集成神经网络和多目标进化算法的卷烟产品参数优化设计方法及应用研究
|
摘要
随着烟草行业卷烟工艺水平的发展,计算机辅助卷烟产品设计系统越来越受到卷烟企业的重视。该系统在综合分析历史数据和经验知识的基础上,建立符合卷烟生产实际的各类模型,并以此为指导进行产品优化设计。本文以国内某大型卷烟企业在该领域的课题研究为背景,重点研究该企业计算机辅助产品设计系统中的核心模块——卷烟产品参数优化设计模块。在研究过程中发现,卷烟产品参数优化设计是一个十分复杂的黑盒多目标优化问题,主要反映在:
(1)优化对象“工艺参数”与优化目标“质量指标”之间的映射关系十分复杂,难以建立常规的数学优化模型,对于给定的一组工艺参数取值,只能通过现场实验才能获得准确的质量指标评价;
(2)优化目标由相互冲突的多个目标组成,这些目标在大多数情况下不能直接进行优劣关系的比较,目标之间相互冲突,在不降低某一目标性能的情况下不能通过参数优化任意提高其他目标的性能。
针对上述问题,本文提出了一种基于神经网络和多目标进化算法混合策略的集成计算智能方法:首先,利用人工神经网络对历史数据进行训练,获得能反应参数优化过程中参数向量空间到目标向量空间非线性映射关系的神经网络模型;其次,将训练好的神经网络模型嵌入到多目标进化算法中,以此作为进化过程中个体的适应度评价函数,使得多目标进化算法可以直接应用于产品参数优化设计过程。具体地,本文的主要研究内容和创新成果概括如下:
1、提出了一种集成神经网络和多目标进化算法的产品参数优化设计方法现实世界中,很多产品参数优化设计问题均可归纳为黑盒多目标优化问题。
黑盒多目标优化问题具有系统建模困难和多个目标必须协调优化的特点。本文通过结合跨越BP神经网络和改进的非劣解排序遗传算法(NSGA-II),借助不同智能计算方法的优点,互补不足:(1)可以充分利用跨越BP神经网络建模的优点,解决复杂系统建模困难的问题,并为NSGA-II的进化个体提供适应度评价函数;(2)采用NSGA-II解决复杂系统中的多目标优化问题。
2、提出了基于跨越BP算法的人工神经网络建模方法
复杂系统建模是成功解决产品设计参数优化问题的关键,然而传统BP算法具有收敛速度慢、网络结构选择困难、容易陷入局部极小等缺点。本文从连接方式、结构优化以及优化策略三个方面,对传统BP算法进行了改进:
(1)采用基于跨越连接的误差反向传播算法对网络进行训练。有跨越连接的神经网络摒弃了传统神经网络只有前后层相连的拓扑结构,能以更加简洁的结构逼近神经网络的理想状态,加快网络收敛速度。国防科学技术大学研究生院博士学位论文
(2)提出了一种BP神经网络结构优化算法。该算法通过引入有方向的均方误差,在有跨越连接的多层前馈人工神经网络结构方程式的基础上,分别导出隐层层数和隐层神经元数判别式。
(3)采用基于MOEA和BP混合算法的神经网络建模方法。由于要维持具有一定规模的群体,多目标进化算法必须同时处理搜索空间中的若干点而不像梯度下降法那样只处理单点,从而有助于搜索全局最优点,免予陷入局部最小。这样就可以避免传统BP人工神经网络采用梯度下降法所带来的缺点,同时也确保了良好的收敛速度。
3、提出了基于NN和MOEA的卷烟工艺参数优化设计方法
卷烟工艺设计主要分为打叶复烤工艺设计、制丝工艺设计和辅料配套工艺设计,它们的本质都是基于各类参数指标关系模型的多目标参数优化过程,且属于黑盒多目标优化范畴。因此,可采用集成神经网络和多目标进化算法的产品参数优化设计方法(ICIA–NN & MOEA)优化求解。在具体应用过程中,结合工艺设计实际提出了基于NN和MOEA的卷烟工艺参数优化设计方法,并将其应用于二次润叶工序工艺参数多目标优化问题,取得了令人满意的效果。
4、提出了基于NN和MOEA的卷烟配方参数优化设计方法
卷烟配方设计主要分为叶组配方设计和糖香料配方设计,它们的本质都是基于感官质量评价模型的多目标参数优化过程,且属于黑盒多目标优化范畴。与工艺参数优化设计不同的是,配方参数优化设计涉及到感官质量评价问题,这是一个主观性较强的评价过程,难以直接建立类似于工艺参数指标关系模型的单料烟比例与感官质量指标关系模型。针对上述问题,本文提出了基于NN和MOEA的卷烟配方参数优化设计方法,该方法与工艺参数多目标优化设计方法相比,主要有两点不同:(1)借鉴卷烟配方实践中的感官质量评分标准,将感官质量评价结果转换为感官质量得分,实现了非数值型指标向数值型指标的转变;(2)以烟叶化学成分为中间环节,分别建立单料烟比例与烟叶化学成分关系式和烟叶化学成分与感官质量得分关系模型,成功实现了由单料烟比例到感官质量评价的非线性映射。最后,将基于NN和MOEA的卷烟配方参数优化设计方法应用于配方创新和配方维护,取得了令人满意的效果。
With the development of the cigarette industry, the computer assistance design system of cigarette product obtains more attention by cigarette enterprises than before. This system would establish some effective cigarette models based on history data and experience, and has been taken to instruct the optimization of product design. This thesis mainly studies the parameter optimization module of computer assistance cigarette product design system. Because the cigarette parameter optimization design is an extremely complex black-box multi-objective optimization question, this paper proposed a kind of new optimization method based on a mixed strategy of neural network and multi-objective evolutionary algorithm. Specifically, the main contents and fruits of this thesis are outlined as follows:
1、Research on an integrated design approach based on neural network and multi-objective evolutionary algorithm in product parameter optimization.
In the real world, many product parameter optimization design questiones are the black-box multi-objective optimization question. The black-box multi-objective optimization question has the character that the system modelling is difficulty and many goals must be optimized coordinate. This thesis integrates the BP neural network and NSGA-II. First, we would make use of artificial neural networks to train historical data and establish a neural network model that can respond the non-linearity mapping relations of the parameter vector space to the goal vector space; next, in the processing of parameter optimization, we would make use of the neural network model to obtain individual fitness.
2、Research on a modeling method of artificial neural networks based on the cross connection BP algorithm.
The complex system modelling is the key to solve product design parameter optimization question. However, the traditional BP algorithm has some shortcomings, which includes the slow convergence rate, the network-architecture-choosing difficulty, falling into partial minimum easily and so on. This thesis has made some improvement to the traditional BP algorithmfrom in the above three aspects.
