用户名: 密码: 验证码:
支持向量机及其在制浆过程重要参数软测量中的应用研究
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
节能减排、环境保护及消费者对纸张质量要求的提高客观上要求造纸行业必须进一步提高企业的自动化及信息化程度。制浆过程卡伯值、碱回收过程及洗涤过程的黑液浓度的在线测量一直是浆纸企业关注的热点,也是影响浆纸工业信息化及自动化发展的难点。支持向量机(Support VectorMachines,SVM)是一种核函数学习机器,它遵循结构风险最小化原则,具有理论完备、适应性强、推广能力好、全局优化等优点,是当前国际工业自动化领域的一个研究热点。与传统机器学习方法相比,SVM具有良好的发展与应用潜力。本文以SVM在制浆过程一些重要工艺参数在线软测量应用中的若干问题为主线,结合相应过程工艺知识,针对现有软测量模型在实际应用过程中普遍存在诸如速度慢、精度低、缺乏在线校正等问题,提出或改进了若干算法。仿真研究及应用效果表明本文提出的算法是有效的。本文的主要贡献可总结如下:
     (1)基于SVM算法的卡伯值分类在线自适应软测量建模。针对现有蒸煮过程卡伯值软测量模型存在精度低、在线适应能力弱的缺点,提出基于SVM卡伯值分类在线自适应软测量模型。工业过程一般工作在几个有限工作点附近,蒸煮过程也不例外。根据过程这个特点,本文采用模糊c均值聚类方法将蒸煮过程卡伯值软测量样本点划归成若干类(由于采样数据所限,本文选两类)。划分原则:每类中训练样本间最大程度相似;不同类中训练样本最大程度不同。然后分别建立各类软测量模型。通过这种方式,就把一个大类的卡伯值软测量模型细化成各个小类卡伯值软测量模型,相应地提高了卡伯值模型的静态精度。通过对国内外蒸煮过程卡伯值常用的软测量方法优缺点及国内使用情况分析发现,大多数已有模型提高精度的方法主要集中在静态模型精度上,比如增加过程信息量等,也就是在如何提供模型静态精度上开展工作较多,但对模型使用过程中精度研究相对较少。通过工艺分析及国内蒸煮现状研究,发现国内蒸煮过程工况变化比较频繁。如果卡伯值软测量模型在线自适应能力弱,模型使用过程中,精度必然降低。工况经常变化的蒸煮过程卡伯值软测量不适合采用具有批处理式、离线建模的一般支持向量回归机算法。适合采用适应工况变化的增量自适应SVM算法。也就是在模型使用过程中不断采集新的样本,在原模型基础上再学习。这样的算法既兼顾了原有历史模型,又给模型引入了新工况要素。为了降低所建模型在使用过程中计算机内存占有量,提高模型运行速度,必须选择合适样本替换策略。考虑蒸煮过程典型慢时变特点,从算法和蒸煮特性两方面考虑,选用基于支持向量数据域描述策略和基于滑动时间窗法结合形式的样本替换策略。与普通的卡伯值软测量模型相比本文提出的模型优点:一方面充分地考虑到历史训练的结果,减少再建模型训练时间;另一方面对历史数据无需继续保存,减少算法对计算机存储空间要求。通过仿真及应用证明这种基于数据挖掘增量在线自适应支持向量机模型适合蒸煮过程卡伯值软测量建模。
     (2)基于在线自适应残差补偿LS-SVM算法的碱回收蒸发工段出效黑液浓度软测量建模。针对碱回收蒸发工段出效黑液浓度难于测量的情况,本文提出了在线自适应残差补偿LS-SVM模型。通过对碱回收过程工艺及多效蒸发器原理的学习、分析,影响碱回收蒸发过程黑液浓度的主要因素有进效稀黑液浓度和流量及多效蒸发器的总有效温差,其它因素为固形物成分组成、冷却水温度、环境温度、传热系数等。黑液蒸发一般由多效蒸发器组成,每效蒸发器又存在蒸汽、黑液和冷凝水等流程。蒸发过程的非线性比较严重。蒸发器内部管道有时出现的各种液柱脉动、出效黑液浓度和流量有时发生的振荡现象。由于蒸发器本身设备管道长、多,在蒸发系统的一端的量(流量、温度、浓度等)发生变化后,常常要过很长时间另一端才能做出反应,所以黑液浓度实际系统影响因素更复杂。目前黑液浓度软测量模型都采取以压力、温度为辅助变量,黑液浓度为主导变量建模,忽视了其它因素影响,这样建立模型精度并不高。本文建立的残差自适应补偿系统就考虑了其它影响因素。该模型将工业现场自适应校正模型和自适应残差补偿模型结合,从两个方面提高黑液波美度模型精度:一是选用最小二乘支持向量机在线算法进行模型训练和在线修正,使模型适应当前工况;二是增加残差补偿自适应环节,对模型输出的值进行在线修正,以使软测量模型的测量结果更准确。普通最小二乘支持向量机算法简单,计算量大,速度快,但所建模型会丧失支持向量机内在稀疏性,丧失支持向量机建模时鲁棒性。模型在运行时,样本数目增多,运行时间变长,占用内存加大。如果对算法不进行有效改进,实际中将无法使用。本文采用的在线自适应最小二乘支持向量机算法,用一个新样本代替一个时间最久样本,采用增量形式对模型在线校正,减少运算的复杂性,节约计算机内存。残差补偿自适应模型采用多元线性回归法建立残差与各影响因素间关系模型,采用丢弃最早误差值方法进行残差补偿模型的自适应校正。仿真及应用结果证明本文提出蒸发过程出效黑液浓度软测量模型精度高,能够适应工况变化。
     (3)基于模糊在线自适应LS-SVM算法的洗涤过程黑液浓度软测量建模。洗涤过程黑液浓度软测量研究相对较少,已有模型泛化能力差、精度低。针对已有模型存在的缺点,本文提出了自适应能力强的基于模糊在线自适应LS-SVM黑液浓度软测量模型。洗涤过程的黑液浓度在线测量也很困难。通过对洗涤过程工艺分析,其黑液浓度受纸浆流量、纸浆厚度、洗浆水、真空度、转鼓速度等影响,传统方法难以实时测量。考虑到工厂实际情况,在不增加仪器,不影响生产等情况下确定建模参数。通过在洗涤工段现场分析,与资深工程师讨论等形式,决定以进浆浓度、进浆流量、清水流量为洗涤过程中黑液浓度软测量模型辅助变量,以黑液浓度(首段)为主导变量,建立洗涤黑液浓度软测量模型。基于现场采集辅助变量、主导变量所建立的黑液浓度软测量模型存在问题:仅能大致反映实际工业对象本质的变化趋势。由此可见,其模型误差必然存在。模型投入使用后,由于洗涤系统的时变性、非线性和不确定性,随着时间推移,洗涤过程的特性及其工作点都可能发生一些变化,现在工况下的参数测量用原来采样数据所建模型测量,误差必然加大。在线自适应最小二乘算法就能适应工况的变化,在线地校正模型。实际洗涤过程采集的样本点都含有不同程度的噪声。为了提高模型精度,不同的样本对应不同的惩罚权系数,这样能够消除部分噪声和孤立点的影响。样本替换策略在增加一个新样本点的同时删除Lagrange乘子绝对值最小样本点。仿真与应用结果证明本文提出的软测量模型适合洗涤过程黑液浓度测量。
     (4)基于SVM算法的矛盾数据发现方法初步研究。洗涤过程测量仪表的校正不准确、不及时,仪表失灵、过程有较大的干扰、过程基准漂移、操作人员误操作、不定期清洗管道、管道堵塞等因素导致仪表示值的波动、数据失真等产生矛盾数据。这些矛盾数据是突发的,平时很难得到其具体样本。通过对SVM回归理论的研究,提出了主导变量的矛盾数据解决方案,借助例子对方案进行了验证,并将此方案实施在纸浆洗涤过程的主导变量的矛盾数据甄别中。
     本文以蒸煮过程卡伯值、碱回收及洗涤过程黑液浓度软测量为研究对象,以支持向量机为工具,围绕如何提高模型在线测量精度提出了一系列软测量方案,仿真及应用证明了这些方案的先进性。这些研究成果对其它的工业软测量模型也有一定参考价值。
In order to save energy, reduce emission, protect environment and improvethe quality of paper, it is necessary to further improve the level of automationand informationization in pulp and paper industry. Measurement of someparameters such as Kappa number during pulping, Black liquor consistency inwashing process and alkali recovery has always been a hotspot in pulp and paperindustry, and it has been one of typical difficulties in enterprise automation andinformationization development. Support Vector Machine is a kernel functionlearning machine following the principle of the structural risk