神经网络规则抽取及其在带钢热镀锌质量控制参数设定中的应用研究
|
摘要
人工神经网络作为动态系统识别、数据挖掘的一种常用的智能工具,已广泛应用于模式识别、图像处理、自动控制、质量建模、机器人、信号处理、管理、商业、医疗和军事等领域。然而,它无法能将相应物理系统直接转化为能被理解的知识表达结果,其“黑箱”特性极大地限制了神经网络的进一步应用。本文主要研究神经网络的规则抽取方法,使神经网络模型输入变量与输出变量之间的关系具有更直观的可理解性,并将从神经网络中抽取的“知识”结合产品质量预测模型实现质量控制参数自动设定的目的,避免了人工凭经验来进行参数设定,对于深入认识生产规律、改善生产工艺、提高产品质量有着重要的意义。
论文主要的创新性成果有以下两方面的内容:
(1)提出了基于优化激活函数的神经网络规则抽取方法,并将其应用于热镀锌生产中的数据挖掘和知识发现,克服传统神经网络产品质量监控模型中“解释性差”的难题。通过加入指数变量的惩罚项,使激活函数的输出值更趋于0和1的二值化,提高了规则的覆盖率,以变量区间的形式成功的提取出产品原材料参数、生产控制参数与产品质量间的对应关系,为生产过程监控和质量管理提供有效的分析方法和控制手段。应用锌层重量的实际生产数据进行模型验证,分析结果表明:用本文方法提取到的知识规则覆盖率达到94.7%。
(2)提出了多变量产品质量模型中过程控制参数的设定方法。首先运用规则抽取方法,从现有的生产数据中找出知识规则作为过程参数的取值区间,同时根据生产数据建立质量预测模型并利用它在规则范围内准确的预测出控制参数值,从而达到过程参数预测控制的目的。本文提出的过程参数预测控制方法搜索空间小,且算法速度快,应用某钢厂实际的镀锌生产数据验证了本文方法的有效性。
Although ANN (Artificial Neural Network) has been widely used in pattern recognition, control and decision-making, system modeling, the inherent“black box”characteristic of ANN has greatly limited their further application. This work studies the rule extraction method of ANN, make clear the relationship between inputs and outputs of ANN and make the model easy to understand. What is more, we apply the“knowledge”extracted from ANN to production quality modeling in order to automatic set the quality parameters combining with quality prediction model, which avoid the artifical seting by people’s experience. It has the important significance for deeply understanding the production law, improving the production technology and the product quality.
The main content of this work is as follows:
(1) The method of the rule extraction of ANN based on optimized activation functions is put forward, which is applied to data mining and knowledge discovery in the production of hot-dip galvanizing and overcomes the defect of“poor explanation”of the traditional ANN. The penalty term of the exponential variable is applied to make the values of the activation function have a better approximation to binary values: 0 or 1,which is helpful for rule extraction. The corresponding relationships among the raw materials parameters, control parameters and the product quality are extracted in the form of production rules, which provides effective means of analysis and control in production process monitoring and quality management. The results of model verification using actual product data of zinc coat weight showed that the coverage rate of the knowledge rules extracted from our method has reached 94.7%.
(2) The prediction-control method of process control parameters in the multivariable production quality model is proposed. The knowledge rules in existing production data are extracted to be constraint conditions of parameters setting problem. Simultaneity, the quality prediction modeling was built based on process control parameters, which was used to find the exactly control parameters in the range of the rules. The proposed method has the advantages of small search space of problem domain and fast convergence during optimization, and has been successfully applied to off-line navigation system for zinc weight control in hot-dip galvanizing strip.
论文主要的创新性成果有以下两方面的内容:
(1)提出了基于优化激活函数的神经网络规则抽取方法,并将其应用于热镀锌生产中的数据挖掘和知识发现,克服传统神经网络产品质量监控模型中“解释性差”的难题。通过加入指数变量的惩罚项,使激活函数的输出值更趋于0和1的二值化,提高了规则的覆盖率,以变量区间的形式成功的提取出产品原材料参数、生产控制参数与产品质量间的对应关系,为生产过程监控和质量管理提供有效的分析方法和控制手段。应用锌层重量的实际生产数据进行模型验证,分析结果表明:用本文方法提取到的知识规则覆盖率达到94.7%。
(2)提出了多变量产品质量模型中过程控制参数的设定方法。首先运用规则抽取方法,从现有的生产数据中找出知识规则作为过程参数的取值区间,同时根据生产数据建立质量预测模型并利用它在规则范围内准确的预测出控制参数值,从而达到过程参数预测控制的目的。本文提出的过程参数预测控制方法搜索空间小,且算法速度快,应用某钢厂实际的镀锌生产数据验证了本文方法的有效性。
Although ANN (Artificial Neural Network) has been widely used in pattern recognition, control and decision-making, system modeling, the inherent“black box”characteristic of ANN has greatly limited their further application. This work studies the rule extraction method of ANN, make clear the relationship between inputs and outputs of ANN and make the model easy to understand. What is more, we apply the“knowledge”extracted from ANN to production quality modeling in order to automatic set the quality parameters combining with quality prediction model, which avoid the artifical seting by people’s experience. It has the important significance for deeply understanding the production law, improving the production technology and the product quality.