First, Neural networks with any kind of connections can always be sorted as cross-connected ones. According to traditional multi-layer feed-forward neural network, we elaborated the concept of completely-fully connected neural network and then put forward a cross-connected multi-layer feed-forward neural network algorithm. It can be theoretical proved that the cross-connected neural network can reach ideal results with more concise framework compareing with the non-cross connected neural network.
Next, it is difficult for us to choose the neural network structure. On the basis of the network structure equation of multi-layer feed-forward neural network with cross connection, discriminants of quantity of hidden layers and discriminants of quantity of perceptrons each layer are given. According to the discriminants, a new neural network structure optimization algorithm is proposed.
Third, a novel approach, combining MOEA with BP, is presented to evolve the neural network. The multi-objective evolution algorithm can work in search space simultaneous by certain scale population, not like gradient method which only deal with one point, thus is helpful in searching the overall optimum point and ensuring the good convergence rate.
3、An approach of cigarette process parameter optimization based on neural network and multi-objective evolutionary algorithm
According to the cigarette craft practice, the parameter optimization model in cigarette product process design is given, and belongs to the black-box multi-objective optimization question. Based on neural network and multi-objective evolutionary algorithm this thesis introduced the integrated design approach to solve this problem, and proposed a approach of cigarette process parameter optimization. This approach was used to ordering-cylinder, and obtained satisfactory effect.
4、An approach of cigarette formulation parameter optimization based on neural network and multi-objective evolutionary algorithm
The cigarette formulation parameter optimization is different with the process parameter optimization question for organoleptic character. The appraisal of organoleptic character is an extremely complex process, and it is difficult to get the relation model directly. Also, the organoleptic character belongs to the non-value index; the appraisal result of organoleptic character cannot use in the neural network modeling directly. In view of the above difficulties, this thesis proposed a parameter optimization method in cigarette product design with organoleptic character.
First, on the basis of the organoleptic character grading standard in cigarette formulation practice, we could transform the organoleptic character appraisal result into the organoleptic character score, realized the non-value index to the value index transformation. Next, we take tobacco leaf chemical composition as the middle link, separately established the relational model of tobacco proportion with chemical composition and that of chemical composition with organoleptic character score. Then the relational model of tobacco proportion with organoleptic character was given. In this foundation, according to the common parameter optimization method of cigarette product parameter design, we proposed the cigarette formulation parameter optimization method based on neural network and multi-objective evolutionary algorithm. Finally, this approach was used to the formulation design and the formulation maintenance successfully.
(1)优化对象“工艺参数”与优化目标“质量指标”之间的映射关系十分复杂,难以建立常规的数学优化模型,对于给定的一组工艺参数取值,只能通过现场实验才能获得准确的质量指标评价;
(2)优化目标由相互冲突的多个目标组成,这些目标在大多数情况下不能直接进行优劣关系的比较,目标之间相互冲突,在不降低某一目标性能的情况下不能通过参数优化任意提高其他目标的性能。
针对上述问题,本文提出了一种基于神经网络和多目标进化算法混合策略的集成计算智能方法:首先,利用人工神经网络对历史数据进行训练,获得能反应参数优化过程中参数向量空间到目标向量空间非线性映射关系的神经网络模型;其次,将训练好的神经网络模型嵌入到多目标进化算法中,以此作为进化过程中个体的适应度评价函数,使得多目标进化算法可以直接应用于产品参数优化设计过程。具体地,本文的主要研究内容和创新成果概括如下:
1、提出了一种集成神经网络和多目标进化算法的产品参数优化设计方法现实世界中,很多产品参数优化设计问题均可归纳为黑盒多目标优化问题。
黑盒多目标优化问题具有系统建模困难和多个目标必须协调优化的特点。本文通过结合跨越BP神经网络和改进的非劣解排序遗传算法(NSGA-II),借助不同智能计算方法的优点,互补不足:(1)可以充分利用跨越BP神经网络建模的优点,解决复杂系统建模困难的问题,并为NSGA-II的进化个体提供适应度评价函数;(2)采用NSGA-II解决复杂系统中的多目标优化问题。
2、提出了基于跨越BP算法的人工神经网络建模方法
复杂系统建模是成功解决产品设计参数优化问题的关键,然而传统BP算法具有收敛速度慢、网络结构选择困难、容易陷入局部极小等缺点。本文从连接方式、结构优化以及优化策略三个方面,对传统BP算法进行了改进:
(1)采用基于跨越连接的误差反向传播算法对网络进行训练。有跨越连接的神经网络摒弃了传统神经网络只有前后层相连的拓扑结构,能以更加简洁的结构逼近神经网络的理想状态,加快网络收敛速度。国防科学技术大学研究生院博士学位论文
(2)提出了一种BP神经网络结构优化算法。该算法通过引入有方向的均方误差,在有跨越连接的多层前馈人工神经网络结构方程式的基础上,分别导出隐层层数和隐层神经元数判别式。
(3)采用基于MOEA和BP混合算法的神经网络建模方法。由于要维持具有一定规模的群体,多目标进化算法必须同时处理搜索空间中的若干点而不像梯度下降法那样只处理单点,从而有助于搜索全局最优点,免予陷入局部最小。这样就可以避免传统BP人工神经网络采用梯度下降法所带来的缺点,同时也确保了良好的收敛速度。
3、提出了基于NN和MOEA的卷烟工艺参数优化设计方法
卷烟工艺设计主要分为打叶复烤工艺设计、制丝工艺设计和辅料配套工艺设计,它们的本质都是基于各类参数指标关系模型的多目标参数优化过程,且属于黑盒多目标优化范畴。因此,可采用集成神经网络和多目标进化算法的产品参数优化设计方法(ICIA–NN & MOEA)优化求解。在具体应用过程中,结合工艺设计实际提出了基于NN和MOEA的卷烟工艺参数优化设计方法,并将其应用于二次润叶工序工艺参数多目标优化问题,取得了令人满意的效果。
4、提出了基于NN和MOEA的卷烟配方参数优化设计方法
卷烟配方设计主要分为叶组配方设计和糖香料配方设计,它们的本质都是基于感官质量评价模型的多目标参数优化过程,且属于黑盒多目标优化范畴。与工艺参数优化设计不同的是,配方参数优化设计涉及到感官质量评价问题,这是一个主观性较强的评价过程,难以直接建立类似于工艺参数指标关系模型的单料烟比例与感官质量指标关系模型。针对上述问题,本文提出了基于NN和MOEA的卷烟配方参数优化设计方法,该方法与工艺参数多目标优化设计方法相比,主要有两点不同:(1)借鉴卷烟配方实践中的感官质量评分标准,将感官质量评价结果转换为感官质量得分,实现了非数值型指标向数值型指标的转变;(2)以烟叶化学成分为中间环节,分别建立单料烟比例与烟叶化学成分关系式和烟叶化学成分与感官质量得分关系模型,成功实现了由单料烟比例到感官质量评价的非线性映射。最后,将基于NN和MOEA的卷烟配方参数优化设计方法应用于配方创新和配方维护,取得了令人满意的效果。
With the development of the cigarette industry, the computer assistance design system of cigarette product obtains more attention by cigarette enterprises than before. This system would establish some effective cigarette models based on history data and experience, and has been taken to instruct the optimization of product design. This thesis mainly studies the parameter optimization module of computer assistance cigarette product design system. Because the cigarette parameter optimization design is an extremely complex black-box multi-objective optimization question, this paper proposed a kind of new optimization method based on a mixed strategy of neural network and multi-objective evolutionary algorithm. Specifically, the main contents and fruits of this thesis are outlined as follows:
1、Research on an integrated design approach based on neural network and multi-objective evolutionary algorithm in product parameter optimization.