minimization,which possesses the advantages of complete theory, global optimization, goodadaptability and better generalization. It is now a studying hotspot ofinternational industrial automation. Compared with the traditionalmachine-learning method, Support Vector Machine has great potential indevelopment and application. Surrounding with several application problems ofonline soft-measurement based on Support Vector Machine for some importantprocess parameters in pulp, the paper combines the process techniques andknowledge, and improves some algorithms to solve problems of slow speed, lowprecision and lack of online revise in application of the current soft-measurementmodel.Simulation and application results show the obvious effect of theimproved algorithms in the paper. The main contributions of this paper aresummed up as follows:
     (1)The research of soft measurement model for Kappa number based onthe on-line adaptive SVM. Pointed to the disadvantages such as low precision, poor online adaptive capacity of the former soft measurement model for Kappanumber in the pulp cooking process, a SVM-based online adaptive model ispresented. The cooking is conducted in several finite localities, so FuzzyC-means algorithm is adopted to divide sample points into several sectionsfollowing up the partition principals that make the training sample of each classmost similar and the training sample of different class most dissimilar. Then thesimilar soft measurement model is established. Through subdividing a maincategory of soft measurement model into some small categories, the staticprecision of the soft measurement model enhanced correspondingly. The meritsand demerits of common soft measurement methods for Kappa number at homeand abroad and the domestic applications are discussed, which shows that moreresearches are focused on improving the modeling static precision instead of thedynamic precision. Studies on the process analysis and the current cookingsituation show that there have been great changes taken place in domesticcooking operating conditions. Because of the variety of the domestic cookingoperating conditions, there is no doubt that the precision reduces during theprocess of modeling if the model has no ability of online self-adaptation. Ingeneral, Support Vector Regression is offline and batching, which isinappropriate for Kappa number measurement in cooking with constantoperating regime variety, so the algorithm should be improved to set up modelbased on delta self-adaptation Support Vector Machine adequate for operatingregime variety. That is to say, during the process of using the model, newsamples are collected constantly and the model is relearned repeatedly based onthe prime one. The algorithm injects the new operating regime factors into theoriginal model appropriately. In order to reduce the share of computer memoryand improve the model speed, it is necessary to select the appropriate samplereplacement strategy. Considering that cooking is a typical slowly time-variedprocess, the paper adopts the sample replacing strategy which combines themethod based on Support Vector data description and the method based onmoving time window to satisfy the algorithm and performance of cookingprocess. Compared with general methods, on one hand, the model makes full useof the historical training results and decreases follow-up time significantly; onthe other hand, the history data is no longer to be saved to reduce the storage space required by the algorithm. Simulation and application results show that thedelta on-line self-adaptation Support Vector Machine model based on datamining is suitable for the measurement of Kappa number in cooking process.