The main content of this work is as follows:
(1) The method of the rule extraction of ANN based on optimized activation functions is put forward, which is applied to data mining and knowledge discovery in the production of hot-dip galvanizing and overcomes the defect of“poor explanation”of the traditional ANN. The penalty term of the exponential variable is applied to make the values of the activation function have a better approximation to binary values: 0 or 1,which is helpful for rule extraction. The corresponding relationships among the raw materials parameters, control parameters and the product quality are extracted in the form of production rules, which provides effective means of analysis and control in production process monitoring and quality management. The results of model verification using actual product data of zinc coat weight showed that the coverage rate of the knowledge rules extracted from our method has reached 94.7%.
(2) The prediction-control method of process control parameters in the multivariable production quality model is proposed. The knowledge rules in existing production data are extracted to be constraint conditions of parameters setting problem. Simultaneity, the quality prediction modeling was built based on process control parameters, which was used to find the exactly control parameters in the range of the rules. The proposed method has the advantages of small search space of problem domain and fast convergence during optimization, and has been successfully applied to off-line navigation system for zinc weight control in hot-dip galvanizing strip.
引文
[1]李九岭.带钢连续热镀锌[M].北京:冶金工业出版社,1981
[2] Tzafestas S, Triantafyllakis A. Deterministic scheduling in computing and manufacturing systems: A survey of models and algorithms [J].Mathematics and Computers in Simulation, 1998(35): 397- 434
[3]朱立.钢材热镀锌[M].北京:化学工业出版社,2006
[4]吴澄.现代集成制造系统导论—概念、方法、技术和应用[M] .北京:清华大学出版社,2002.6
[5]万维汉.流程工业过程的先进控制及其应用[C].首届全国有色金属自动化技术与应用学术年会论文集,2003,24:188-191
[6]钟秉林,黄仁.机械故障控制学第三版[M].北京:机械工业出版社,2007
[7]胡包钢,王泳,杨双红,曲寒冰.如何增加人工神经元网络的透明度[J].模式识别与人工智能.2007, 2
[8]边军,张福波,刘相华,等.我国热镀锌机组连续退火技术的现状与展望[J].金属热处理,2004,29(7):13-16
[9]纪凤玲.连续热镀锌生产工艺及现状分析[J].有色金属设计,1997,24(1):50-54
[10]宋加.我国热镀锌钢板生产及镀锌技术的发展[J].轧钢,2006,23(3):42-46
[11]王兰军.水泥回转窑故障控制系统的研究[D].浙江大学材料加工工程学院,2004
[12]陈秀政,宋金城,王建敏.多元线性回归在坐标测量中的应用[J].宇航计测技术,2007,27(2):4-8
[13]王惠文,吴载斌,孟洁.偏最小二乘回归的线性与非线性方法[M].北京:国防工业出版社,2006
[14]楼顺天,施阳.基于Matlab的系统分析与设计:神经网络[M].西安电子科技大学出版社.1996
[15] Martin T. Hagan, Howard B. Demuth, Mark H. Beale. Neural Network Design. PWS Publishing Company. 19996
[16]周开利,康耀红.神经网络模型及其Matlab仿真程序设计.北京:清华大学出版社,2005.7
[17] Hornik K,Stinchcombe M,White H.Multilayer feed-forward networks are universal approximators[J].Neural Networks,1989,2(5):359-366
[18]胡包钢,王泳,杨双红,等.如何提高人工神经元网络的透明度?[J].模式识别与人工智能,2007,20(1):72-84
[19]杨斌,聂在平,夏耀先,等.基于贝叶斯神经网络的非参数回归[J].电子科技大学学报,2002,31(2):159-162
[20] Lampinen J,Vehtari A.Bayesian approach for neural networks-review and case studies [J].Neural Networks,2001,14(3):257-274
[21] Abu-Mostafa Y S. Learning from Hints in Neural Networks. Journal of Complexity, 1990, 6(2): 192-198
[22] Jang J S R. ANFIS: Adaptive-Network-Based Fuzzy Inference Systems[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1993, 23(3): 665-685
[23] Towell G, Shavlik J. Knowledge-based Artificial Neural Networks[J]. Artificial Intelligence,1994,70(1/2):119-165
[24] Hu B G, Qu H B, Wang Y, et al. A Generalized Constraint Neural Networks Model: Associating Partially Known Relationships for Nonlinear Regressions [EB/OL]. Submitted to IEEE Transactions on Neural Networks, 2005. http://liama.ia.ac.cn/hubg/paper.html
[25]张朝晖,陆玉昌,张钹.利用神经网络发现分类规则[J].计算机学报,1999,22(1):108-112
[26]侯广坤,张劲峰.基于决策树的神经网络规则抽取方法[J].中山大学学报,2000,39(4):27-30
[27] Fu L M.Rule generation form neural networks[J].IEEE Transactions on Systerms,Man,and Cybernetics,1994,24(8):1114-1124
[28] Setiono R.Extracting rules from neural networks by pruning and hidden unit splitting [J]. Neural Computation, 1997,(9): 205-225.