In the real world, many product parameter optimization design questiones are the black-box multi-objective optimization question. The black-box multi-objective optimization question has the character that the system modelling is difficulty and many goals must be optimized coordinate. This thesis integrates the BP neural network and NSGA-II. First, we would make use of artificial neural networks to train historical data and establish a neural network model that can respond the non-linearity mapping relations of the parameter vector space to the goal vector space; next, in the processing of parameter optimization, we would make use of the neural network model to obtain individual fitness.
2、Research on a modeling method of artificial neural networks based on the cross connection BP algorithm.
The complex system modelling is the key to solve product design parameter optimization question. However, the traditional BP algorithm has some shortcomings, which includes the slow convergence rate, the network-architecture-choosing difficulty, falling into partial minimum easily and so on. This thesis has made some improvement to the traditional BP algorithmfrom in the above three aspects.
First, Neural networks with any kind of connections can always be sorted as cross-connected ones. According to traditional multi-layer feed-forward neural network, we elaborated the concept of completely-fully connected neural network and then put forward a cross-connected multi-layer feed-forward neural network algorithm. It can be theoretical proved that the cross-connected neural network can reach ideal results with more concise framework compareing with the non-cross connected neural network.
Next, it is difficult for us to choose the neural network structure. On the basis of the network structure equation of multi-layer feed-forward neural network with cross connection, discriminants of quantity of hidden layers and discriminants of quantity of perceptrons each layer are given. According to the discriminants, a new neural network structure optimization algorithm is proposed.
Third, a novel approach, combining MOEA with BP, is presented to evolve the neural network. The multi-objective evolution algorithm can work in search space simultaneous by certain scale population, not like gradient method which only deal with one point, thus is helpful in searching the overall optimum point and ensuring the good convergence rate.
3、An approach of cigarette process parameter optimization based on neural network and multi-objective evolutionary algorithm
According to the cigarette craft practice, the parameter optimization model in cigarette product process design is given, and belongs to the black-box multi-objective optimization question. Based on neural network and multi-objective evolutionary algorithm this thesis introduced the integrated design approach to solve this problem, and proposed a approach of cigarette process parameter optimization. This approach was used to ordering-cylinder, and obtained satisfactory effect.
4、An approach of cigarette formulation parameter optimization based on neural network and multi-objective evolutionary algorithm
The cigarette formulation parameter optimization is different with the process parameter optimization question for organoleptic character. The appraisal of organoleptic character is an extremely complex process, and it is difficult to get the relation model directly. Also, the organoleptic character belongs to the non-value index; the appraisal result of organoleptic character cannot use in the neural network modeling directly. In view of the above difficulties, this thesis proposed a parameter optimization method in cigarette product design with organoleptic character.
First, on the basis of the organoleptic character grading standard in cigarette formulation practice, we could transform the organoleptic character appraisal result into the organoleptic character score, realized the non-value index to the value index transformation. Next, we take tobacco leaf chemical composition as the middle link, separately established the relational model of tobacco proportion with chemical composition and that of chemical composition with organoleptic character score. Then the relational model of tobacco proportion with organoleptic character was given. In this foundation, according to the common parameter optimization method of cigarette product parameter design, we proposed the cigarette formulation parameter optimization method based on neural network and multi-objective evolutionary algorithm. Finally, this approach was used to the formulation design and the formulation maintenance successfully.
引文
[1]刘铮,郝亚琳. 2007年中国烟草行业实现工商税利3880亿元[J]. 2008, (2008年01月14日)
[2]国家烟草专卖局.中国卷烟科技发展纲要[J]. 2003
[3]栗和平,朱晓琴. OGr19Ni9中板热处理工艺优化试验[J].山西冶金, 2000, 77 (2)
[4]叶健松,李勇军,潘健生.大型支承辊热处理过程的数值模拟[J].机械工程材料, 2002, 26 (6)
[5]张丽萍.均匀设计和最优化方法在热处理中的应用[J].航天工艺, 1994, (4)
[6]喻云水.湿状态下竹胶合板模板力学性能与数值模拟研究[D].长沙:中南林业科技大学, 2006
[7] J. B H. Optimization through evolutionary and recombination [M]. Washington D C: Spartan, 1962
[8] M F R. A learning machine:Part 1 [J]. IBM Journal, 1958, 2 (1): 2-13