     (2)The research of soft measurement model for Black liquor consistencyin the evaporation process of alkali recovery based on the on-line adaptive errorcompensation LS-SVM.According to the conditions that the export Black liquorconsistency is difficult to measure in the evaporation process of alkali recovery,this paper presents the Online adaptive error compensation LS-SVM model.Through studying and analyzing the alkali recovery process and the principle ofmultiple-effect evaporator, it is known that Black liquor consistency in theevaporation process of alkali recovery is influenced by many factors such asinitial diluted concentration and flow of black liquor, total effective difference intemperature of multiple-effect evaporator, the major components, thetemperature of cooling water, ambient temperature, the heat transfer coefficientand so on. Black liquid evaporation usually consists of multiple-effectevaporators, and each evaporator consists of vapors, black liquor andcondensation water. The evaporation process has a serious non-linearperformance due to liquid column pulsation in evaporator’s internal pipelinesand oscillation phenomenon caused by temperature, concentration and flow ofblack liquor. Owing to long size and high capacity of evaporator’s internalpipeline, one end of system may take a long time to respond to physic quantumchanges of the other end, so the actual influencing factors of Black liquorconsistency is more complex. Present models of the soft measurement for Blackliquor consistency takes pressure and temperature as accessorial variables andexport Black liquor consistency as main variable while ignoring other factors. Asa result the accuracy of the model is not high. Residual adaptive compensationsystem established in this paper considers other factors. The paper improves theprecision of the model in two ways: firstly, the algorithm of on-line Least SquareSupport Vector Machine is adopted in model training and on-line correction tomake the model adapt the current work situation; secondly, the tache of residualadaptive compensation is added to make the result of soft measurement moreaccurate by on-line modifying method. The common Least Square SupportVector Machine algorithm is simple and has more calculation and high velocity, but it loses the inner sparsity and robustness of Support Vector Machinemodeling. If the algorithm is not been improved, it will be unusable in actual dueto increasing training examples, long running time and large memory occupancy.Online adaptive Least Square Support Vector Machine removes the mostlongstanding sample while adding a new sample, and corrects the model onlinein delta mode to reduce the arithmetic complexity and save computer memory.The residual compensation adaptive model is established with multiple linearregression method, and its adaptive efficiency method is built using the currentdata and discarding the earliest error. Simulation and application results showthat the soft-measurement model for export Black liquor consistency offers highaccuracy and it has been able to adapt to changes in working conditions.
     (3)The research of soft measurement model for Black liquor consistencyin washing process based on the fuzzy on-line adaptive LS-SVM.The research ofits soft measurement model is relatively less. The former models have poorgeneralization capability and low accuracy. In view of the shortcomings of theformer models, a new model is proposed based on the fuzzy on-line adaptiveLS-SVM. Online measurement for Black liquor consistency in washing processis difficult. The analysis results of the technology of washing process show thatBlack liquor consistency is influenced by pulp flow, pulp thickness, wash slurrywater, vacuum degree, drum speed and so on. Therefore, it is difficult to executethe real-time measurement by traditional methods. Considering the actualsituation of the factory, the modeling parameters are determined in the conditionwithout increasing the instrument or affecting the normal production. Throughthe on-site analysis and discussions with senior engineers, it is determined to useinlet Black liquor consistency, inlet Black liquor flow and water flow in washingprocess as the accessorial variables, and the first period Black liquor consistencyas the leading variable to set up the soft measurement model for the first periodBlack liquor consistency. The soft measurement model based on the on-site datacollection can only toughly reflect the substantial change trend of the industryobject, so there are inescapable errors existing in this model. When the model isput into use, the object characteristics and working point may change with timeprogressing because of its nonlinear, time-varying and uncertain properties, sothat the errors of parameter measurement may increase with the original data sampling of models. Online adaptive Least Square Support Vector Machine canadapt to the changes of the working conditions and correct the models online.Samples from washing process contain different levels of noise. In ordering toimprove the accuracy of the model, one sample point is signified by one degreeof Support Vector defined in order to show the weight of the sample points in themodel training process. The sample replacement strategy of the model is toreplace the sample point which is the minimum sample point of the Lagrangemultiplier absolute value with a new sample point. Simulation and applicationresults show that the soft-measurement model for Black liquor consistency issuitable for washing process.