[29] Towell G G,Shavlik J W.Extracting refined rules from knowledge-based neural networks[J].Machine Learning,1993,13(1):71-101
[30] Craven M W, Shavlik J W.Extracting tree-structured representation of trained networks [A].Advances in Neural Information Proc Systems[C].Cambridge,MA:MIT Press,1996:24-30
[31] Garson G D. Interpreting neural-network connection weights[J].AI Expert, 1991, 6(4): 47-51
[32] Olden J D, Jackson D A.Illuminating the black-box: A randomization approach for understanding variable contributions in artificial neural networks[J]. Ecological Modeling, 2002, 154(1): 135-150
[33] Melnik O, Pollack J. Using graphs to analyze high-dimensional classifiers[C]. Proceedings of the International Joint Conference on Neural Networks, Italy: IEEE, 2000, 425-430
[34] Gallant S I. Connectionist expert systems. Communications of the ACM, 1988, 31(2): 152-169
[35] Saito K, Nakano R. Rule extraction from facts and neural networks. In: Proceedings of the International Neural Network Conference, Paris, France, 1990, 379-382
[36] Fu L. Rule. learning by searching on adapted nets. In: Proceedings of the 9th National Conference on Artificial Intelligence, Anaheim, CA, 1991, 590-595
[37] Giles C L, Miller C B, Chen D, Chen H H, Sun G Z, Lee Y C. Learning and extracting finite state automata with second-order recurrent neural networks. Neural Computation, 1992, 4(3): 393-405
[38] Towell G G, Shavlik J W. Interpretation of artificial neural networks: mapping knowledge-based neural networks into rules. In: Moody J, Hanson S, Lippman R, eds. Advances in Neural Information Processing Systems 4, San Mateo, CA: Morgan Kaufmann, 1992, 977-984
[39] Omlin C W, Giles C L, Miller C B. Heuristics for the extraction of rules from discrete time recurrent neural networks. In: Proceedings of the International Joint Conference on Neural Networks, Baltimore, MD, 1992, vol.1, 33-38
[40] Craven M W, Shavlik J W. Learning symbolic rules using artificial neural networks. In: Proceedings of the 10th International Conference on Machine Learning, Amherst, MA, 1993, 73-80
[41] Sestito S, Dillon T. Knowledge acquisition of conjunctive rules using multilayered neural networks. International Journal of Intelligent Systems, 1993, 8(7): 779-805
[42] Craven M W, Shavlik J W. Using sampling and queries to extract rules from trained neural networks. In: Proceedings of the 11th International Conference on Machine Learning, New Brunswick, NJ, 1994, 37-45
[43] Carpenter G A, Tan A W. Rule extraction: from neural architecture to symbolic representation. Connection Science, 1995, 7(1): 3-27
[44] Thrun S. Extracting rules from artificial neural networks with distributed representations. In: Tesauro G, Touretzky D, Leen T, eds. Advances in Neural Information Processing Systems 7, Cambridge, MA: MIT Press, 1995, 505-512]
[45] Setiono R, Liu H. Understanding neural networks via rule extraction. In: Proceedings of the 14th International Joint Conference on Artificial Intelligence, Montreal, Canada, 1995, 480-485
[46] Andrews R, Diederich J, Tickle A B. Survey and critique of techniques for extracting rules from trained artificial neural networks. Knowledge-Based Systems, 1995, 8(6): 373-389
[47] Tickle A B, Andrews R, Golea M, Diederich J. The truth will come to light: directions and challenges in extracting the knowledge embedded within trained artificial neural networks. IEEE Transactions on Neural Networks, 1998, 9(6): 1057-1068
[48] Alexander J A, Mozer M C. Template-based procedures for neural network interpretation. Neural Networks, 1999, 12(3): 479-498
[49] Krishnan R, Sivakumar G, Bhattacharya P. Extracting decision trees from trained neural networks. Pattern Recognition, 1999, 32(12): 1999-2009
[50] Maire F. Rule-extraction by backpropagation of polyhedra. Neural Networks, 1999, 12(4-5): 717-725
[51] Taha I A, Ghosh J. Symbolic interpretation of artificial neural networks. IEEE Transactions on Knowledge and Data Engineering, 1999, 11(3): 448-463
[52] Setiono R. Extracting M-of-N rules from trained neural networks. IEEE Transactions on Neural Networks, 2000, 11(2): 512-519