[9] M F R, B D, H N J. A learning machine: Part 2 [J]. IBM Journal, 1958, 3 (7): 282-287
[10] P B J E. Evolutionary operation: a method for increasing industrial productivity [J]. Appl Statistics, 1957, 6 (2): 81-101
[11] H H J. Outline for a logical theory of adaptive systems [J]. J Assoc Compute Mach, 1962, (3): 297-314
[12] I R. Cybernetic Solution Path of an Experimental Problem [M]. UK:Farnborough, 1965
[13] P S H. Pojekt MHD-Staustrahlrohr: Experimentelle Optimierung einer Zweiphasenduse [M]. Berlin: Germany: AEG Forschungsistitut, 1968
[14] J F L. Autonomous automata [J]. Industry Research, 1962, (4): 14-19
[15]梁瑞鑫.基于混沌理论和人工免疫理论优化方法的研究——及其在高炉操作参数优化中的应用[D].北京:北京科技大学, 2002
[16]费洪晓,黄勤径,戴弋.基于SVM与遗传算法的燃煤锅炉燃烧多目标优化系统[J].计算机应用研究, 2008, 25 (3)
[17]刘志新.高速铣削过程动力学建模及其物理仿真研究[D].天津:天津大学, 2006
[18]李清.虚拟数控铣床加工过程仿真系统及相关技术的研究[D].天津:天津大学, 2004
[19]任迪峰.中药材干燥过程中质量退化及优化干燥工艺的研究[D].北京:中国农业大学, 2002
[20] Chankong V, Haimes Y Y. Multiobjective decision making theory and methodology [M]. New York: North-Holland, 1983
[21] Hans A E. Multicriteria optimization for highly accurate systems [M]. New York: plenum press, 1988
[22] L C J. Multiobjective Programming and Planing [M]. New York: Academic Press, 1978
[23] E S R. Multiple Criteria Optimization: Theory, Computation, and Application [M]. New York:: Wiley, 1986
[24] N.Srinivas, K.Deb. Multiobjective optimization using nondominated sorting in genetic algorithms [J]. Evolutionary Computation, 1994, 2 (3): 221-248
[25]崔逊学.基于多目标优化的进化算法研究[D].合肥:中国科学技术大学, 2001
[26] A I F. Basic Artificial Intelligence Application [J]. Materials Evaluation, 2000, 58 (1): 33-34
[27] Sang-Hui P, Seok-Pil L. EMG Pattern Recognition Based on Artificial Intelligence Techniques [J]. IEEE Trans on Rehabilitation Engineering, 1998, 6 (4): 400-405
[28] K.Deb. Multiobjective Optimization Using Evolutionary Algorithms [M]. Chichester, U.K.: 2001
[29] C.A.Coello, Veldhuizen D A V, G.B.Lamont. Evolutionary Algorithms for Solving Multi-Objective Problems [M]. Norwell: MA:Kluwer, 2002
[30] RS.Rosenberg. Simulation of Genetic Populations with Biochemical Properties [D]. Michigan: University of Michigan, 1967
[31] Schaffer J D. Some experiments in machine learning using vector evaluated genetic algorithms [D]. Tennessee: Vanderbilt University Electrical Engineering, 1984
[32] Goldberg D E, Richardson J. Proc. Second Int. Conf. on Genetic Algorithms [C]. Lawrence Erlbaum: 1987
[33] Goldberg. Genetic Algorithms in Search, Optimization and Machine Learning [M]. Massachusetts: Addison Wesley, 1989
[34] Fonseca C M, Fleming P J. Genetic Algorithm: Proceedings of the Fifth International Conference [C]. San Mateo: 1993
[35] Fonseca C M, Fleming P J. An overview of evolutionary algorithm in multiobjective optimization [J]. Evolutionary Computation, 1995, 3 (1): 1-16
[36] Fourman M P. Genetic Algorithm: Proceedings of the First International Conference [C]. Grefenstette: 1985
[37] Horn J, Nafpliotis N. Proceedings of the First IEEE Conference on Evolutionary Computation,IEEE World Congress on Computational Intelligence [C]. Piscataway,NJ: IEEE Service Center, 1994
[38] Zitzler E, Thiele L. Multiobjective Evolutionary Algorithms: A Comparative Case Study and the Strength Pareto Approach [J]. IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 1999, 3 (4): 257-271
[39] Zitzler E, Laumanns M, Thiele L. SPEA2: Improving the Strength Pareto Evolutionary Algorithm [J]. 2001
[40] Knowles J, Corne D. Proceedings of the 1999 Congress on Evolutionary Computation [C]. Piscataway,NJ: IEEE Press, 1999
[41] Deb K, Pratap A, Agarwal S, et al. A Fast and Elitist Multiobjective Genetic Algorithm: NSGA-II [J]. IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2002, 6 182-197
[42] M.Erickson, A.Mayer, J.Horn. First International Conference on Evolutionary Multi-Criterion Optimization [C]. Springer-Verlag: Lecture Notes in Computer Science, 2001
[43] D.Corne, J.Knowles, M.Oates. Proceedings of the Parallel problem Solving from NatureⅥConference [C]. Paris: Lecture Notes in Computer Science, 2000
[44]张志刚. 570kg/h CO2叶丝膨胀线工艺技术系统研究[J].中国烟草科学, 2004, (1): 14-17
[45]熊安言,王镇增,张志刚. COMAS烘丝机工艺参数与叶丝质量的测试分析[J].烟草科技, 2003, (8): 6-8
[46]张大波,刘志平,李跃锋. HXD运行参数与工作段气流初始温度相关性的回归分析[J].烟草科技, 2004, (9): 12-15
[47]马宇平. HXD在线膨胀工艺参数和膨胀率与卷烟质量的关系[J].烟草科技, 2004, (7): 4-7
[48]席年生,张大波,李跃锋. HXD蒸汽喷射量对叶丝膨胀效果及卷烟内在质量的影响[J].烟草科技, 2005, (3): 3-6
[49]张万祥,戴永生,姚文祥.风选工艺参数对梗丝结构的影响[J].烟草科技, 2004, (1): 10-11
[50]叶春文,杨明权,王兵.烘丝工艺参数对卷烟感官质量的影响[J].烟草科技, 2005, (11): 7-9
[51]周晖.卷烟厂的气力输送技术应用参数设计优化[D].南昌:南昌大学, 2005
[52]潘高伟,王川,甘学文.加料工序的不同工艺条件对烟草香味成分含量变化的影响研究[J].郑州轻工业学院学报(自然科学版), 2007, 22 (6): 43-48
[53]郑光勇,李斌,叶为全.基于RBF神经网络的制丝生产线仿真模型[J].计算机工程与应用, 2005, (21)
[54]陈颀,张云生,张钟录.烟叶复烤机智能控制系统[J].昆明理工大学学报, 2001, 26 (3)
[55]罗海燕,方文青,谢鑫.打叶质量与出片率的关系[J].烟草科技, 2005, (1)
[56]尹献忠,李晓,李显红.降低烟支密度和提高滤嘴吸阻在降焦中的应用[J].烟草科技, 2002, (8)
[57]郭俊成,程晓蕾.白肋烟料剂及处理工艺参数的优化设计[J].中国烟草科学, 1997, (4): 33-36
[58]张敏,童亿刚,戴志渊. SPC技术在制丝质量管理中的初步应用[J].烟草科技, 2004, (9): 10-11
[59]席年生,申玉军,储国海.不同施压条件下梗丝填充值与其膨胀效果的相关性[J].烟草科技, 2003, (9): 3-6
[60]王兵,杨达辉,林平.新的卷烟生产工序品质评价方法的建立[J].烟草科技, 2002, (11): 16-18
[61]陈洪,钱强,杨艳萍.润叶后不同储叶时间与烟叶品质变化关系的研究[J].烟草科技, 2002, (7)
[62]上海派克软件有限公司.上海卷烟厂制丝线质量分析控制系统[J].