     (4)In this thesis, checking the abnormal value based on SVM is discussedand studied. The support vector machine method can discover and solve theproblems in washing process such as the measuring instrument’s calibrationinaccuracies, no-timely, instrument malfunction, larger process interference,benchmark drift, operator’s mistakes, a periodically pipeline cleaning, theinstrument’s value volatility caused by the pipeline stopper, the data conflictcaused by data distortion and so on. The paper discusses the scheme which canfind the abnormal data in leading variables by using SVM regression algorithm,and it is verified by example, then it is used in select abnormal data of leadingvariables during washing process.
     The paper takes the soft measurement for Kappa number during pulping,Black liquor consistency in alkali recovery and washing process as researchsubjects, presents a series of algorithms based on Support Vector Machine toimprove the on-line measurement precision, and proves the validity of theproposed algorithm through simulation and application. The research results areuseful as reference for other industry soft-measurements.
引文
[1]刘焕彬.纸浆性质软测量原理与技术[M].北京:中国轻工业出版社,2009:1,21-29.
    [2]王孟效,胡晓刚,韩卿等.软测量技术在造纸工业中的应用[J].中国造纸,2005,24(11):46-49.
    [3]罗瑜.支持向量机在机器学习中的应用研究[D].成都:西南交通大学,2007.
    [4]Kruif B J,Vries T J A.On Using a Support Vector Machine in Learning Feed-ForwardControl[C].In Proceedings of Int. Conf. on Advanced Intelligent Mechatronics.Como,Italy,July2001:272-277.
    [5]Suykens J A K.Nonlinear Modeling and Support Vector Machines[C].In Proceedings ofthe18th IEEE Instrumentation and Measurement Technology Conference,2001,1:278-294.
    [6] Rychetsky M,Ortmann S,Glesner M.Support Vector Approaches for Engine KnockDetection[C].International Joint Conference on Neural Networks,1999,2:969-974.
    [7]张英.基于支持向量机的过程工业数据挖掘技术研究[D].杭州:浙江大学,2005.
    [8]孙宗海.支持向量机及其在控制中的应用研究[D].杭州:浙江大学,2003.
    [9]杨金芳.支持向量回归在预测控制中的应用研究[D].保定:华北电力大学,2007.
    [10]张健.提高制浆蒸煮过程纸浆Kappa值软测量精度的研究[D].广州:华南理工大学,2005.
    [11]黄细霞.基于支持向量机建模方法及在材料加工中的应用研究[D].上海:上海交通大学,2008.
    [12] Vapnik V N,张学工.统计学习理论的本质[M].北京:清华大学出版社,2000:3-15.
    [13]邓乃杰,田英杰.数据挖掘中的新方法[M].北京:科学出版社,2006:4-26.
    [14]王国胜.支持向量机理论与算法研究[D].北京:北京邮电大学,2007.
    [15]王磊.支持向量机学习算法的若干问题研究[D].成都:电子科技大学,2007.
    [16]于涛.基于混合模型的软测量方法研究及其在发酵过程中的应用[D].北京:北京化工大学,2006.
    [17]张煶.综合建模方法和先进控制技术在两个化工过程中的应用[D].南京:南京工业大学,2003.
    [18]李艳.制浆蒸煮过程纸浆卡伯值软测量技术研究与应用[D].广州:华南理工大学,2003.
    [19]Joseph B.,Brosilow C.B..Inferential control of process:part I.Steady state analysis anddesign[J].AIChE Journal,1978,24(3):485-492.
    [20] Brosilow C.B.,Tong M.. Inferential control of process: part II.The structure anddynamics of inferential control system[J].AIChE Journal,1978,24(3):492-500.
    [21] Joseph B.,Brosilow C.B..Inferential control of process:part III. Construction of optimaland suboptimal dynamic estimators[J].AIChE Journal,1978,24(3):500-509.
    [22]周凌柯.数据校正技术的研究及应用[D].杭州:浙江大学,2005.
    [23]李春富.基于数据的软测量建模方法及其应用的研究[D].北京:清华大学,2004.
    [24]李鸿儒,顾树生,邓长辉.递归神经网络的RPE算法及其在非线性动态系统建模中的应用[J].东北大学学报(自然科学版),2000,21(6):590-593.
    [25]彭云峰.污水处理出水水质软测量预测预报系统开发[D].昆明:昆明理工大学,2003.
    [26]李春富,王桂增,徐博文.聚丙烯熔融指数软测量[J].化工自动化及仪表,2002,29(5):22-25.
    [27]方辉,陈源梅.基于软测量的黑液波美度在线控制[J].计算机测量与控制,2006,14(10):1329-1330.
