1-5.
[7] Riviere J M, Bayart M, Thiriet J M. Intelligent instruments: some modeling approaches [J].Measurement and Control,1996,29(4):179-186.
[8] Van den Bos A, Eykboff P. Model building and parameter estimation as means for intelligentmeasurement [J]. Measurement,1988,6(1):25-32.
[9]蒙建波,朱林章.检测理论的现状及发展展望[J].化工自动化及仪表.1994,31(3):40-41,45.
[10]赵新民,陈海军.模型化测量的发展现状与展望[J].计量学报.1992,13(4):314-319.
[11]李锡钢,魏传,曹先明.基于MSR的催化裂化装置柴油闪点在线测量[J].石油化工自动化,2007(5):51–52.
[12]王华忠,俞金寿.基于混合核函数PCR方法的工业过程软测量建模[J].化工自动化及仪表,2005,32(2):23-25.
[13] Weber R, Brosilow C. The use of secondary measurements to improve control [J]. AIChE J.1972,18(3):614-623.
[14]梁军.工业流化床乙烯气相聚合中的在线测量与模型分析(I)[J].传感技术学报,2003,3:242-248.
[15]梁军.工业流化床乙烯气相聚合中的在线测量与模型分析(II)[J].传感技术学报,2003,3:250-255.
[16]李春富, ZHANG Jie,王桂增.基于PLS模型的自适应间歇过程质量预测[J].清华大学学报(自然科学版),2004,44(10):1360-1363.
[17]成忠,陈德钊.基于极小极大估计器的偏最小二乘方法及应用[J].化学工程,2007,35(9):29-32.
[18] Shaw A.M., Doyle III F.J. Multivariable nonlinear control applications for a high purity distillationcolumn using a recurrent dynamic neuron model [J]. Journal of Process Control.1997,7(4):255-262.
[19] Joseph B and Brosilow C B. Inferential control of processes: Part I Steady state analysis and design[J].AIChE J.1978,15(2):150-159.
[20] Hinchliffe M, Montague G, Willis M, et al. Hybrid Approach to Modeling an Industrial PolyethyleneProcess [J]. AIChE J,2003,49(12):3127-3137.
[21]吴海燕,曹柳林,王晶等.利用离散正交多项式组合神经网络建立聚合物分子量分布灰箱模型[J].化工学报,2009,60(11):2833-2837.
[22]张宇,李柠,黄道.基于多神经网络模型的酯化反应软测量[J].华东理工大学学报(自然科学版),2005,31(2):208-211,226.
[23]陈国初,徐余法,俞金寿.基于粒子群模糊神经网络的丙烯腈收率软测量建模[J].系统仿真学报,2007,19(23):5370-5372.
[24] Keerthi S S, Shevade S K, Bhattacharyya C, et al. A fast iterative nearest point algorithm for supportvector machine classifier design [J]. IEEE Transactions on Neural Networks,2000,11(1):124-136.
[25] Lin Chun-Fu, Wang Sheng-De. Fuzzy support vector machines [J]. IEEE Transactions on NeuralNetworks,2002,13(2):464-471.
[26]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):3242.
[27]马勇,黄德先,金以慧.基于支持向量机的软测量建模方法[J].信息与控制,2004,33(4):120-125.
[28]阎威武,朱宏栋,邵惠鹤.基于最小二乘支持向量机的软测量建模[J].系统仿真学报,2003,15(10):1494-1496.
[29]王旭东,邵惠鹤.基于神经网络的通用软测量技术[J].自动化学报.1998,24(5):702-706.
[30]邵信光,杨慧中.石晨曦. ε不敏感支持向量回归在化工数据建模中的应用[J].东南大学学报(自然科学版),2004,34(增):215-218.
[31]杨希,钱锋,张兵.基于核函数主元分析的SVM建模方法及应用[J].华东理工大学学报(自然科学版),2007,33(2):259-262,270.