[63]王德吉.复烤机智能控制系统研究[D].郑州:郑州大学, 2003
[64]简辉,杨学良,王保兴.复烤温度对烟叶化学成分及感官质量的影响[J].烟草科技, 2006, (2)
[65]李跃锋,姜焕元,刘志平.烟叶温度和含水率与打叶质量的关系[J].烟草科技, 2005, (2)
[66]李斌,于川芳,杨述元.卷烟材料对烟气特征的预测模型[J].烟草科技, 2005, (9)
[67]范黎,苗芊,赵航.卷烟吸阻和滤棒压降测量值与大气压力的关系[J].烟草科技, 2004, (12)
[68]赵同林,李兵役,田兴友.烟支密度与烟支重量、吸阻、硬度及标准偏差的关系[J].烟草科技, 2005, (4)
[69]于川芳,罗登山,王芳.卷烟“三纸一棒”对烟气特征及感官质量的影响(一) [J].中国烟草学报, 2001, 7 (2)
[70]于川芳,罗登山,王芳.卷烟“三纸一棒”对烟气特征及感官质量的影响(二) [J].中国烟草学报, 2001, 7 (3)
[71]彭黔荣.烟叶的化学成分与烟叶质量的人工神经网络预测[D].成都:四川大学, 2004
[72]何琴,高建华,刘伟.广义回归神经网络在烤烟内在质量分析中的应用[J].安徽农业大学学报, 2005, 32 (3): 406-410
[73]高大启,吴守一.并联神经网络在烤烟内在品质评定中的运用[J].农业机械学报, 1999, 30 (1): 58-62
[74]张志刚,王二彬,苏东赢.卷烟常规化学成分与焦油的线性回归分析[J].烟草科技, 2003, (11)
[75]于川芳,卢斌斌,牟定荣.卷烟劲头与其烟丝、烟气主要化学成分的相关性[J].烟草科技, 2006, (9)
[76]于建军,章新军,毕庆文.烤烟烟叶理化特性对烟气烟碱、CO、焦油量的影响[J].中国烟草科学, 2003, (3): 5-8
[77]刘丁伟,胡建军,熊燕.总粒相物与烟支重量、吸阻和抽吸口数的相关分析、通径分析[J].烟草科技, 2005, (8)
[78]陈景云,胡建军.烟叶化学成分-品质综合评价物元模型的建立与应用[J].烟草科技, 2003, (10)
[79]国防科技大学信息系统与管理学院.长沙卷烟厂企业产品管理信息综合服务系统需求分析书[J]. 2004
[80]国防科技大学信息系统与管理学院.长沙卷烟厂业务调查报告[J]. 2004
[81]毛多斌,马宇平,梅业安.卷烟配方和香精香料[M].北京:化学工业出版社, 2001
[82]中华人民共和国国家标准.卷烟[J]. GB5606-1996, 1996
[83] Heckman R A, Dube M F, Lynm D, et al. The role of tobacco leaf precursors in cigarette flavor [J]. Rec.Adv.Tob.Sci., 1981, (7): 107-153
[84]何新贵,梁久祯.利用目标函数梯度的遗传算法[J].软件学报, 2001, 12 (7): 981-986
[85]杨朋林,贺新.人工神经网络与遗传算法结合的研究[J].现代电子技术, 2002, (143): 105-107
[86]张克进,徐敏,俞金寿. GA-ANN算法在产品质量估计中的应用[J].华东理工大学学报, 2000, 26 (52): 512-516
[87] Kesbeng W, L G H. A hybrid intelligent method for modeling the EDM process [J]. International Journal of Machine Tools & Manufacture, 2003, 43 (10): 995-999
[88]刘建萍.制造智能技术新进展[J].新技术新工艺, 2001, (7): 2-4
[89]殷勇,吴守一,高大启.基于遗传算法的卷烟质量评定神经网络模型[J].农业机械学报, 1999, 30 (3): 71-75
[90]王清,马广富,弥曼.一种基于遗传算法的神经网络控制方法研究[J].系统仿真学报, 2006, 18 (4): 1070-1072
[91]王科俊,王克成.神经网络建模、预报与控制[M].哈尔滨:哈尔滨工程大学出版社, 1996
[92] D J N, Gupta, S S R. Comparing back-propagation with a genetic algorithm for neural network training [J]. Omega, 1999, (27): 679-684
[93] F C D, T R C, L M R. Combining a neural network with a genetic algorithm for process parameter optimization [J]. Engineering Application of Artificial intelligence, 2000, (3): 391-396
[94] T.Hagan M, B.Demuth H, H.Beale M.神经网络设计[M].北京:机械工业出版社, 2002
[95] [俄]加卢什金.神经网络理论[M].北京:清华大学出版社, 2002
[96]吴佑寿,赵明生.激活函数可调的神经元模型及其有监督学习与应用[J].中国科学(E辑), 2001, 31 (3): 263-272
[97]刘耦耕,李圣清,肖强晖.多层前馈人工神经网络结构研究[J].湖南师范大学自然科学学报, 2004, 27 (1): 26-30
[98] C C F. Back-propagation neural networks for nonlinear self-tuning adaptive control [J]. IEEE Control system Magazine, 1990, (4): 44-48
[99] S N K, K P. Identification and control for dynamic systems using neural networks [J]. IEEE Trans. on Neural Networks, 1990, 1 (1): 4-27
[100] A B S, F V W S. Correlation based model validity tests for nonlinear models [J]. Int J control, 1986, (44): 235-244
[101] M B C. Neural networks for pattern recognition [M]. Oxford. England: Oxford University Press, 1996
[102] Iizaka T, Matsui T, Fukuyama Y. A Novel Daily Peak Load Forecasting Method using Analyzable Structured Neural Network [J]. IEEE T&D Asia, Yokohama, 2002, 1-6
[103] B W. Neural networks application in industry, business and science [J]. Communication of the ACM, 1994, (37): 93-105
[104]宋锐,张静,夏胜平.一种基于BP神经网络群的自适应分类方法及其应用[J].电子学报, 2001, (12)
[105] D G. Effective back-propagation training with variable step size [J]. Neural Networks, 1997, 10 (1): 69-82
[106] D S. Methods to speed up error BP learning algorithm [J]. ACM Computing Survey, 1995, (27): 519-592
[107] S Y. Global optimization for NN training [J]. IEEE Computer, 1996, (3): 45-54
[108]崔光照,毕娟,许进.前向神经网络结构优化的研究进展[J].郑州轻工业学院学报(自然科学版), 2003, 18 (3): 11-13
[109]王正志,薄涛.进化计算[M].长沙:国防科技大学出版社, 2000
[110]孙达.卷烟制丝工序加料均匀性检测项目设计[D].上海:上海交通大学, 2004
[111]戴志渊. SAS统计分析系统在上海卷烟厂制丝车间质量管理中的应用[D].上海:复旦大学, 2000
[112]孟冬玲.烟草和烟气中酚类物质的分析方法研究及应用[D].昆明:昆明理工大学, 2004
[113] R B A, L B R. In Computer Science and Statistics: Proc. of the 20st Interface [C]. 1988