    [28]傅永峰.软测量建模方法研究及其工业应用[D].杭州:浙江大学,2007.
    [29] Zamprogna E., Barolo M.,Seborg D.C..Optimal selection of soft sensors inputs forbatch distillation columns using principal component analysis[J].Journal of ProcessControl,2005,15(1):39-52.
    [30]马朝阳,苏宏业,傅永峰.基于KPCA-SVR方法的复合肥养分含量建模[J].中国科技大学学报(增刊),2005,35:314-321.
    [31] Luo J.X.,Shao H.,Selecting secondary measurements for soft sensor modeling usingrough sets theory[C].Proceedings of the4thWorld Congress on Intelligent Control andAutomation,June2002,Shanghai,China,415-419.
    [32]刘瑞兰,陈渭泉,苏宏业.基于GA-PLS算法的最优辅助变量选择及其在软测量建模中的应用[J].南京邮电大学学报(自然科学版),2006,26(1):76-80.
    [33]仲蔚.软测量与先进控制策略研究及其在石油化工过程中的应用[D].广州:华南理工大学,1999.
    [34]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:化学工业出版社,2000:6-20.
    [35] Narasimhan S.,Mah R.S.H.,Generalized likelihood ratio method for gross erroridentification[J].AIChE Journal,1987,33(9):1514-1521.
    [36] Michael L.,Mark A.,Modeling chemical processes using prior knowledge and neuralnetworks[J].AIChE Journal,1994,40(8):1328-1340.
    [37]陈剑.数据校正理论及其应用研究[D].杭州:浙江大学,2000.
    [38]于静江,周春晖.过程控制中的软测量技术[J].控制理论与应用,1996,13(2):137-144.
    [39]刘瑞兰.软测量技术若干问题的研究及其工业应用[D].杭州:浙江大学,2004.
    [40] Zhou L.,Lu Y.Z..Modeling and control for nonlinear time-delay system via patternrecognition approach[C].Preprints of2ndIFAC workshop on artificial intelligence in realtime control,1989,Shenyang,China,7-12.
    [41] FengR.,ZhangY.J.,Zhang Y.Z. Drifting modeling method using weighed supportvecotor machines with application to soft sensor[J].Acta Automatica Sinica,2004,30(3):436-441.
    [42]李春富,zhang J,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.
    [43]乔弘.火电厂热工参数软测量关键技术和方法研究[D].北京:华北电力大学,2009.
    [44] Sarkar P,Gupta S K.Steady state simulation of continuous-flow stirred-tank slurrypropylene polymerization reactor.Polymer Engineering and Science[J].1992,32(11):732-742.
    [45]Sarkar P,Gupta S K.Dynamic simulation of propylene polymerization in continuous flowstirred tank reactors[J]. Polymer Engineering and Science,1993,33(6):368-374.
    [46]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,1994,25(5):23-28.
    [47] Sato C,Ohtani T,Nishitani H.Modeling simulation and nonlinear control of a gas-phasepolymerization process[J].Computers Chem.Engng,2000,24(2-7):945-951.
    [48]黄克瑾.精馏过程的模型化及仿真[D].杭州:浙江大学,1992.
    [49]单鸿亮,王文海,孙优贤.间隙蒸煮过程软测量建模综述[J].中国造纸学报,2003,18(2):195-200.
    [50] Tham M T,Montague G A,Morris A J.Soft-sensors for process estimation and inferentialcontrol[J].Process Control,1991,1(1):3-14.
    [51]Guilandoust M T, Morris A J, Tham M T. Adaptive estimation algorithm forinferentialcontrol[J].Ind.Eng.Chem.Res.,1988,27:1658-1664.
    [52] Quintero-Marmol E,Luyben W L,Georgakis C.Application of an extended Luenbergerobserver to the control of multicomponent batch distillation[J].Ind. Eng.Chem.Res.,1991,30(8):1870-1880.
    [53]仲蔚,刘爱伦,俞金寿.多变量系统的软测量建模研究[J].控制与决策2000,15(2):209-212.
    [54]熊智华,王雄,徐用懋.一种利用多神经网络结构建立非线性软测量模型的方法[J].控制与决策,2000,15(2):173-176.
    [55]阎威武,朱宏栋,邵惠鹤.基于最小二乘支持向量机软测量建模[J].系统仿真学报,2003,15(10):1494-1496.
    [56]徐晔,杜文莉,钱锋.基于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):3873-3876.
    [57]连承波,赵永军,李汉林.基于支持向量机回归的煤层含气量预测[J].西安科技大学学报,2008,28(4):707-710.
    [58]邵年华,沈冰,秦胜英.核主成分支持向量机模型在蒸发预测中的应用[J].北京师范大学学报(自然版),2010,46(3):163-165.
    [59]程跃,程文明,郑严.支持向量机在中药浓缩浓度软测量中的应用[J].计算机工程与应用,2010,46(5):240-242.
    [60]张淑宁.湿法冶金铜萃取组分含量软测量方法研究[D].沈阳:东北大学,2009.
