明胶浓度软测量建模研究
|
摘要
明胶浓度是明胶生产过程的重要参数之一，它在一定程度上反映了明胶质量的高低，因此明胶浓度的检测速度影响着整个明胶的生产进程。国内明胶企业大都采用人工定时采样、离线检测的方法测量胶液浓度，但这样不仅容易污染样本，也使测量出现滞后。为改变现有胶液浓度的测量方式，提高明胶生产的自动化水平，借助软测量建模的优势，通过测量易获得的辅助变量推导出难测量的主导变量，实现明胶浓度的在线测量。目前国内对明胶浓度在线测量的研究较少，已有的研究成果中所建的模型均为单一模型，所选的辅助变量为时间和温度，但是单模型存在建模过程特性匹配不佳、精度和外推能力差等缺陷。
本文以青海明胶公司提胶工序为研究背景，以提胶工序中较难测量的胶液浓度为研究对象，认真分析明胶生产工艺，对从工业现场采集的数据进行分析，选取与胶液浓度密切相关且较易测量的时间、温度和比重作为辅助变量。为减小各数据因数量级差异对测量结果造成的影响，本文对数据进行了归一化处理，然后建立一系列软测量多模型，以实现明胶浓度的在线测量。
本文分别建立了基于BP_Adaboost的多模型、基于模糊C均值聚类(FuzzyC-means, FCM)算法的支持向量机多模型和基于GK聚类(Gustafson_Kessel, GK)算法的径向基函数多模型，并将所建多模型用于测量明胶浓度。与相同算法的单模型测量结果进行对比，结果显示多模型具有更高的测量精度。
为使所建立的软仪表能真正发挥实际效用，本文最后将MATLAB与VC进行混合编程。选用MATLAB引擎与VC编程方法相结合，将MATLAB强大的计算功能与VC良好的运行环境完美匹配，建立人机交互界面，从而实现真正意义上软仪表的现场应用。
Concentration of gelatin is one of the important parameters of the gelatinproduction process, it can be reflects the level of quality of gelatin, so the gelatinconcentration detection speed affect the gelatin production process. Domestic gelatinmanufacturers mostly use artificial timing sampling method to measure liquidconcentrations, but this is not easily contaminated sample, but also measured with alag. In order to change the existing glue concentration measurement, improve thelevel of automation of the gelatin production, we deduced dominant variable difficultto measure by measuring auxiliary variable which are easily accessible with theadvantage of the soft sensor modeling to achieve the online measurement of theconcentration of gelatin.
This paper takes the Qinghai Gelatin company refining processes technology asthe background and takes the concentration of gelatin as study object, after carefulanalysis of the gelatin production process and analyze data collected from anindustrial site, we select time, temperature and specific gravity which are easiermeasurement closely related to the glue concentration as an auxiliary variable.Toreduce the impact on the measurement results by the data due to differences inmagnitude, the data were normalized, and then create a series of soft measurementmodel to achieve the online measurement of the concentration of gelatin.
In this paper, multi-models based on BP_Adaboost and the fuzzy C-meansclustering (Fuzzy C-means FCM) algorithm and support vector machine and GKclustering (Gustafson_Kessel, GK) algorithm of radial basis function is establishedrespectively and the multi-models used to measure the concentration of gelatin.werecompared with the single model use the same algorithm, measurement results showthat the multi-model has a higher measurement accuracy.
To make the soft instrument really play the practical utility, we use the MATLABand VC for mixing programme. Combine the MATLAB engine and the VCprogramming methods and match the powerful computing features in MATLAB andVC operating environment perfectly, and create interactive interface, a true sense ofthe soft sensor field applications is realized at last.