[114] R M H, G I A. Inductive learning algorithms for complex systems modeling [M]. Boca Raton, London, Tokyo: CRC Press Inc, 1994
[115]岳超源.决策理论与方法[M].北京:科学出版社, 2003
[116]汪祖柱,程家兴.一种混合交叉策略的多目标演化算法及其性能分析[J].系统仿真学报, 2005, 17 (10)
[117]李桂琴,袁庆丰,王克胜, et al.制造过程多目标优化的集成计算智能方法[J].计算机集成制造系统, 2006, 12 (12): 2039-2043
[118]刘宝碇,赵瑞清.随机规划与模糊规划[M].北京:清华大学出版社,1998
[119]魏发远,李世其.基于混合遗传算法的多目标动态优化设计[J].宇航学报, 2004, 25 (6): 609-615
[120] S S V, K J V, D K B. Solution of constrained optimization problems by multi-objective genetic algorithm [J]. Computers and Chemical Engineering, 2002, 26 (10): 1481-1492
[121]崔逊学.多目标进化算法及其应用[M].北京:国防工业出版社, 2006
[122]胡开文.烟叶打叶复烤工艺与设备[M].北京:化学工业出版社, 2002
[123]王允白.烟叶主要化学成分与评吸香味关系研究[D].北京:中国农业科学院, 1996
[124]高家合,秦西云,谭仲夏.烟叶主要化学成分对评吸质量的影响[J].山地农业生物学报, 2004, 23 (6): 497-501
[125]闫克玉,王建民,屈剑波.河南烤烟评吸质量与主要理化指标的相关分析[J].烟草科技, 2001, (10): 5-9
[126]彭黔荣,蔡元青,王东山.根据常规化学指标识别烟叶品质的BP神经网络模型[J].中国烟草学报, 2005, 11 (5)
[127]谭梅,赵葵银,陈强.基于神经网络的卷烟工艺风力系统的压力控制[J].自动化与仪表, 2005, (4)
[128]殷勇,吴守一.遗传RBF神经网络在卷烟香气质量评定中的应用[J].农业机械学报, 2001, 17 (6): 147-149
[129]王强.基于支持向量机的卷烟叶组配方设计数据挖掘预测模型及应用研究[D].长沙:国防科学技术大学, 2006
[2]国家烟草专卖局.中国卷烟科技发展纲要[J]. 2003
[3]栗和平,朱晓琴. OGr19Ni9中板热处理工艺优化试验[J].山西冶金, 2000, 77 (2)
[4]叶健松,李勇军,潘健生.大型支承辊热处理过程的数值模拟[J].机械工程材料, 2002, 26 (6)
[5]张丽萍.均匀设计和最优化方法在热处理中的应用[J].航天工艺, 1994, (4)
[6]喻云水.湿状态下竹胶合板模板力学性能与数值模拟研究[D].长沙:中南林业科技大学, 2006
[7] J. B H. Optimization through evolutionary and recombination [M]. Washington D C: Spartan, 1962
[8] M F R. A learning machine:Part 1 [J]. IBM Journal, 1958, 2 (1): 2-13
[9] M F R, B D, H N J. A learning machine: Part 2 [J]. IBM Journal, 1958, 3 (7): 282-287
[10] P B J E. Evolutionary operation: a method for increasing industrial productivity [J]. Appl Statistics, 1957, 6 (2): 81-101
[11] H H J. Outline for a logical theory of adaptive systems [J]. J Assoc Compute Mach, 1962, (3): 297-314
[12] I R. Cybernetic Solution Path of an Experimental Problem [M]. UK:Farnborough, 1965
[13] P S H. Pojekt MHD-Staustrahlrohr: Experimentelle Optimierung einer Zweiphasenduse [M]. Berlin: Germany: AEG Forschungsistitut, 1968
[14] J F L. Autonomous automata [J]. Industry Research, 1962, (4): 14-19
[15]梁瑞鑫.基于混沌理论和人工免疫理论优化方法的研究——及其在高炉操作参数优化中的应用[D].北京:北京科技大学, 2002
[16]费洪晓,黄勤径,戴弋.基于SVM与遗传算法的燃煤锅炉燃烧多目标优化系统[J].计算机应用研究, 2008, 25 (3)
[17]刘志新.高速铣削过程动力学建模及其物理仿真研究[D].天津:天津大学, 2006
[18]李清.虚拟数控铣床加工过程仿真系统及相关技术的研究[D].天津:天津大学, 2004
[19]任迪峰.中药材干燥过程中质量退化及优化干燥工艺的研究[D].北京:中国农业大学, 2002
[20] Chankong V, Haimes Y Y. Multiobjective decision making theory and methodology [M]. New York: North-Holland, 1983
[21] Hans A E. Multicriteria optimization for highly accurate systems [M]. New York: plenum press, 1988
[22] L C J. Multiobjective Programming and Planing [M]. New York: Academic Press, 1978
[23] E S R. Multiple Criteria Optimization: Theory, Computation, and Application [M]. New York:: Wiley, 1986
[24] N.Srinivas, K.Deb. Multiobjective optimization using nondominated sorting in genetic algorithms [J]. Evolutionary Computation, 1994, 2 (3): 221-248
[25]崔逊学.基于多目标优化的进化算法研究[D].合肥:中国科学技术大学, 2001
[26] A I F. Basic Artificial Intelligence Application [J]. Materials Evaluation, 2000, 58 (1): 33-34
[27] Sang-Hui P, Seok-Pil L. EMG Pattern Recognition Based on Artificial Intelligence Techniques [J]. IEEE Trans on Rehabilitation Engineering, 1998, 6 (4): 400-405
[28] K.Deb. Multiobjective Optimization Using Evolutionary Algorithms [M]. Chichester, U.K.: 2001
[29] C.A.Coello, Veldhuizen D A V, G.B.Lamont. Evolutionary Algorithms for Solving Multi-Objective Problems [M]. Norwell: MA:Kluwer, 2002
[30] RS.Rosenberg. Simulation of Genetic Populations with Biochemical Properties [D]. Michigan: University of Michigan, 1967
[31] Schaffer J D. Some experiments in machine learning using vector evaluated genetic algorithms [D]. Tennessee: Vanderbilt University Electrical Engineering, 1984
[32] Goldberg D E, Richardson J. Proc. Second Int. Conf. on Genetic Algorithms [C]. Lawrence Erlbaum: 1987