[32] Frank P M. Fault diagnosis in dynamic systems using analytical and knowledge-based redundant-asurvey and some new results [J]. Automatica,1990,26(3):459-474.
[33]郭明.基于数据驱动的流程工业性能监控与故障诊断研究[D].杭州:浙江大学,2004.
[34]张杰,阳宪惠.多变量统计过程控制[M].北京:化学工业出版社,2000.
[35] Ku W, Storey R H, Georgakis C. Disturbance detection and isolation by dynamic principal componentanalysis[J]. Chemometrics and Intelligent Laboratory Systems,1995,30(1):179-196.
[36] Li R Y, Rong G. Fault isolation by partial dynamic principal component analysis in dynamic process [J].Chinese Journal of Chemical Engineering,2006,14(4):486-493.
[37] Kampjarvi P, Sourander M, Komulainen T, et al. Fault detection and isolation of an on-line analyzer foran ethylene cracking process [J]. Control Engineering Practice,2008,16(1):1-13.
[38]杨文婧,曹柳林.改进的非线性PCA方法及其在过程监控中的应用[J].北京化工大学学报,2008,35(4):95-99.
[39] Nomikos P, MacGregor J F. Monitoring batch processes using multiway principal component analysis [J].AICHE J,1994,40(8):1361-1375.
[40] Zhao X, Wen X J, Shao H H. Combination method of principal component analysis and support vectormachine for on-line process monitoring and fault diagnosis [J]. Journal of Donghua University (EnglishEdition),2006,23(1):53-58.
[41]赵立杰,王纲,李元.非线性主元分析故障检测和诊断方法及应用[J].信息与控制,2001,30(4):359-364.
[42]李元,谢植,周东华.基于CMPCA的化工生产过程性能检测与故障诊断[J].信息与控制,2004,33:253-256.
[43]李元,王纲,曹锐.基于MPCA与DTW的间歇反应过程的性能监视及故障诊断[J].沈阳化工学院学报,2003,17(4):285-289.
[44] Lee G, Han C, Yoon E S. Multiple-fault diagnosis of the Tennessee Eastman process based on systemdecomposition and dynamic PLS [J]. Industrial and Engineering Chemistry Research,2004,43(25):8037-8048.
[45] Kourti T, Nomikos P, MacGregor J F. Analysis, monitoring and fault diagnosis of batch processes usingmultiblock and multiway PLS [J]. Journal of Process Control,1995,5(4):277-284.
[46]蒋丽英,王树青.基于MPCA-MDPLS的间歇过程的故障诊断[J].化工学报,2005,56(3):482-486.
[47] Li Y F, Wang Z F, Yuan J Q. On-line Fault Detection Using SVM-based Dynamic MPLS for BatchProcesses[J]. Chinese Journal of Chemical Engineering,2006,14(6):754-758.
[48] Comon P. Independent component analysis–a new concept [J]. Signal Processing,1991,36(3):287-314.
[49] Kano M, Tanaka S, Hasebe S. Monitoring independent components for fault detection [J]. AIChE J,2003,49(4):969-976.
[50] Albazzaz H, Wang X Z. Introduction of dynamics to approach for batch process monitoring usingindependent component analysis [J]. Chemical Engineering Communications,2007,194(2):218-233.
[51] Ge Z Q, Song Z H. Process monitoring based on independent Component Analysis-Principal ComponentAnalysis (ICA-PCA) and similarity factors [J]. Industrial and Engineering Chemistry Research,2007,46(7):2054-2063.
[52] Chen G J, Liang J, Qian J X. ICA and its Application to Chemical Process Monitoring and FaultDiagnosis [J]. Journal of Chemical Industry and Engineering (China),2003,54(10):1474-1477.
[53]杨慧中,高岩,张素贞等.独立成分分析和支持向量机混合方法在过程监控中的应用[J].计算机与应用化学,24(3):295-298.
[54]赵旭,阎威武,邵惠鹤.基于核Fisher判别分析方法的非线性统计过程监控与故障诊断[J].化工学报,2007,58(4):951-956.