本文以青海明胶公司提胶工序为研究背景，以提胶工序中较难测量的胶液浓度为研究对象，认真分析明胶生产工艺，对从工业现场采集的数据进行分析，选取与胶液浓度密切相关且较易测量的时间、温度和比重作为辅助变量。为减小各数据因数量级差异对测量结果造成的影响，本文对数据进行了归一化处理，然后建立一系列软测量多模型，以实现明胶浓度的在线测量。
本文分别建立了基于BP_Adaboost的多模型、基于模糊C均值聚类(FuzzyC-means, FCM)算法的支持向量机多模型和基于GK聚类(Gustafson_Kessel, GK)算法的径向基函数多模型，并将所建多模型用于测量明胶浓度。与相同算法的单模型测量结果进行对比，结果显示多模型具有更高的测量精度。
为使所建立的软仪表能真正发挥实际效用，本文最后将MATLAB与VC进行混合编程。选用MATLAB引擎与VC编程方法相结合，将MATLAB强大的计算功能与VC良好的运行环境完美匹配，建立人机交互界面，从而实现真正意义上软仪表的现场应用。
Concentration of gelatin is one of the important parameters of the gelatinproduction process, it can be reflects the level of quality of gelatin, so the gelatinconcentration detection speed affect the gelatin production process. Domestic gelatinmanufacturers mostly use artificial timing sampling method to measure liquidconcentrations, but this is not easily contaminated sample, but also measured with alag. In order to change the existing glue concentration measurement, improve thelevel of automation of the gelatin production, we deduced dominant variable difficultto measure by measuring auxiliary variable which are easily accessible with theadvantage of the soft sensor modeling to achieve the online measurement of theconcentration of gelatin.
This paper takes the Qinghai Gelatin company refining processes technology asthe background and takes the concentration of gelatin as study object, after carefulanalysis of the gelatin production process and analyze data collected from anindustrial site, we select time, temperature and specific gravity which are easiermeasurement closely related to the glue concentration as an auxiliary variable.Toreduce the impact on the measurement results by the data due to differences inmagnitude, the data were normalized, and then create a series of soft measurementmodel to achieve the online measurement of the concentration of gelatin.
In this paper, multi-models based on BP_Adaboost and the fuzzy C-meansclustering (Fuzzy C-means FCM) algorithm and support vector machine and GKclustering (Gustafson_Kessel, GK) algorithm of radial basis function is establishedrespectively and the multi-models used to measure the concentration of gelatin.werecompared with the single model use the same algorithm, measurement results showthat the multi-model has a higher measurement accuracy.
To make the soft instrument really play the practical utility, we use the MATLABand VC for mixing programme. Combine the MATLAB engine and the VCprogramming methods and match the powerful computing features in MATLAB andVC operating environment perfectly, and create interactive interface, a true sense ofthe soft sensor field applications is realized at last.
引文
[1]牟忠臣，李国庆.明胶溶液浓度与比重、温度的关系[J].明胶科学与技术，1995，15(3)：142-144.
[2] Khalfe Nadeem M., Lahiri Kumar Sandip, Sawke Kumar Sunil. Soft sensor forbetter control of carbon dioxide removal process in ethylene glycol plant [J].Chemical Industry and Chemical Engineering Quarterly,2011,17(1):17-20.
[3]强明辉，邹有锋.基于RBF神经网路的发酵过程菌体浓度的软测量[J].自动化与仪器仪表，2012(2)：179-180.
[4] Geng Liang, Hong Wang,Wen Li,Yan Bai. Optimized combustion system incoal-fired boiler: using optimization control and fieldbus[J]. Transactions of theInstitute of Measurement and Control,2011,33(2):291-312.
[5]傅永峰.软测量建模方法研究及其工业应用[D].浙江大学博士论文，2007.
[6]李琦.炼油精馏过程软测量与综合优化控制的研究[D].大连理工大学博士论文，2008.
[7]丁云，于静江，周春晖.原油蒸馏塔的质量估计和优化管理[J].石油炼制与化工，1994(5)：23-28.
[8]侯卫锋.催化重整流程模拟与优化技术及其应用研究[D].浙江大学博士论文，2006.
[9] X. Hulhoven, A. Vande Wouwer, Ph Bogaerts. Hybrid extendedluenberger-asymptotic observer for bioprocess state estimation. Chemical EngineeringScience,2006,61(21):7151-7160.
[10] Rocco Tarantino, Ferenc Szigeti, Eliezer Colina-Morles. Generalized luenbergerobserver-based fault-detection filter design: An industrial application[J]. ControlEngineering Practice,2000,8(6):665-671.
[11] Davide Fissore, David Edouard, Hassan Hammouri, Antonello A. Barresi.Nonlinear soft-sensor design for unsteady-state voc afterburners[J]. AIChEJournal,2006,52(1):282-291.
[12] Olivier Bernard, Antoine Sciandra, Gauthier Sallet. Anon-linear software senaorto monitor the internal nitrogen quota of phytoplanktonic cells[J]. QceanologicaActa,2001,24(5):435-442.