[33] Goldberg. Genetic Algorithms in Search, Optimization and Machine Learning [M]. Massachusetts: Addison Wesley, 1989
[34] Fonseca C M, Fleming P J. Genetic Algorithm: Proceedings of the Fifth International Conference [C]. San Mateo: 1993
[35] Fonseca C M, Fleming P J. An overview of evolutionary algorithm in multiobjective optimization [J]. Evolutionary Computation, 1995, 3 (1): 1-16
[36] Fourman M P. Genetic Algorithm: Proceedings of the First International Conference [C]. Grefenstette: 1985
[37] Horn J, Nafpliotis N. Proceedings of the First IEEE Conference on Evolutionary Computation,IEEE World Congress on Computational Intelligence [C]. Piscataway,NJ: IEEE Service Center, 1994
[38] Zitzler E, Thiele L. Multiobjective Evolutionary Algorithms: A Comparative Case Study and the Strength Pareto Approach [J]. IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 1999, 3 (4): 257-271
[39] Zitzler E, Laumanns M, Thiele L. SPEA2: Improving the Strength Pareto Evolutionary Algorithm [J]. 2001
[40] Knowles J, Corne D. Proceedings of the 1999 Congress on Evolutionary Computation [C]. Piscataway,NJ: IEEE Press, 1999
[41] Deb K, Pratap A, Agarwal S, et al. A Fast and Elitist Multiobjective Genetic Algorithm: NSGA-II [J]. IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2002, 6 182-197
[42] M.Erickson, A.Mayer, J.Horn. First International Conference on Evolutionary Multi-Criterion Optimization [C]. Springer-Verlag: Lecture Notes in Computer Science, 2001
[43] D.Corne, J.Knowles, M.Oates. Proceedings of the Parallel problem Solving from NatureⅥConference [C]. Paris: Lecture Notes in Computer Science, 2000
[44]张志刚. 570kg/h CO2叶丝膨胀线工艺技术系统研究[J].中国烟草科学, 2004, (1): 14-17
[45]熊安言,王镇增,张志刚. COMAS烘丝机工艺参数与叶丝质量的测试分析[J].烟草科技, 2003, (8): 6-8
[46]张大波,刘志平,李跃锋. HXD运行参数与工作段气流初始温度相关性的回归分析[J].烟草科技, 2004, (9): 12-15
[47]马宇平. HXD在线膨胀工艺参数和膨胀率与卷烟质量的关系[J].烟草科技, 2004, (7): 4-7
[48]席年生,张大波,李跃锋. HXD蒸汽喷射量对叶丝膨胀效果及卷烟内在质量的影响[J].烟草科技, 2005, (3): 3-6
[49]张万祥,戴永生,姚文祥.风选工艺参数对梗丝结构的影响[J].烟草科技, 2004, (1): 10-11
[50]叶春文,杨明权,王兵.烘丝工艺参数对卷烟感官质量的影响[J].烟草科技, 2005, (11): 7-9
[51]周晖.卷烟厂的气力输送技术应用参数设计优化[D].南昌:南昌大学, 2005
[52]潘高伟,王川,甘学文.加料工序的不同工艺条件对烟草香味成分含量变化的影响研究[J].郑州轻工业学院学报(自然科学版), 2007, 22 (6): 43-48
[53]郑光勇,李斌,叶为全.基于RBF神经网络的制丝生产线仿真模型[J].计算机工程与应用, 2005, (21)
[54]陈颀,张云生,张钟录.烟叶复烤机智能控制系统[J].昆明理工大学学报, 2001, 26 (3)
[55]罗海燕,方文青,谢鑫.打叶质量与出片率的关系[J].烟草科技, 2005, (1)
[56]尹献忠,李晓,李显红.降低烟支密度和提高滤嘴吸阻在降焦中的应用[J].烟草科技, 2002, (8)
[57]郭俊成,程晓蕾.白肋烟料剂及处理工艺参数的优化设计[J].中国烟草科学, 1997, (4): 33-36
[58]张敏,童亿刚,戴志渊. SPC技术在制丝质量管理中的初步应用[J].烟草科技, 2004, (9): 10-11
[59]席年生,申玉军,储国海.不同施压条件下梗丝填充值与其膨胀效果的相关性[J].烟草科技, 2003, (9): 3-6
[60]王兵,杨达辉,林平.新的卷烟生产工序品质评价方法的建立[J].烟草科技, 2002, (11): 16-18
[61]陈洪,钱强,杨艳萍.润叶后不同储叶时间与烟叶品质变化关系的研究[J].烟草科技, 2002, (7)
[62]上海派克软件有限公司.上海卷烟厂制丝线质量分析控制系统[J].
[63]王德吉.复烤机智能控制系统研究[D].郑州:郑州大学, 2003
[64]简辉,杨学良,王保兴.复烤温度对烟叶化学成分及感官质量的影响[J].烟草科技, 2006, (2)
[65]李跃锋,姜焕元,刘志平.烟叶温度和含水率与打叶质量的关系[J].烟草科技, 2005, (2)
[66]李斌,于川芳,杨述元.卷烟材料对烟气特征的预测模型[J].烟草科技, 2005, (9)
[67]范黎,苗芊,赵航.卷烟吸阻和滤棒压降测量值与大气压力的关系[J].烟草科技, 2004, (12)
[68]赵同林,李兵役,田兴友.烟支密度与烟支重量、吸阻、硬度及标准偏差的关系[J].烟草科技, 2005, (4)
[69]于川芳,罗登山,王芳.卷烟“三纸一棒”对烟气特征及感官质量的影响(一) [J].中国烟草学报, 2001, 7 (2)
[70]于川芳,罗登山,王芳.卷烟“三纸一棒”对烟气特征及感官质量的影响(二) [J].中国烟草学报, 2001, 7 (3)
[71]彭黔荣.烟叶的化学成分与烟叶质量的人工神经网络预测[D].成都:四川大学, 2004
[72]何琴,高建华,刘伟.广义回归神经网络在烤烟内在质量分析中的应用[J].安徽农业大学学报, 2005, 32 (3): 406-410
[73]高大启,吴守一.并联神经网络在烤烟内在品质评定中的运用[J].农业机械学报, 1999, 30 (1): 58-62
[74]张志刚,王二彬,苏东赢.卷烟常规化学成分与焦油的线性回归分析[J].烟草科技, 2003, (11)
[75]于川芳,卢斌斌,牟定荣.卷烟劲头与其烟丝、烟气主要化学成分的相关性[J].烟草科技, 2006, (9)
[76]于建军,章新军,毕庆文.烤烟烟叶理化特性对烟气烟碱、CO、焦油量的影响[J].中国烟草科学, 2003, (3): 5-8
[77]刘丁伟,胡建军,熊燕.总粒相物与烟支重量、吸阻和抽吸口数的相关分析、通径分析[J].烟草科技, 2005, (8)
[78]陈景云,胡建军.烟叶化学成分-品质综合评价物元模型的建立与应用[J].烟草科技, 2003, (10)
[79]国防科技大学信息系统与管理学院.长沙卷烟厂企业产品管理信息综合服务系统需求分析书[J]. 2004
[80]国防科技大学信息系统与管理学院.长沙卷烟厂业务调查报告[J]. 2004
[81]毛多斌,马宇平,梅业安.卷烟配方和香精香料[M].北京:化学工业出版社, 2001
[82]中华人民共和国国家标准.卷烟[J]. GB5606-1996, 1996