[55] Jiang L Y, Xie L, Wang S Q. Fault diagnosis for batch processes by improved multi-model Fisherdiscriminant analysis [J]. Chinese Journal of Chemical Engineering,2006,14(3):343-348.
[56] Chiang L H, Kotanchek M E, Kordon A K. Fault diagnosis based on Fisher discriminant analysis andsupport vector machines [J]. Computers and Chemical Engineering,2004,28(8):1389-1401.
[57]何小荣.化工过程优化[M].北京:清华大学出版社,2003.
[58]邓正龙.化工中的优化方法[M].北京:化学工业出版社,1992.
[59] Holland J H. Adaption in Nature and Artificial System[M]. Boston: MIT Press,1991.
[60] Kirkpatrick S, Gelatt J C D, Vecchi M P. Optimization by simulated annealing [J]. Science,1983,220:671-680.
[61]王举,袁希钢,陈中州.用于多产品间歇化工过程排序的模拟退火算法[J].化工学报,2000,51(6):751-756.
[62]王浩平,项曙光.间歇精馏过程模拟优化研究进展[J].计算机仿真,2004,21(2):4-6.
[63]姚兆玲,袁希钢,王举.考虑环境影响的多产品间歇化工过程优化排序[J].化工学报,2002,53(2):167-171.
[64]岳金彩,郑世清,韩方煜.多目标遗传算法及在过程优化综合中的应[J].计算机与应用化学,2006,22(8):748-752.
[65] Nascimento C A O, Giudici R. Neural network based approach for optimization applied to an industrialnylon-6,6polymerisation process [J]. Computer and Chemical Engineering,1988,22(7):595-600.
[66]王举,陈中州,袁希钢.用于间歇化工过程最优设计的遗传算法.计算机与应用化学,1999,16(1):24-28.
[67]汤文生,合烨,陈小安.基于BP神经网络和遗传算法的硫化工艺参数优化[J].橡胶工业,2008,55(2):105-109.
[68]蒲黎明,俞欢军,陈德钊.连续多蚁群算法的构建及其在过程动态优化中的应用[J].高校化学工程学报,2008,22(5):871-876.
[69]郭相坤,王晓静,许德平等.间歇自由基聚合反应器的蚁群优化[J].计算机与应用化学,2009,26(6):732-736.
[70]邢杰,萧德云.混合粒子群优化算法及其应用[J].化工学报,2008,59(7):1707-1710.
[71]莫愿斌,陈德钊.多目标过程系统优化的粒子群算法求解[J].化工学报2008,22(1):94-98.
[72] Kohonen T. Self-Organizing Maps [M]. Berlin: Springer-Verlag,1997,50-75.
[73] Mhasker H N. Neural networks for optimal approximation of smooth and analytic functions[J]. NeuralComputation,1996,32(8):1731-1742.
[74] Poggio T, Girosi F. Regilarization algorithms for learning that are equivalent to multilayer networks[J].Science,1990,1247(32):978-982.
[75]史忠植.神经网络[M].北京:高等教育出版社,2009,48-63.
[76] Haykin S. Adaptive Filter Theory.3rd. Upper Saddle River [M]. NJ: Prentice Hall,1996,56-89.
[77]俞金寿.工业过程先进控制[M].北京:中国石化出版社,2002.
[78]傅庚,唐小我,曾勇.广义递归方差倒数组合预测方法研究[J].电子科技大学学报,1995,24(2):211-217.
[79]陈友华.基于预测有效度的组合预测模型研究[J].预测,1997,16(3):72-73.
[80]王应明,罗英.调和平均组合预测中的参数估计技术[J].系统工程与电子技术,1997,19(10):45-49.
[81]王应明.广义加权几何平均组合预测方法研究[J].厦门大学学报(自然科学版),1998,37(1):6-10.
[82]陈华友.组合预测方法有效性理论及其应用[M].北京:科学出版社,2008.