[53] Setiono R, Leow W K. FERNN: an algorithm for fast extraction of rules from neural networks. Applied Intelligence, 2000, 12(1-2): 15-25
[54] Tsukimoto H. Extracting rules from trained neural networks. IEEE Transactions on Neural Networks, 2000, 11(2): 377-389
[55] Zhou Z-H, Chen S-F, Chen Z-Q. A statistics based approach for extracting priority rules from trained neural networks. In: Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks, Como, Italy, 2000, vol.3, 401-406
[56] Duch W, Adamczak R, Grabczewski K. A new methodology of extraction, optimization and application of crisp and fuzzy logical rules. IEEE Transactions on Neural Networks, 2001, 12(2): 277-306
[57] Saito K, Nakano R. Extracting regression rules from neural networks. Neural Networks, 2002, 15(10): 1279-1288
[58] Castro J L, Mantas C J, Benitez J M. Interpretation of artificial neural networks by means of fuzzy rules. IEEE Transactions on Neural Networks, 2002, 13(1): 101-116
[59] Setiono R. Extracting rules from pruned neural networks for breast cancer diagnosis. Artificial Intelligence in Medicine, 1996, 8(1): 37-51
[60] Hayashi Y, Setiono R, Yoshida K. A comparison between two neural network rule extraction techniques for the diagnosis of hepatobiliary disorders. Artificial Intelligence in Medicine, 2000, 20(3): 205-216
[61] Craven M W, Shavlik J W. Understanding time-series networks: a case study in rule extraction. International Journal of Neural Systems, 1997, 8(4): 373-384
[62] Fu L. A neural-network model for learning domain rules based on its activation function characteristics. IEEE Transactions on Neural Networks, 1998, 9(5): 787-795
[63] Setiono R, Thong J Y L, Yap C. Symbolic rule extraction from neural networks: an application to identifying organizations adopting IT. Information and Management, 1998, 34(2): 91-101
[64] Baesens B, Setiono R, Mues C, Vanthienen J. Using neural network rule extraction and decision tables for credit risk evaluation. Management Science, 2003, 49(3): 312-329
[65] Duch W, Adamczak R, Grabczewski K. Extraction of crisp logical rules using constrained backpropagation networks. In: Proceedings of the International Conference on Neural Networks, Houston, TX, 1997, 2384-2389
[66] Quinlan J R. Introduction of decision trees [ J ]. Ma2 chine Learning, 1986 (1) : 84 - 100
[67] J.M. Tarela, E. Alonso and M.V. Martinez, A representation method for PWL functions oriented to parallel processing, Mathl. Comput. Modelling 13 (lo), 5-83 (1990)
[68] R. Setiono, J.Y.L Thong, An approach to generate rules from neural networks for regression problems. European Journal of Operational Research, 155 (2004) 239-250
[69] J.M. Tarela, K. Basterretxea, I. Del Campo, M.V. Martinez, E. Alonso, Optimised PWL Recursive Approximation and its Application to Neuro-Fuzzy Systems, Mathematical and Computer Modeling, 35 (2002) 867-883
[70] Andrews R, Diederich J, Tickle A B. Andrews Survey and critique of techniques for extracting rules from trained artificial neural networks[J]. Knowledge-Based Systems, 1995, 8(6): 373-389
[71] Blum A, Chawla S. Learning from Labeled and Unlabeled Data using Graph Mincuts[C]. Proc. 18th International Conf. on Machine Learning, San Mateo: Morgan Kaufmann, 2001, 19-26
[72]赵玲玲,翁苏明,曾华军译.模式分析的核方法[M].机械工业出版社,2006
[73] Formmann K M,Paramonov V A.Vertical a new process for t he hot-dip coating of steel sheet[A].The Use and Manufacture of Zinc and Zinc Alloy Coated Sheet Steel Products into 21st Century[C].Chicago:The Iron and Steel Society,1995,189～192
[74]何晓群,刘文卿.应用回归分析[M].中国人民大学出版社,2007
[75] Belkin M,Niyogi P,Sindhwani V.Manifold Regularization: A geometric framework for learning from examples[J].Journal of Machine Learning Research.2006,7(11):2399-2434
[76] Towell G G,Shavlik J W.Knowledge-based artificial neural networks[J].Artificial Intelligence,1994,70(1-2):119-165
[77] Benitez J M, Castro J L, Requena I. Are artificial neural networks black boxes? IEEE Transactions on Neural Networks, 1997, 8(5): 1156-1164
[78] Setiono R.Extracting rules from neural networks by pruning and hidden unit splitting [J]. Neural Computation, 1997,(9): 205-225.