[83] Heckman R A, Dube M F, Lynm D, et al. The role of tobacco leaf precursors in cigarette flavor [J]. Rec.Adv.Tob.Sci., 1981, (7): 107-153
[84]何新贵,梁久祯.利用目标函数梯度的遗传算法[J].软件学报, 2001, 12 (7): 981-986
[85]杨朋林,贺新.人工神经网络与遗传算法结合的研究[J].现代电子技术, 2002, (143): 105-107
[86]张克进,徐敏,俞金寿. GA-ANN算法在产品质量估计中的应用[J].华东理工大学学报, 2000, 26 (52): 512-516
[87] Kesbeng W, L G H. A hybrid intelligent method for modeling the EDM process [J]. International Journal of Machine Tools & Manufacture, 2003, 43 (10): 995-999
[88]刘建萍.制造智能技术新进展[J].新技术新工艺, 2001, (7): 2-4
[89]殷勇,吴守一,高大启.基于遗传算法的卷烟质量评定神经网络模型[J].农业机械学报, 1999, 30 (3): 71-75
[90]王清,马广富,弥曼.一种基于遗传算法的神经网络控制方法研究[J].系统仿真学报, 2006, 18 (4): 1070-1072
[91]王科俊,王克成.神经网络建模、预报与控制[M].哈尔滨:哈尔滨工程大学出版社, 1996
[92] D J N, Gupta, S S R. Comparing back-propagation with a genetic algorithm for neural network training [J]. Omega, 1999, (27): 679-684
[93] F C D, T R C, L M R. Combining a neural network with a genetic algorithm for process parameter optimization [J]. Engineering Application of Artificial intelligence, 2000, (3): 391-396
[94] T.Hagan M, B.Demuth H, H.Beale M.神经网络设计[M].北京:机械工业出版社, 2002
[95] [俄]加卢什金.神经网络理论[M].北京:清华大学出版社, 2002
[96]吴佑寿,赵明生.激活函数可调的神经元模型及其有监督学习与应用[J].中国科学(E辑), 2001, 31 (3): 263-272
[97]刘耦耕,李圣清,肖强晖.多层前馈人工神经网络结构研究[J].湖南师范大学自然科学学报, 2004, 27 (1): 26-30
[98] C C F. Back-propagation neural networks for nonlinear self-tuning adaptive control [J]. IEEE Control system Magazine, 1990, (4): 44-48
[99] S N K, K P. Identification and control for dynamic systems using neural networks [J]. IEEE Trans. on Neural Networks, 1990, 1 (1): 4-27
[100] A B S, F V W S. Correlation based model validity tests for nonlinear models [J]. Int J control, 1986, (44): 235-244
[101] M B C. Neural networks for pattern recognition [M]. Oxford. England: Oxford University Press, 1996
[102] Iizaka T, Matsui T, Fukuyama Y. A Novel Daily Peak Load Forecasting Method using Analyzable Structured Neural Network [J]. IEEE T&D Asia, Yokohama, 2002, 1-6
[103] B W. Neural networks application in industry, business and science [J]. Communication of the ACM, 1994, (37): 93-105
[104]宋锐,张静,夏胜平.一种基于BP神经网络群的自适应分类方法及其应用[J].电子学报, 2001, (12)
[105] D G. Effective back-propagation training with variable step size [J]. Neural Networks, 1997, 10 (1): 69-82
[106] D S. Methods to speed up error BP learning algorithm [J]. ACM Computing Survey, 1995, (27): 519-592
[107] S Y. Global optimization for NN training [J]. IEEE Computer, 1996, (3): 45-54
[108]崔光照,毕娟,许进.前向神经网络结构优化的研究进展[J].郑州轻工业学院学报(自然科学版), 2003, 18 (3): 11-13
[109]王正志,薄涛.进化计算[M].长沙:国防科技大学出版社, 2000
[110]孙达.卷烟制丝工序加料均匀性检测项目设计[D].上海:上海交通大学, 2004
[111]戴志渊. SAS统计分析系统在上海卷烟厂制丝车间质量管理中的应用[D].上海:复旦大学, 2000
[112]孟冬玲.烟草和烟气中酚类物质的分析方法研究及应用[D].昆明:昆明理工大学, 2004
[113] R B A, L B R. In Computer Science and Statistics: Proc. of the 20st Interface [C]. 1988
[114] R M H, G I A. Inductive learning algorithms for complex systems modeling [M]. Boca Raton, London, Tokyo: CRC Press Inc, 1994
[115]岳超源.决策理论与方法[M].北京:科学出版社, 2003
[116]汪祖柱,程家兴.一种混合交叉策略的多目标演化算法及其性能分析[J].系统仿真学报, 2005, 17 (10)
[117]李桂琴,袁庆丰,王克胜, et al.制造过程多目标优化的集成计算智能方法[J].计算机集成制造系统, 2006, 12 (12): 2039-2043
[118]刘宝碇,赵瑞清.随机规划与模糊规划[M].北京:清华大学出版社,1998
[119]魏发远,李世其.基于混合遗传算法的多目标动态优化设计[J].宇航学报, 2004, 25 (6): 609-615
[120] S S V, K J V, D K B. Solution of constrained optimization problems by multi-objective genetic algorithm [J]. Computers and Chemical Engineering, 2002, 26 (10): 1481-1492
[121]崔逊学.多目标进化算法及其应用[M].北京:国防工业出版社, 2006
[122]胡开文.烟叶打叶复烤工艺与设备[M].北京:化学工业出版社, 2002
[123]王允白.烟叶主要化学成分与评吸香味关系研究[D].北京:中国农业科学院, 1996
[124]高家合,秦西云,谭仲夏.烟叶主要化学成分对评吸质量的影响[J].山地农业生物学报, 2004, 23 (6): 497-501
[125]闫克玉,王建民,屈剑波.河南烤烟评吸质量与主要理化指标的相关分析[J].烟草科技, 2001, (10): 5-9
[126]彭黔荣,蔡元青,王东山.根据常规化学指标识别烟叶品质的BP神经网络模型[J].中国烟草学报, 2005, 11 (5)
[127]谭梅,赵葵银,陈强.基于神经网络的卷烟工艺风力系统的压力控制[J].自动化与仪表, 2005, (4)
[128]殷勇,吴守一.遗传RBF神经网络在卷烟香气质量评定中的应用[J].农业机械学报, 2001, 17 (6): 147-149
[129]王强.基于支持向量机的卷烟叶组配方设计数据挖掘预测模型及应用研究[D].长沙:国防科学技术大学, 2006