[13] Antonello A. Barresi, Salvatore A. Velardi, Roberto Pisano, Valeria Rasetto,Alberto Vallan, Miquel Galan. In-line control of the lyophilization process. A gentlepat approach using software sensors[J]. International Journal of Refrigeration In Press,Accepted Manuscript,2008,32(5):1003-1014.
[14] Ph Bogaerts, A. V. A. Vande Wouwer. Parameter identification for stateestimation-application to bioprocess software sensors[J]. Chemucal EnginweeringScience,2004,59(12):2465-2476.
[15] E. Akkari, S. Chevallier, L. Boillereaux. Real-time estimation of food defrostingby software sensors[J]. AIChE Journal,2006,52(4):1473-1480.
[16] Oscar A. Z. Sotomayor, Song Won Park, Claudio Garcia. Software sensor foron-line estimation of the microbial activity in activated sludge systems[J]. ISATransactions,2002,41:127-143.
[17] Ph Bogaerts, A. Vande Wouwer. software sensors for bioprocesses[J]. ISATransactions,2003,42:547-558.
[18] D. Chapelle, P. Moireau, P. Le Tallec. Robust filtering for joint state-parameterestimation in distributed mechanical systems[J]. Computer Methods in AppliedMechanics and Engineering,2008,197:659-677.
[19] Rajesh Rajamani, Karl Hedrick. Adaptive observer for active automotivesuspensions-theory and experiment. IEEE Trans. on Control SystemsTechnology,1995,3(1):86-93.
[20]孙帆，钱锋.乙二醇生产过程中环氧乙烷浓度的软测量研究[J].计算机与应用化学，2010，27(1)：06-10.
[21]赵忠盖，刘飞.基于因子回归模型的软测量方法[J].计算机与应用化学，2010，27(1)：38-40.
[22] Long Han. Kernel Partial Least Squares(K-PLS) for Scientfic Data Mining[D].Rensselaer Polytechnic Institute,2007.
[23]王永吉，杨慧中.基于K近邻的支持向量机的多模型建模[J].江南大学学报，2010，9(1)：8-10.
[24]边肇祺.模式识别[M].北京：清华大学出版社，1999.
[25] Carl Edward Rasmussen, Chris Williams. Gaussian processes for machineLearning. the MIT Press,2006.
[26] Fernando di Sciascio, Adriana N. Amicarelli. Biomass estimation in batchbiotechnological processes by bayesian gaussian process regression[J]. Computers&Chemical Engineering,2008,32(12):3264-3273.
[27] Alexandra Grancharova, Jus Kocijan, Tor A. Johansen. Explicit stochastic controlof combustion plants based on gaussian process models[J]. Automatica,2008,44(6):1621-1631.
[28] J. Kocijan, B. Likar. Gas-liquid separator modelling and simulation withgaussian-process models[J]. Simulation Modelling Practice and Theory,2008,16(8):910-922.
[29]陈丽华，马生全，李丽.基于人工神经网络样本优化法识别的河流溶解氧模型[J].海南师范大学学报(自然科学版)，2008(4)：337-339.
[30] Ruilan Liu, Hongye Su, Shengjing Mu, Tao Jia, Weiquan Chen, Jian Chu. Fuzzyneural network model of4-cba concentration for industrial purified terephthalic acidoxidation process[J]. Chinese Journal of Chemical Engineering,2004,12(2):234-239.
[31] G. Zahedi, A. Elkamel, A. Lohi, A. Jahanmiri,M. R. Rahimpor. Hybrid artificialneural network-first principle model formulation for the unsteady state simulation andanalysis of a packed bed reactor for co2hydrogenation to methanol[J]. ChemicalEngineering Journal,2005,115(1-2):113-120.
[32]袁平，毛志忠，王福利.基于网络多模型软测量技术及应用[J].系统仿真学报，2006，18(6)：1458-1465.
[33]陈金凤，杨慧中，邓玉俊.一种基于LDA和FCM的BPA多模型软测量方法[J].华东理工大学学报，2010，2：126-129.
[34] Jialin Liu, On-line soft sensor for polyethylene process multiple productiongrades[J]. Control Engineering Practice,2007,15(7):769-778.
[35]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，2005，31(2)：208-211.
[36]熊志化，张卫庆，赵瑜等.基于混合高斯过程的多模型热力参数测量软仪表[J].中国电机工程学报，2005，7：30-33.