[79] Saito K, Nakano R. Rule extraction from facts and neural networks. In: Proceedings of the International Neural Network Conference, Paris, France, 1990, 379-382
[80] J. Ross Quinlan, C4.5: Programs for Machine Learning, Morgan Kaufmann, San Mateo, California, 1993
[81] Murphy P M, Pazzani M J. ID2-of-3: constructive induction of M-of-N concepts for discriminators in decision trees. In: Proceedings of the 8th International Workshop on Machine Learning, Evanston, IL, 1991, 183-187
[82] Wu X. Knowledge Acquisition from Databases, Norwood, NJ: Ablex, 1995
[83] Palade V, Negoita G, Ariton V. Genetic algorithm optimization of knowledge extraction from neural networks. In: Proc of the 6th Conf on Neural Information Processing. Perth, Australia, 1999, 752-758
[84]宋秦豹,沈钧毅.神经网络数据挖掘中的数据准备问题[J].计算机工程与应用,2002,12:102-104
[85] Stone M. Cross-validatory choice and assessment of statistical predictions[J]. Journal of the Royal Statistical Society, 1974, 36(2): 111-147
[86] Schenker B, Agarwal M. Cross-validated Structure Selection for Neural Networks[J]. Computers chem, 1996, 20(2): 175-186
[87] Andersen T, Martinez T. Cross Validation and MLP Architecture Selection[C]. Proceedings of the IEEE International Joint Conference on Neural Network, Washington DC: IEEE, 1999, 1614-1619
[88] Setiono R. A penalty-function approach for pruning feedforward neural networks[J]. Neural Computation,1997,9(1):185-204
[89] Hagan M T, Menhaj M B. Training feedforward networks with the Marquardt algorithm[J]. IEEE Transactions on Neural Networks, 1994, 5(6): 989-993
[90] UCI–MLR: University of California Irvine Machine Learning Repository. California Irvine, Tex: University of California Irvine Libraries, 1987[2007]. http://archive.ics.uci.edu/ml/index.html
[91]陈兆乾,刘宏,周戎,陈世福.一种混合型多概念获取算法HMCAP及其应用.计算机学报,1996,19(10):753-761
[92]李开泰,黄艾香.张量分析及其应用[M].科学出版社,2005
[93] Muslea I, Minton S, Craig K A. Active+Semi-Supervised Learning=Robust Multi-View Learning[C]. 19th International Conference. on Machine Learning, Sydney, Australia: Morgan Kaufmann Publishers, 2002, 435-442
[94] Setiono R. Extracting rules from neural networks by pruning and hidden unit splitting[J]. Neural Computation, 1997, (9): 205-225.
[95]万百五.工业大系统优化与产品质量控制[M].北京:科学出版社,2003
[96]万维汉,万百五,杨金义.闪速炉的神经网络冰镍质量模型与稳态优化控制研究[J].自动化学报,1999,25(6):800-803
[97]刘国海,张浩,戴先中.神经网络逆系统在电机变频调速系统中的应用[J].电工技术学报,2003,18 (3) :67-71
[98]桑保华,薛晓中.多变量解耦控制方法[J].火力与指挥控制,2007,32(11):245-248
[99]常玉清,李玉朝,吕哲,王福利.基于两级神经网络的发酵过程多变量前馈解耦控制[J].东北大学学报,2007,28(7):925-928
[100]张晓婕.多变量时变系统CARMA模型近似解耦法[J].中国计量学院学报,2004,15(4):284-286
[101]王晰,李少远.多模型分层递阶自适应前馈解耦控制器[J].控制与决策,2005,20(1):17-22
[102]柴天佑.多变量自适应解耦控制及应用[M ].北京:科学出版社,2001
[103]张杰,杨翠娥.电加热锅炉系统神经网络PID解耦控制器[J].系统工程理论与实践,2002,22(1):123-127
[104]周涌,陈庆伟.基于动态神经网络解耦线性化的内模控制[J].南京理工大学学报,2004,28(6):566-570
[105]梁伟平,张健宇.基于神经网络的除氧器水位解耦控制[J].华北电力大学学报,2005,32(1):66-68
[106]何关钰.限制解耦控制的充分必要条件[J].控制理论与应用,1998,15(1):125-129
[107]杨辉,王金章.多变量解耦模糊控制器的研究[J].控制与决策,1988,3 (1):17-21
[108] Czogalaand E,Zimmermann H J.Some aspects of synthesis of probabilistic fuzzy controllers[J]. Fuzzy Sets and Systems,1984,13(2): 169-178
[109] Kiszka J B,Gupta M M,Trojan G M.Multivariable fuzzy controller under godel a implicat ion[J].Fuzzy Sets and Systems,1990,34(3):301-321