[83] Wolpert D H. Stacked generalization [J]. Neural Networks,1992,5(2):241-259.
[84] Sridhar D V, Bartlett E B, Seagrave R C. Information theoretic approach for combining neural networkprocess models[J]. Neural Networks,1999,12(6):915-926.
[85] Sridhar D V,Seagrave R C, Bartlett E B. Process modelling using stacked neural networks[J]. AIChE J,1996,42(9):2529-2539.
[86]夏陆岳.聚氯乙烯智能制造技术研究[D].杭州:浙江工业大学,2004.
[87]王新楼.小波去噪方法分析与MATLAB仿真[J].工业控制计算机,2008,21(6):55-60.
[88]杜浩藩,丛爽.基于MATLAB的小波去噪方法研究[J].计算机仿真,2003,20(7):119-221.
[89]杨慧中,钟豪,丁锋.基于多重小波变换的信号去噪及其在软测量中的应用[J].仪器仪表学报,2007,28(7):1245-1449.
[90]王海清,余世明.基于故障诊断性能优化的主元个数选取方法[J].化工学报,2004,55(2):214-219.
[91] Rudolph G. Convergence analysis of canonical genetic algorithms[J]. IEEE Transactions on NeuralNetworks1994,5(1):96-101.
[92] Kennedy J, Eberhart R C. Particle swarm optimization[C]. Proceedings of IEEE InternationalConference on Neural Networks, Perth: IEEE,1995,1942-1948.
[93] Wilson E O. Sociobiology: The new synthesis [M]. Cambridge, MA: Belknap Press.1975.
[94]曾建潮,介婧,崔志华.微粒群算法[M].北京:科学出版社,2004.
[95] Zhou C, Chen T. Chaotic annealing for optimization [J].Physical Review E,1997,55(3):2580-2587.
[96] Choi C, Lee J. Chaotic local search Algorithm [J]. Artificial Life and Robotics,1998,2(1):41-47.
[97]梁瑞鑫,郑德玲.基于区间套混沌搜索的混合优化方法[J].北京科技大学学报,2002,2(1):41-47.
[98]张彤,土宏伟,王子才.变尺度混沌优化方法及其应用[J].控制与决策,1999,14(3):285-288.
[99]王子才,张彤,土宏伟.基于混沌变量的模拟退火优化方法[J].控制与决策,1999,14(4):381-384.
[100]盛煜翔.基于混沌粒子群算法的闪蒸过程优化研究[D].浙江：浙江工业大学，2009.
[101]Victoire T A A, Jeyakumar A E. Hybrid PSO-SQP for economic dispatch with valve-point effect[J].Electric Power Systems Research,2004,71(1):51-59.
[102]骆晨钟,邵惠鹤.用混沌搜索求解非线性约束优化问题[J].系统工程理论与实践,2000,26(8):54-56.
[103]Kiparissides C. Polymerization reactor modeling: a review of recent development and future directions[J]. Chemical Engineering Science,1996,51(10):1637-1659.
[104]Scorah M J, Dhib R, Penlidis A. Modelling of free radical polymerization of styrene and methylmethacrylate by a tetrafunctional initiator [J]. Chemical Engineering Science,2006,61(15):4827-4859.
[105]Dong M K, Piet D I. Modeling of branching density and branching distribution in low-densitypolyethylene polymerization [J]. Chemical Engineering Science,2008,63(8):2035-2046.
[106]Mariano A, Adriana B. Modeling and optimization of a high-pressure ethylene polymerization reactorusing gPROMS [J]. Computers and Chemical Engineering,2008,32(3):396-408.
[107]Yi H S, Kin J H, Han C, et al. Plantwide optimal grade transition for an industrial high-densitypolyethylene plant [J]. Industrial and Engineering Chemistry Research,2003,42(1):91-98.
[108]Flores T A, Biegler L T, Saldivar G E. Optimal grade transitions in the high-impact polystyrenepolymerization process [J]. Industrial and Engineering Chemistry Research,2006,45(18):6175-6189.