[37]林金星，沈炯，李益国.基于递阶g-k聚类的热工过程多模型建模方法[J].中国电机工程学报，2006，26(11)：23-28.
[38]杨翊鹏，李少远.基于满意聚类的非线性系统多模型建模方法[J].上海交通大学学报，2003(4)：489-492.
[39]王福利，桑海峰，何大阔等.基于多支持向量机的诺西肽发酵中菌体浓度软测量[J].系统仿真学报，2006，18(7)：1983-1986.
[40]高丽敏.明胶浓度的软测量建模及参数优化[D].兰州:兰州理工大学电信学院，2011.
[41] Yoichiro Maeda, Toru Tsubouchi. Parallel Genetic Algorithm Used FuzzyAdaptive Search Method[J].SICE Annual conference in Fukui.August2003,2353-2356.
[42] De Assis A.J.Filho R.M.. Soft Sensors Development for On-line Bioreactor StateEstimation[J].Computers and Chemical Engineering.2000,24:1099-1103.
[43] Arthur Kordon. Robust soft sensors on integration of genetic programming,analytical neural networks, and support vector machines Evolutionary Computation[J]. Proceedings of the2002Congress on, May2002(1):896-1019.
[44] MATLAB中文论坛.MAT，LAB神经网络30个案例分析[M].北京：北京航空航天大学出版社，2010.
[45]周雅婷.明胶史[J].明胶科学与技术，2011，31(2)：84-91.
[46]杨丽萍，张新明，白云清等.利用动物皮、骨生产胶原多肽[J].山东食品发酵，2005，137(2)：20-21.
[47]姚龙坤，顾平远，黄文涛.明胶在食品工业中的应用简述[J].明胶科学与技术，2009(29)：131-132.
[48]黄明智，黄雅钦.明胶的疗效与保健功能及其发展前景[J].明胶科学与技术，2005，25(1)：4-15.
[49] hbjllyy.http://baike.baidu.com/view/232149.htm,2012-03-05.
[50]杨穆，黄明智.明胶类血浆代用品[J].明胶科学与技术，2001，21(3)：113-119.
[51]周蓓蓓.明胶浓度软测量建模及参数优化[D].兰州:兰州理工大学电信学院，2010.
[52]周世海.骨明胶生产[J].明胶科学与技术，2005,25(3)：132-141.
[53]明胶科学与技术信息编辑组.骨明胶工艺及设备[J].明胶科学与技术，2005，25(4)：173-182.
[54]邓文.骨明胶生产的浸酸工艺[J].明胶科学与技术，2005,25(2)：68-71.
[55]邓文，邓斌才.骨明胶生产后工序[J].明胶科学与技术，2006,26(4)：177-187.
[56] Toru Otsuru，Reiji Tomiku，Noriko Okamoto etal. Ensemble averaged surfacenormal impedance measurement method in a reverberation room[J]. AcousticalScience and Technology，2011,32(2)：69-72.
[57]WeiXu, LingboZhang, XingshengGu. Soft sensor for ammonia concentration atthe ammonia converter outlet based on an improved particle swarm optimization andBP neural network[J]. Chemical Engineering Research&Design; Transactions of theInstitution of Chemical Engineers，2011,89(10)：2102-2109.
[58] Li-hongJIANG，Ai-guoWANG，Nai-yuanTIAN etal. BP Neural Network ofContinuous Casting Technological Parameters and Secondary Dendrite Arm Spacingof Spring Steel[J]. Journal of Iron and Steel Research International，2011,18(8)：25-29.
[59] Shengdong Li, Xueqiang Lv, Chenlin Zhan etal. Regional Division of policepatrols Based on Adaptive FCM Clustering and MMAS[J]. Journal ofComputers,2011,6(2):313-320.
[60] S. Ramathilagam, R. Pandiyarajan,A. Sathya,etal. Modified fuzzy c-meansalgorithm for segmentation of T1–T2-weighted brain MRI[J]. Journal ofComputational and Applied Mathematics.2011,235(6).
[61] Wen-Tsao Pan. Combining fuzzy sammon mapping and fuzzy clustering approachto perform clustering effect analysis: Take the banking service satisfaction as anexample[J]. Expert Systems with Applications,2010,37(6):4139-4145.