[110] Walichiewicz L.Decomposition of linguistic rule in the design of a multidimensional fuzzy control algorithm[J].Cybernetics SystRes,1984,2 ( 3) :185-193
[111]任新宇,樊思齐.航空发动机多变量自学习模糊解耦控制[J].推进技术,2004,25(6):535-537
[112]邱焕耀,毛宗源.采用模糊控制的感应电动机解耦变结构控制系统的研究[J].自动化学报,1998,24(3):391-394
[113]林莉军.宝钢热镀锌机组锌层重量控制模型的应用[J].宝钢技术,2007,4: 33?36
[114]王宸煜,王敏,董溯攀.人工神经网络在镀锌钢板点焊性能估测中的应用[J].上海交通大学学报,2001,35(3):420?423
[115] Sanchez A P, Blanco I D, Vega A A C, et al. Virtual Sensor Design for Coating Thickness Estimation in a Hot Dip Galvanising Line Based on Interpolated SOM Local Models[C]. IEEE Industrial Electronics Society, 2002 28th Annual Conference of the Industrial Electronics Society, New York: IEEE, 2002, 1584?1589
[116]姚林,阳建宏,何飞,等.基于核偏最小二乘的锌层重量预测模型[J].控制工程,2008,15(2):154?157
[117]姚林,阳建宏,徐金梧,等.基于偏最小二乘回归模型的带钢热镀锌质量监控方法[J].北京科技大学学报,2007,29(6):627-631
[2] Tzafestas S, Triantafyllakis A. Deterministic scheduling in computing and manufacturing systems: A survey of models and algorithms [J].Mathematics and Computers in Simulation, 1998(35): 397- 434
[3]朱立.钢材热镀锌[M].北京:化学工业出版社,2006
[4]吴澄.现代集成制造系统导论—概念、方法、技术和应用[M] .北京:清华大学出版社,2002.6
[5]万维汉.流程工业过程的先进控制及其应用[C].首届全国有色金属自动化技术与应用学术年会论文集,2003,24:188-191
[6]钟秉林,黄仁.机械故障控制学第三版[M].北京:机械工业出版社,2007
[7]胡包钢,王泳,杨双红,曲寒冰.如何增加人工神经元网络的透明度[J].模式识别与人工智能.2007, 2
[8]边军,张福波,刘相华,等.我国热镀锌机组连续退火技术的现状与展望[J].金属热处理,2004,29(7):13-16
[9]纪凤玲.连续热镀锌生产工艺及现状分析[J].有色金属设计,1997,24(1):50-54
[10]宋加.我国热镀锌钢板生产及镀锌技术的发展[J].轧钢,2006,23(3):42-46
[11]王兰军.水泥回转窑故障控制系统的研究[D].浙江大学材料加工工程学院,2004
[12]陈秀政,宋金城,王建敏.多元线性回归在坐标测量中的应用[J].宇航计测技术,2007,27(2):4-8
[13]王惠文,吴载斌,孟洁.偏最小二乘回归的线性与非线性方法[M].北京:国防工业出版社,2006
[14]楼顺天,施阳.基于Matlab的系统分析与设计:神经网络[M].西安电子科技大学出版社.1996
[15] Martin T. Hagan, Howard B. Demuth, Mark H. Beale. Neural Network Design. PWS Publishing Company. 19996
[16]周开利,康耀红.神经网络模型及其Matlab仿真程序设计.北京:清华大学出版社,2005.7
[17] Hornik K,Stinchcombe M,White H.Multilayer feed-forward networks are universal approximators[J].Neural Networks,1989,2(5):359-366
[18]胡包钢,王泳,杨双红,等.如何提高人工神经元网络的透明度?[J].模式识别与人工智能,2007,20(1):72-84
[19]杨斌,聂在平,夏耀先,等.基于贝叶斯神经网络的非参数回归[J].电子科技大学学报,2002,31(2):159-162
[20] Lampinen J,Vehtari A.Bayesian approach for neural networks-review and case studies [J].Neural Networks,2001,14(3):257-274
[21] Abu-Mostafa Y S. Learning from Hints in Neural Networks. Journal of Complexity, 1990, 6(2): 192-198
[22] Jang J S R. ANFIS: Adaptive-Network-Based Fuzzy Inference Systems[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1993, 23(3): 665-685
[23] Towell G, Shavlik J. Knowledge-based Artificial Neural Networks[J]. Artificial Intelligence,1994,70(1/2):119-165
[24] Hu B G, Qu H B, Wang Y, et al. A Generalized Constraint Neural Networks Model: Associating Partially Known Relationships for Nonlinear Regressions [EB/OL]. Submitted to IEEE Transactions on Neural Networks, 2005. http://liama.ia.ac.cn/hubg/paper.html
[25]张朝晖,陆玉昌,张钹.利用神经网络发现分类规则[J].计算机学报,1999,22(1):108-112
[26]侯广坤,张劲峰.基于决策树的神经网络规则抽取方法[J].中山大学学报,2000,39(4):27-30
[27] Fu L M.Rule generation form neural networks[J].IEEE Transactions on Systerms,Man,and Cybernetics,1994,24(8):1114-1124
[28] Setiono R.Extracting rules from neural networks by pruning and hidden unit splitting [J]. Neural Computation, 1997,(9): 205-225.