    [61]李晓光.混合建模方法研究及其在化工过程中的应用[D].北京:北京化工大学,2008.
    [62]闫友彪,陈元琰.机器学习的主要策略综述[J].计算机应用研究,2004,(7):4-10.
    [63] Vapnik V.N..Statistical Learning Theory[M].Wiley,New York,1998.
    [64] Vapnik V.N.. The Nature of Statistical Learning Theory[M].Berlin,Springer,1995.
    [65]张国云.支持向量机算法及应用研究[D].湖南:湖南大学,2006.
    [66] Anthony M.Probabilistic Analysis of Learning in Artificial Neural Networks:The PACModel and Its Variants[J].Neural Computing Surveys,1997,1:1-47.
    [67]Vapnik V.N..An overview of statistical learning theory[J].IEEE Transactions on NeuralNetworks,1999,10(5):988-999.
    [68] Shawe-Taylor, J. Bartlett. Stuctural risk minimization over data dependenthierarchies[J].IEEE Transactions on Information Theory,1998,44(5):1926-1940.
    [69]唐发明.基于统计学习理论的支持向量机算法研究武汉[D].武汉:华中科技大学,2005.
    [70]曾志强.支持向量分类机的训练与简化算法研究[D].杭州:浙江大学,2007.
    [71]孙德山.支持向量机分类与回归方法研究[D].长沙:中南大学,2004.
    [72]张春华.支持向量机中最优化问题的研究[D].北京:中国农业大学,2004.
    [73]姜文翰.模式识别中的样本选择研究及其应用[D].南京:南京理工大学,2008.
    [74]陶卿,曹进德,孙德敏.基于支持向量机分类的回归方法[J].软件学报,2002,13(5):1024-1028.
    [75]K.work,T.J..Support Vector mixture for Classification and Regression Problems[J].Proceedings of Fourteeth International Conference on Pattern Recognition,1998,(1):255-258.
    [76]W.G.Vermeulen, P.J. vander Wolk,A.P.dewejie.Prediction of Jominy Hardness Profilesof Steels Using Artifieial Neural Newtorks[J].Jounral of Materials Engineering andPerofmrance,5(l):57-63,1996.
    [77]田英杰.支持向量回归机及其应用研究[D].北京:中国农业大学,2005.
    [78] S.S.Keerthi,C.J.Lin.Asymptotic behaviors of support vector machines with Gaussiankernel[J].Neural Computation,2003,15(7):1667-1689.
    [79]李烨,蔡云泽,许晓鸣.支持向量机在产品成分估计中的应用研究[J].自动化仪表,2006,27(4):8-11.
    [80]李妍妍.SVM理论及其在船舶机炉协调智能控制中的应用研究[D].哈尔滨:哈尔滨工程大学,2007.
    [81]王春林.大型电站锅炉配煤及燃烧优化的支持向量机建模及实验研究[D].杭州:浙江大学,2007.
    [82] Sch lkopf B,Smola A J,Williamson R C.New support vector algorithms[J].NeurslComputation,2000,12(5):1207-1245.
    [83] Y.J.Lee and O.L.Mangasarian,RSVM:Reduced support vector machines[C].InProceedings of the First SIAM International Conference on Data Mining,2001.
    [84] J.Suykens, J.Vandewalle.Least squares support vector machine classifiers[J].NeuralProceeding Letters,1999,9(3):293-300.
    [85]李忠伟.支持向量机学习算法研究[D].哈尔滨:哈尔滨工程大学,2006.
    [86]常群.支持向量机的核方法及其模型选择[D].哈尔滨:哈尔滨工业大学,2007.
    [87]范听炜.支持向量机算法的研究及其应用[D].杭州:浙江大学,2003.
    [88]张永.基于模糊支持向量机的多类分类算法研究[D].大连:大连理工大学,2008.
    [89]邝守敏.制浆工艺及设备[M].北京:中国轻工业出版社,2000.
    [90] E.W.马科隆,T.M.格雷斯(美)编,曹邦威译.最新碱法制浆技术[M].北京:中国轻工业出版社,1998.
    [91] Vroom K.E.The H factor:a means of expressing cooking times and temperatures as asingle variable[J].Pulp Paper Mag.Can.1957,58(C):228.
    [92]管永刚.碱法蒸煮实用技术[M].天津:天津大学,1992:1.
    [93]雷美,沈文浩.纸浆卡伯值在线测量的研究进展[J].中国造纸,2006,25(3):52-56.
    [94] Lin,C.P., Mao, M.Y. and Jane, C.Y., Tappi,1978,61(2):72.
    [95]李向阳.间歇蒸煮过程纸浆Kappa值软测量方法研究与应用[D].广州:华南理工大学,2001.
    [96]罗琪.硫酸盐法蒸煮过程纸浆Kappa值软测量技术的研究[D].广州:华南理工大学,1998.
    [97]巫淑萍,邱公伟.硫酸盐法间歇蒸煮化学浆的动力学数学模型及其计算机控制方案研讨[J].中国造纸,1988(2):35-40.