[109]Sharmin R, Sundararaj U, Shah Sirish, et al. Inferential sensors for estimation of polymer qualityparameters: Industrial application of a PLS-based soft sensor for a LDPE plant [J]. ChemicalEngineering Science,2006,61(19):6372-6384.
[110]Liu J. On-line soft sensor for polyethylene process with multiple production grades [J]. ControlEngineering Practice,2007,15(7):769-778.
[111]金学兰,袁璞,胡品慧等.丙烯聚合反应动态模拟[J].化工自动化及仪表,2005,32(3):7-10.
[112]邬正义,施健.基于鲁棒最小二乘支持向量机的聚丙烯熔融指数预报[J].华东理工大学学报(自然科学版),2006,32(7):889-892.
[113]Shi J, Liu X G, Sun Y X. Melt index prediction by neural networks based on independent componentanalysis and multi-scale analysis [J]. Neurocomputing,2006,70(1-3):280-287.
[114]Zhang J, Qi B X, Yong M. Inferential estimation of polymer melt index using sequentially trainedbootstrap aggregated neutral networks [J]. Chemical Engineering Technology,2006,29(4):442-448.
[115]McAuley K B, MacGregor J F. On-line inference of polymer polymer properties in an industrialpolyethylene reactor [J]. AIChE J,1991,37(6):825-835.
[116]Rallo R, Ferre-GineJ, Arenas A, et al. Neural virtual sensor for the inferential prediction of productquality from process variables [J]. Computer and Chemical Engineering,2002,26:1735-1754.
[117]Kano M, Hasebe S, Hashimoto I. A new multivariate statistical process monitoring method usingprincipal component analysis [J]. Computers and Chemical Engineering,2001,25(7-8):1103-1113,2001.
[118]Seongkyu Y, John F, MacGregor J F. Fault diagnosis with multivariate statistical models part I: usingsteady state fault signatures [J]. Journal of Process Control,2001,11(4):387-400.
[119]Lee J M, Yoo C K, Choi S W. Nonlinear process monitoring using kernel principal component analysis[J]. Chemical Engineering Science,2004,59(1):223-234.
[120]李玉贵.液相本体法聚丙烯生产及应用[M].北京:中国石化出版社,1992.
[121]McAuley K.B., MacGregor J.F. Optimal grade transitions in a gas-phase polyethylene reactor[J]. AIChEJournal,1992,38:1564-1576.
[122]金学兰,袁璞,胡品慧.丙烯聚合反应动态建模及在牌号切换中的应用[J].系统仿真学报,2006,18(8):2089-2091,2096.
[123]费正顺,胡斌,叶鲁彬等.带路径约束的聚烯烃牌号切换操作优化方法[J].化工学报,2010,61(4):893-900.
[124]王靖岱,陈纪忠,阳永荣.迭代动态规划在树脂牌号切换最优化模型中的应用[J].高校化学工程学报,2000,14(3):264-269.
[125]黄骅.聚丙烯生产过程建模及牌号切换策略研究[D].杭州:浙江工业大学，2009.
[126]MahadevanR,Doyle III F J.Control-relevantsc heduling of polymer grade transitions[J]. AIChE,2002,48(8):1754-1764.
[127]荚亮.聚丙烯连续生产过程的牌号切换研究[D].杭州:浙江大学,2006.
[128]Vassiliadis V S, Sargent R W H, Pantelides C C. Solution of a class of multistage dynamic optimizationproblems (I): problems without path constraints [J]. Industrial and Engineering Chemistry Research,1994,33(9):2111-2122.
[129]Biegler L T, Grossmann I E. Retrospective on optimization [J]. Computers and Chemical Engineering,2004,28(8):1169-1192.
[130]Schlegel M, Marquardt W. Detection and exploitation of the control switching structure in the solutionof dynamic optimization problems[J]. Journal of Process Control,2006,16(3):275-290.