[29] Towell G G,Shavlik J W.Extracting refined rules from knowledge-based neural networks[J].Machine Learning,1993,13(1):71-101
[30] Craven M W, Shavlik J W.Extracting tree-structured representation of trained networks [A].Advances in Neural Information Proc Systems[C].Cambridge,MA:MIT Press,1996:24-30
[31] Garson G D. Interpreting neural-network connection weights[J].AI Expert, 1991, 6(4): 47-51
[32] Olden J D, Jackson D A.Illuminating the black-box: A randomization approach for understanding variable contributions in artificial neural networks[J]. Ecological Modeling, 2002, 154(1): 135-150
[33] Melnik O, Pollack J. Using graphs to analyze high-dimensional classifiers[C]. Proceedings of the International Joint Conference on Neural Networks, Italy: IEEE, 2000, 425-430
[34] Gallant S I. Connectionist expert systems. Communications of the ACM, 1988, 31(2): 152-169
[35] Saito K, Nakano R. Rule extraction from facts and neural networks. In: Proceedings of the International Neural Network Conference, Paris, France, 1990, 379-382
[36] Fu L. Rule. learning by searching on adapted nets. In: Proceedings of the 9th National Conference on Artificial Intelligence, Anaheim, CA, 1991, 590-595
[37] Giles C L, Miller C B, Chen D, Chen H H, Sun G Z, Lee Y C. Learning and extracting finite state automata with second-order recurrent neural networks. Neural Computation, 1992, 4(3): 393-405
[38] Towell G G, Shavlik J W. Interpretation of artificial neural networks: mapping knowledge-based neural networks into rules. In: Moody J, Hanson S, Lippman R, eds. Advances in Neural Information Processing Systems 4, San Mateo, CA: Morgan Kaufmann, 1992, 977-984
[39] Omlin C W, Giles C L, Miller C B. Heuristics for the extraction of rules from discrete time recurrent neural networks. In: Proceedings of the International Joint Conference on Neural Networks, Baltimore, MD, 1992, vol.1, 33-38
[40] Craven M W, Shavlik J W. Learning symbolic rules using artificial neural networks. In: Proceedings of the 10th International Conference on Machine Learning, Amherst, MA, 1993, 73-80
[41] Sestito S, Dillon T. Knowledge acquisition of conjunctive rules using multilayered neural networks. International Journal of Intelligent Systems, 1993, 8(7): 779-805
[42] Craven M W, Shavlik J W. Using sampling and queries to extract rules from trained neural networks. In: Proceedings of the 11th International Conference on Machine Learning, New Brunswick, NJ, 1994, 37-45
[43] Carpenter G A, Tan A W. Rule extraction: from neural architecture to symbolic representation. Connection Science, 1995, 7(1): 3-27
[44] Thrun S. Extracting rules from artificial neural networks with distributed representations. In: Tesauro G, Touretzky D, Leen T, eds. Advances in Neural Information Processing Systems 7, Cambridge, MA: MIT Press, 1995, 505-512]
[45] Setiono R, Liu H. Understanding neural networks via rule extraction. In: Proceedings of the 14th International Joint Conference on Artificial Intelligence, Montreal, Canada, 1995, 480-485
[46] Andrews R, Diederich J, Tickle A B. Survey and critique of techniques for extracting rules from trained artificial neural networks. Knowledge-Based Systems, 1995, 8(6): 373-389
[47] Tickle A B, Andrews R, Golea M, Diederich J. The truth will come to light: directions and challenges in extracting the knowledge embedded within trained artificial neural networks. IEEE Transactions on Neural Networks, 1998, 9(6): 1057-1068
[48] Alexander J A, Mozer M C. Template-based procedures for neural network interpretation. Neural Networks, 1999, 12(3): 479-498
[49] Krishnan R, Sivakumar G, Bhattacharya P. Extracting decision trees from trained neural networks. Pattern Recognition, 1999, 32(12): 1999-2009
[50] Maire F. Rule-extraction by backpropagation of polyhedra. Neural Networks, 1999, 12(4-5): 717-725
[51] Taha I A, Ghosh J. Symbolic interpretation of artificial neural networks. IEEE Transactions on Knowledge and Data Engineering, 1999, 11(3): 448-463
[52] Setiono R. Extracting M-of-N rules from trained neural networks. IEEE Transactions on Neural Networks, 2000, 11(2): 512-519
[53] Setiono R, Leow W K. FERNN: an algorithm for fast extraction of rules from neural networks. Applied Intelligence, 2000, 12(1-2): 15-25
[54] Tsukimoto H. Extracting rules from trained neural networks. IEEE Transactions on Neural Networks, 2000, 11(2): 377-389
[55] Zhou Z-H, Chen S-F, Chen Z-Q. A statistics based approach for extracting priority rules from trained neural networks. In: Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks, Como, Italy, 2000, vol.3, 401-406
[56] Duch W, Adamczak R, Grabczewski K. A new methodology of extraction, optimization and application of crisp and fuzzy logical rules. IEEE Transactions on Neural Networks, 2001, 12(2): 277-306
[57] Saito K, Nakano R. Extracting regression rules from neural networks. Neural Networks, 2002, 15(10): 1279-1288
[58] Castro J L, Mantas C J, Benitez J M. Interpretation of artificial neural networks by means of fuzzy rules. IEEE Transactions on Neural Networks, 2002, 13(1): 101-116
[59] Setiono R. Extracting rules from pruned neural networks for breast cancer diagnosis. Artificial Intelligence in Medicine, 1996, 8(1): 37-51