[62] P. C. Nayak,K. P. Sudheer. Fuzzy model identification based on clusterestimation for reservoir inflow forecasting[J]. HydrologicalProcesses,2008,22(6):827=841.
[63]梁锦锦.支持向量机和支持向量域描述的若干问题研究[D].西安：西安电子科技大学，2009.
[64]孙宗海.支持向量机及其在控制中的应用研究[D].兰州：兰州大学，2003.
[65]牟洪波.基于BP和RBF神经网络的木材缺陷监测研究[D].东北林业大学，2010.
[66]骆平，范瑜.基于RBFNN的高压断路器机械故障诊断系统[J].中国电力，2010,43(5):30-34.
[67]赵勇.基于GA-RBFNN算法的列车车轮踏面损伤识别[J].计算机工程与应用，2012,1(5).
[68]姜万录，雷亚飞，张齐生等.基于RBFNN建模的动态流量软测量方法研究[J].仪器仪表学报，2008，29(9):1888-1893.
[69]董维国.深入浅出MATLAB7.X混合编程[M].北京：机械工业出版社，2006.
[70]王素立，高洁，孙新德.MATLAB混合编程与工程应用[M].北京：清华大学出版社，2008.
[2] Khalfe Nadeem M., Lahiri Kumar Sandip, Sawke Kumar Sunil. Soft sensor forbetter control of carbon dioxide removal process in ethylene glycol plant [J].Chemical Industry and Chemical Engineering Quarterly,2011,17(1):17-20.
[3]强明辉，邹有锋.基于RBF神经网路的发酵过程菌体浓度的软测量[J].自动化与仪器仪表，2012(2)：179-180.
[4] Geng Liang, Hong Wang,Wen Li,Yan Bai. Optimized combustion system incoal-fired boiler: using optimization control and fieldbus[J]. Transactions of theInstitute of Measurement and Control,2011,33(2):291-312.
[5]傅永峰.软测量建模方法研究及其工业应用[D].浙江大学博士论文，2007.
[6]李琦.炼油精馏过程软测量与综合优化控制的研究[D].大连理工大学博士论文，2008.
[7]丁云，于静江，周春晖.原油蒸馏塔的质量估计和优化管理[J].石油炼制与化工，1994(5)：23-28.
[8]侯卫锋.催化重整流程模拟与优化技术及其应用研究[D].浙江大学博士论文，2006.
[9] X. Hulhoven, A. Vande Wouwer, Ph Bogaerts. Hybrid extendedluenberger-asymptotic observer for bioprocess state estimation. Chemical EngineeringScience,2006,61(21):7151-7160.
[10] Rocco Tarantino, Ferenc Szigeti, Eliezer Colina-Morles. Generalized luenbergerobserver-based fault-detection filter design: An industrial application[J]. ControlEngineering Practice,2000,8(6):665-671.
[11] Davide Fissore, David Edouard, Hassan Hammouri, Antonello A. Barresi.Nonlinear soft-sensor design for unsteady-state voc afterburners[J]. AIChEJournal,2006,52(1):282-291.
[12] Olivier Bernard, Antoine Sciandra, Gauthier Sallet. Anon-linear software senaorto monitor the internal nitrogen quota of phytoplanktonic cells[J]. QceanologicaActa,2001,24(5):435-442.
[13] Antonello A. Barresi, Salvatore A. Velardi, Roberto Pisano, Valeria Rasetto,Alberto Vallan, Miquel Galan. In-line control of the lyophilization process. A gentlepat approach using software sensors[J]. International Journal of Refrigeration In Press,Accepted Manuscript,2008,32(5):1003-1014.
[14] Ph Bogaerts, A. V. A. Vande Wouwer. Parameter identification for stateestimation-application to bioprocess software sensors[J]. Chemucal EnginweeringScience,2004,59(12):2465-2476.
[15] E. Akkari, S. Chevallier, L. Boillereaux. Real-time estimation of food defrostingby software sensors[J]. AIChE Journal,2006,52(4):1473-1480.
[16] Oscar A. Z. Sotomayor, Song Won Park, Claudio Garcia. Software sensor foron-line estimation of the microbial activity in activated sludge systems[J]. ISATransactions,2002,41:127-143.
[17] Ph Bogaerts, A. Vande Wouwer. software sensors for bioprocesses[J]. ISATransactions,2003,42:547-558.