    [98] Ming Rao,Jean Corbin,Qun Wang.Soft sensors for quality prediction in batch chemicalpulping processes[C].Proceedings of the1993International Symposium on IntelligentControl,Chicago,Illinois USA,1993:150-155.
    [99] H.C.Kim,X.Shen,M.Rao,J.Zurcher.Quality prediction by neural network for pulp andpaper process[C].Conference on Electrical and Computer Engineering,Vancouver,Canada,1993,vol.1:104-107.
    [100] M.T.Musvi, C.Domnisoru.A Neuro-fuzzy system for prediction of pulp digesterK-number[C]. International Joint Conference on Neural Networks,Washington,USA,1999,Vol.6:4253-4258.
    [101] Belarbi K,Bettou K. Fuzzy Neural Networks for Estimation and Fuzzy ControllerDesign Simulation Study for a Pulp Batch Digester [J]. Journal of Process Control,10,2000:35.
    [102] Dani Juricic. Model based control of the Kappa number in the pulp cookingprocess[C].Electrotechnical Conference,Ljubljana,Slovenia,1991,Proceedings,6th Mediterranean,1991, vol.2:836-839.
    [103] Anjan Kumar Datta,Jay H.Lee.Model-based monitoring and control of batch pulpdigester[C].Proceedings of the American Control Conference,Baatlmore, Maryland,1994, vol.1:500-504.
    [104] A.K. Datta,J.H.Lee,G. A.Krishnagopalan.Reducing batch-to-batch variability of pulpquality through model-based estimation[J]. Pulp Paper Canada,1997,98:119-122.
    [105] Helena,Cristina,Aguiar. Neural Network and Hybrid Model:A Discussion aboutDifferent Modeling Techniques to Predict Pulping Degree with Industrial Date[J]. Chemical Engineering Science,2001,56:565.
    [106] Lee,Jay H.,A.K.Datta.Nonlinear inferential control of pulp digesters[J].AIChE J.,1994,40(1):50-64.
    [107]C.H.Wells, E.C.Johns,F.L.Chapman.Computer control of batch digesters using aKappa number model[J].Tappi Journel,1975,58(8):177-181.
    [108]Dani Juricic.Model based computer control of Kappa number in magnefite pulpcooking[C].Proceedings of the Third IEEE Conference on Control Applications,Glasgow,UK,1994, vol.2:775-780.
    [109] Philip A. Wisnewski, Francis J.Doyle.Model-based predictive control studies for acontinuous pulp digester[J].IEEE Transactions on Control Systems Technology.2001,9(3):435-444.
    [110]管永刚.落叶松KP法蒸煮数学模型参数的确定[J].中国造纸,1999(2):22-24.
    [111]管永刚.关于硫酸盐法蒸煮数学模型的若干问题[J].中国造纸.1997,16(2):55-58.
    [112]王海燕,宋光燕,杨静.蒸煮液有效碱浓度与纸浆硬度控制[J].纸和造纸,1999(2):37.
    [113]于玲,祝和云.神经元网络用于纸浆的高锰酸钾值的软测量[J].中国造纸学报,1997,13(sup):63-67.
    [114]郑启富.制浆蒸煮过程卡伯值测定的研究[J].中国造纸学报,2002,17(2):92.
    [115]李艳.李向阳.朱学峰.基于粗集理论和RBF神经网络的软测量技术研究[J].自动化学报,2000,6(Supplement B):168-172.
    [116]李海生.支持向量机回归算法与应用研究[D].广州:华南理工大学,2005.
    [117] Tamra S. Pattern classification based on fuzzy relations[J].IEEE Trans. SMC,1971,1(1):217-242.
    [118] Backer E and Jain A K.A clustering performance measure based on fuzzy setdecomposition[J].IEEE Trans.PAMI,1981,3(1):66-74.
    [119] Zkim Le. Fuzzy relation compositions and pattern recognition[J]. Inf Sci.1996,89,107-130.
    [120] Matula D W.Graph theoretic technigues for cluster analysis algorithms[C].AdvancedSeminar on Classification and Clustering, J Van Ryzin, Ed. New York, Academic,1997,95-129.
    [121] Bezdek J C,Harris J O.Convex decompositions of fuzzy ISO DATA clusteringalgorithm[J]. IEEE Tans.PAMI,1980,1(2):1-8
    [122] Esogbue A O. Optimal clustering of fuzzy data via fuzzy dynamicprogramming[J]. FSS,1986,26(1):127-130.
    [123] Mantaras R L D,Valverde L.New results in fuzzy clustering based on the concept ofindistinguishablity relation[J]. IEEE Trans. PAMI,1988,10(5):754-757.
    [124] Dunn J C.A fuzzy relative of the ISO DATA process and its use in detecting compactwell separated clusters[J].Cybernet,3,1974,32-57.
    [125] Bezdek J C. Pattern Recognition with Fuzzy Objective FunctionAlgorithms[M]. Plenum Press,New York,1981.
    [126] Syed N., Liu H., Sung K. Incremental Learning with Support vectorMachines[C]. Proceeding of IJCAI Conference,Sweden,August1999.
    [127] klinkenberg R., Joachims T. Detecting Concept Drift With Support VectorMachines[C].Proceedings of the17thInternational Conference on Machine Learning,Morgan Kaufmann,2000.