[18] D. Chapelle, P. Moireau, P. Le Tallec. Robust filtering for joint state-parameterestimation in distributed mechanical systems[J]. Computer Methods in AppliedMechanics and Engineering,2008,197:659-677.
[19] Rajesh Rajamani, Karl Hedrick. Adaptive observer for active automotivesuspensions-theory and experiment. IEEE Trans. on Control SystemsTechnology,1995,3(1):86-93.
[20]孙帆，钱锋.乙二醇生产过程中环氧乙烷浓度的软测量研究[J].计算机与应用化学，2010，27(1)：06-10.
[21]赵忠盖，刘飞.基于因子回归模型的软测量方法[J].计算机与应用化学，2010，27(1)：38-40.
[22] Long Han. Kernel Partial Least Squares(K-PLS) for Scientfic Data Mining[D].Rensselaer Polytechnic Institute,2007.
[23]王永吉，杨慧中.基于K近邻的支持向量机的多模型建模[J].江南大学学报，2010，9(1)：8-10.
[24]边肇祺.模式识别[M].北京：清华大学出版社，1999.
[25] Carl Edward Rasmussen, Chris Williams. Gaussian processes for machineLearning. the MIT Press,2006.
[26] Fernando di Sciascio, Adriana N. Amicarelli. Biomass estimation in batchbiotechnological processes by bayesian gaussian process regression[J]. Computers&Chemical Engineering,2008,32(12):3264-3273.
[27] Alexandra Grancharova, Jus Kocijan, Tor A. Johansen. Explicit stochastic controlof combustion plants based on gaussian process models[J]. Automatica,2008,44(6):1621-1631.
[28] J. Kocijan, B. Likar. Gas-liquid separator modelling and simulation withgaussian-process models[J]. Simulation Modelling Practice and Theory,2008,16(8):910-922.
[29]陈丽华，马生全，李丽.基于人工神经网络样本优化法识别的河流溶解氧模型[J].海南师范大学学报(自然科学版)，2008(4)：337-339.
[30] Ruilan Liu, Hongye Su, Shengjing Mu, Tao Jia, Weiquan Chen, Jian Chu. Fuzzyneural network model of4-cba concentration for industrial purified terephthalic acidoxidation process[J]. Chinese Journal of Chemical Engineering,2004,12(2):234-239.
[31] G. Zahedi, A. Elkamel, A. Lohi, A. Jahanmiri,M. R. Rahimpor. Hybrid artificialneural network-first principle model formulation for the unsteady state simulation andanalysis of a packed bed reactor for co2hydrogenation to methanol[J]. ChemicalEngineering Journal,2005,115(1-2):113-120.
[32]袁平，毛志忠，王福利.基于网络多模型软测量技术及应用[J].系统仿真学报，2006，18(6)：1458-1465.
[33]陈金凤，杨慧中，邓玉俊.一种基于LDA和FCM的BPA多模型软测量方法[J].华东理工大学学报，2010，2：126-129.
[34] Jialin Liu, On-line soft sensor for polyethylene process multiple productiongrades[J]. Control Engineering Practice,2007,15(7):769-778.
[35]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，2005，31(2)：208-211.
[36]熊志化，张卫庆，赵瑜等.基于混合高斯过程的多模型热力参数测量软仪表[J].中国电机工程学报，2005，7：30-33.
[37]林金星，沈炯，李益国.基于递阶g-k聚类的热工过程多模型建模方法[J].中国电机工程学报，2006，26(11)：23-28.
[38]杨翊鹏，李少远.基于满意聚类的非线性系统多模型建模方法[J].上海交通大学学报，2003(4)：489-492.
[39]王福利，桑海峰，何大阔等.基于多支持向量机的诺西肽发酵中菌体浓度软测量[J].系统仿真学报，2006，18(7)：1983-1986.
[40]高丽敏.明胶浓度的软测量建模及参数优化[D].兰州:兰州理工大学电信学院，2011.
[41] Yoichiro Maeda, Toru Tsubouchi. Parallel Genetic Algorithm Used FuzzyAdaptive Search Method[J].SICE Annual conference in Fukui.August2003,2353-2356.
[42] De Assis A.J.Filho R.M.. Soft Sensors Development for On-line Bioreactor StateEstimation[J].Computers and Chemical Engineering.2000,24:1099-1103.