    [128] Cauwenberghs G.,Poggio T..Incremental and Decremental Support Vector MachineLearning.Advances in Neural Information Processing Systems,2000.
    [129]汪辉.增量型支持向量机回归训练算法及在控制中的应用[D].杭州:浙江大学,2006.
    [130]王平,田华阁,田学民等.一种基于增量式SVR学习的在线自适应建模方法[J].化工学报,2010,61(8):2040-2045.
    [131]BurgessT.L.. Kraft Recovery Operations Short Course Notes[J].TAPPI Press,Atlanta,1990,133-144.
    [132]R.P.格林,G.霍夫编,潘锡五译.碱法制浆化学药品的回收[M].北京:中国轻工业出版社,1998.
    [133]张志秀,刘星萍和张新荣.黑液波美度的在线软测量[J].中华纸业,2003,24(8):54-55.
    [134]张玲,郑恩让.黑液浓度的一种在线检测方法[J].中华纸业,2000,21(4):41.
    [135]李艳.基于造纸碱回收的黑液浓度软测量技术的研究[J].陕西科技大学学报,2003,21(6):90.
    [136]杨飚,张曾科.基于稳健关联向量回归的黑液波美度软测量[J].清华大学学报(自然科学版),2007,47(S2):1840-1843.
    [137]谭超.基于LS-SVM的黑液波美度软测量[J].中国造纸,2005,24(12):22-24.
    [138]陈伟,李茜,汤伟.一种高级算法在黑液波美度软测量中的应用研究[J].计算机测量与控制,2009,17(8):1520-1522.
    [139]李瑾,汤伟.模糊最小二乘支持向量机在黑液波美度软测量中的应用[J].航天制造技术,2008,4(2):51-53.
    [140]孙计.造纸碱回收过程先进控制研究[D].杭州:浙江大学,2004.
    [141]姜静清.最小二乘支持向量机算法研究及应用研究[D].长春:吉林大学,2007.
    [142]陶少辉.最小二乘支持向量机的改进及其在化学化工中的应用[D].杭州:浙江大学,2006.
    [143]李丽娟.最小二乘支持向量机建模及预测控制算法研究[D].杭州:浙江大学,2008.
    [144]王玲,薄列峰,刘芳.最小二乘隐空间支持向量机[J].计算机学报,2005,28(8):1302-1307.
    [145]张浩然,汪晓东.回归最小二乘支持向量机的增量和在线式学习算法[J].计算机学报,2006,29(3):400-406.
    [146]刘新旺,殷建平,张国敏.基于最小二乘支持向量机的特征增量学习算法[J].计算机工程与科学,2008,30(12):68-71.
    [147]杨滨,杨晓伟,黄岚.自适应迭代最小二乘支持向量机回归算法[J].电子学报,2010,38(7):1621-1625.
    [148]梅炽,胡志坤,彭小奇.基于神经网络和自适应残差补偿的炼铜转炉吹炼终点预报模型[J].中国有色金属,2000,10(5):732-735.
    [149]杨春节,孙优贤,鲍伯良.纸浆洗涤过程控制的现状和趋势[J].中华纸业,1998,3:25-28.
    [150]汤伟,单文娟,王孟效.残碱和黑液波美度的在线软测量方法及实现[J].中国造纸,2011,30(6):47-52.
    [151]郑恩让,张玲,钱丽.软测量技术在波美度测量中的应用[J].仪器仪表学报,2003,24(2):146-147.
    [152]宗大伟,王孟效.纸浆洗涤过程的DC S方案设计[J].中国造纸,2004,23(10):43.
    [153]杨春节.纸浆洗涤过程测控技术的研究进展[J].中国造纸,2002(6):56.
    [154]郑恩让,赵小梅,王孟效.基于神经网络的纸浆洗涤过程数学模型(1)[J].化工自动化及仪表,2000,27(4):6-9.
    [155]汤伟,施颂椒.纸浆洗涤过程双目标优化分布式控制系统[J].控制理论与应,2002,19(4):555-560.
    [156]Mika S,Sch lkopf B,wlliamson R.Kernal PCA and de-noising in feature spaces,Advances in Neural Information Processing Systems[M].MIT Press,Cambridfe,MA,1999,11:536-542.
    [157]Nguyen D D, Ho T B. A bottom-up method for simplifying support vectorsolutions[J].IEEE Transactions on Neural Networks,2006,17(3):792-796.
    [158]Downs T, Gates K E, Masters A. Exact simplification of support vectorsolutions[J].Journal of Machine Learning Research,2001,2:293-297.
    [159]邹伟,孙瑜,周海君.基于DDE机理的WinCC和Matlab网络通讯研究[J].自动化博览2005,(5):61-63.
    [160]李智宇,彭雪明,税爱社.WinCC与Matlab的软件接口方法研究[J].后勤工程学院学报,2011,27(6):92-96.
    [161]魏骏,张世峰,蒋一.MATLAB与WinCC的动态数据交换技术[J].工业控制计算机,2007,20(1):48-49.

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700