[43] Arthur Kordon. Robust soft sensors on integration of genetic programming,analytical neural networks, and support vector machines Evolutionary Computation[J]. Proceedings of the2002Congress on, May2002(1):896-1019.
[44] MATLAB中文论坛.MAT，LAB神经网络30个案例分析[M].北京：北京航空航天大学出版社，2010.
[45]周雅婷.明胶史[J].明胶科学与技术，2011，31(2)：84-91.
[46]杨丽萍，张新明，白云清等.利用动物皮、骨生产胶原多肽[J].山东食品发酵，2005，137(2)：20-21.
[47]姚龙坤，顾平远，黄文涛.明胶在食品工业中的应用简述[J].明胶科学与技术，2009(29)：131-132.
[48]黄明智，黄雅钦.明胶的疗效与保健功能及其发展前景[J].明胶科学与技术，2005，25(1)：4-15.
[49] hbjllyy.http://baike.baidu.com/view/232149.htm,2012-03-05.
[50]杨穆，黄明智.明胶类血浆代用品[J].明胶科学与技术，2001，21(3)：113-119.
[51]周蓓蓓.明胶浓度软测量建模及参数优化[D].兰州:兰州理工大学电信学院，2010.
[52]周世海.骨明胶生产[J].明胶科学与技术，2005,25(3)：132-141.
[53]明胶科学与技术信息编辑组.骨明胶工艺及设备[J].明胶科学与技术，2005，25(4)：173-182.
[54]邓文.骨明胶生产的浸酸工艺[J].明胶科学与技术，2005,25(2)：68-71.
[55]邓文，邓斌才.骨明胶生产后工序[J].明胶科学与技术，2006,26(4)：177-187.
[56] Toru Otsuru，Reiji Tomiku，Noriko Okamoto etal. Ensemble averaged surfacenormal impedance measurement method in a reverberation room[J]. AcousticalScience and Technology，2011,32(2)：69-72.
[57]WeiXu, LingboZhang, XingshengGu. Soft sensor for ammonia concentration atthe ammonia converter outlet based on an improved particle swarm optimization andBP neural network[J]. Chemical Engineering Research&Design; Transactions of theInstitution of Chemical Engineers，2011,89(10)：2102-2109.
[58] Li-hongJIANG，Ai-guoWANG，Nai-yuanTIAN etal. BP Neural Network ofContinuous Casting Technological Parameters and Secondary Dendrite Arm Spacingof Spring Steel[J]. Journal of Iron and Steel Research International，2011,18(8)：25-29.
[59] Shengdong Li, Xueqiang Lv, Chenlin Zhan etal. Regional Division of policepatrols Based on Adaptive FCM Clustering and MMAS[J]. Journal ofComputers,2011,6(2):313-320.
[60] S. Ramathilagam, R. Pandiyarajan,A. Sathya,etal. Modified fuzzy c-meansalgorithm for segmentation of T1–T2-weighted brain MRI[J]. Journal ofComputational and Applied Mathematics.2011,235(6).
[61] Wen-Tsao Pan. Combining fuzzy sammon mapping and fuzzy clustering approachto perform clustering effect analysis: Take the banking service satisfaction as anexample[J]. Expert Systems with Applications,2010,37(6):4139-4145.
[62] P. C. Nayak,K. P. Sudheer. Fuzzy model identification based on clusterestimation for reservoir inflow forecasting[J]. HydrologicalProcesses,2008,22(6):827=841.
[63]梁锦锦.支持向量机和支持向量域描述的若干问题研究[D].西安：西安电子科技大学，2009.
[64]孙宗海.支持向量机及其在控制中的应用研究[D].兰州：兰州大学，2003.
[65]牟洪波.基于BP和RBF神经网络的木材缺陷监测研究[D].东北林业大学，2010.
[66]骆平，范瑜.基于RBFNN的高压断路器机械故障诊断系统[J].中国电力，2010,43(5):30-34.
[67]赵勇.基于GA-RBFNN算法的列车车轮踏面损伤识别[J].计算机工程与应用，2012,1(5).
[68]姜万录，雷亚飞，张齐生等.基于RBFNN建模的动态流量软测量方法研究[J].仪器仪表学报，2008，29(9):1888-1893.
[69]董维国.深入浅出MATLAB7.X混合编程[M].北京：机械工业出版社，2006.
[70]王素立，高洁，孙新德.MATLAB混合编程与工程应用[M].北京：清华大学出版社，2008.