[60] Hayashi Y, Setiono R, Yoshida K. A comparison between two neural network rule extraction techniques for the diagnosis of hepatobiliary disorders. Artificial Intelligence in Medicine, 2000, 20(3): 205-216
[61] Craven M W, Shavlik J W. Understanding time-series networks: a case study in rule extraction. International Journal of Neural Systems, 1997, 8(4): 373-384
[62] Fu L. A neural-network model for learning domain rules based on its activation function characteristics. IEEE Transactions on Neural Networks, 1998, 9(5): 787-795
[63] Setiono R, Thong J Y L, Yap C. Symbolic rule extraction from neural networks: an application to identifying organizations adopting IT. Information and Management, 1998, 34(2): 91-101
[64] Baesens B, Setiono R, Mues C, Vanthienen J. Using neural network rule extraction and decision tables for credit risk evaluation. Management Science, 2003, 49(3): 312-329
[65] Duch W, Adamczak R, Grabczewski K. Extraction of crisp logical rules using constrained backpropagation networks. In: Proceedings of the International Conference on Neural Networks, Houston, TX, 1997, 2384-2389
[66] Quinlan J R. Introduction of decision trees [ J ]. Ma2 chine Learning, 1986 (1) : 84 - 100
[67] J.M. Tarela, E. Alonso and M.V. Martinez, A representation method for PWL functions oriented to parallel processing, Mathl. Comput. Modelling 13 (lo), 5-83 (1990)
[68] R. Setiono, J.Y.L Thong, An approach to generate rules from neural networks for regression problems. European Journal of Operational Research, 155 (2004) 239-250
[69] J.M. Tarela, K. Basterretxea, I. Del Campo, M.V. Martinez, E. Alonso, Optimised PWL Recursive Approximation and its Application to Neuro-Fuzzy Systems, Mathematical and Computer Modeling, 35 (2002) 867-883
[70] Andrews R, Diederich J, Tickle A B. Andrews Survey and critique of techniques for extracting rules from trained artificial neural networks[J]. Knowledge-Based Systems, 1995, 8(6): 373-389
[71] Blum A, Chawla S. Learning from Labeled and Unlabeled Data using Graph Mincuts[C]. Proc. 18th International Conf. on Machine Learning, San Mateo: Morgan Kaufmann, 2001, 19-26
[72]赵玲玲,翁苏明,曾华军译.模式分析的核方法[M].机械工业出版社,2006
[73] Formmann K M,Paramonov V A.Vertical a new process for t he hot-dip coating of steel sheet[A].The Use and Manufacture of Zinc and Zinc Alloy Coated Sheet Steel Products into 21st Century[C].Chicago:The Iron and Steel Society,1995,189～192
[74]何晓群,刘文卿.应用回归分析[M].中国人民大学出版社,2007
[75] Belkin M,Niyogi P,Sindhwani V.Manifold Regularization: A geometric framework for learning from examples[J].Journal of Machine Learning Research.2006,7(11):2399-2434
[76] Towell G G,Shavlik J W.Knowledge-based artificial neural networks[J].Artificial Intelligence,1994,70(1-2):119-165
[77] Benitez J M, Castro J L, Requena I. Are artificial neural networks black boxes? IEEE Transactions on Neural Networks, 1997, 8(5): 1156-1164
[78] Setiono R.Extracting rules from neural networks by pruning and hidden unit splitting [J]. Neural Computation, 1997,(9): 205-225.
[79] Saito K, Nakano R. Rule extraction from facts and neural networks. In: Proceedings of the International Neural Network Conference, Paris, France, 1990, 379-382
[80] J. Ross Quinlan, C4.5: Programs for Machine Learning, Morgan Kaufmann, San Mateo, California, 1993
[81] Murphy P M, Pazzani M J. ID2-of-3: constructive induction of M-of-N concepts for discriminators in decision trees. In: Proceedings of the 8th International Workshop on Machine Learning, Evanston, IL, 1991, 183-187
[82] Wu X. Knowledge Acquisition from Databases, Norwood, NJ: Ablex, 1995
[83] Palade V, Negoita G, Ariton V. Genetic algorithm optimization of knowledge extraction from neural networks. In: Proc of the 6th Conf on Neural Information Processing. Perth, Australia, 1999, 752-758
[84]宋秦豹,沈钧毅.神经网络数据挖掘中的数据准备问题[J].计算机工程与应用,2002,12:102-104
[85] Stone M. Cross-validatory choice and assessment of statistical predictions[J]. Journal of the Royal Statistical Society, 1974, 36(2): 111-147
[86] Schenker B, Agarwal M. Cross-validated Structure Selection for Neural Networks[J]. Computers chem, 1996, 20(2): 175-186
[87] Andersen T, Martinez T. Cross Validation and MLP Architecture Selection[C]. Proceedings of the IEEE International Joint Conference on Neural Network, Washington DC: IEEE, 1999, 1614-1619
[88] Setiono R. A penalty-function approach for pruning feedforward neural networks[J]. Neural Computation,1997,9(1):185-204
[89] Hagan M T, Menhaj M B. Training feedforward networks with the Marquardt algorithm[J]. IEEE Transactions on Neural Networks, 1994, 5(6): 989-993
[90] UCI–MLR: University of California Irvine Machine Learning Repository. California Irvine, Tex: University of California Irvine Libraries, 1987[2007]. http://archive.ics.uci.edu/ml/index.html
[91]陈兆乾,刘宏,周戎,陈世福.一种混合型多概念获取算法HMCAP及其应用.计算机学报,1996,19(10):753-761
[92]李开泰,黄艾香.张量分析及其应用[M].科学出版社,2005
[93] Muslea I, Minton S, Craig K A. Active+Semi-Supervised Learning=Robust Multi-View Learning[C]. 19th International Conference. on Machine Learning, Sydney, Australia: Morgan Kaufmann Publishers, 2002, 435-442
[94] Setiono R. Extracting rules from neural networks by pruning and hidden unit splitting[J]. Neural Computation, 1997, (9): 205-225.
[95]万百五.工业大系统优化与产品质量控制[M].北京:科学出版社,2003
[96]万维汉,万百五,杨金义.闪速炉的神经网络冰镍质量模型与稳态优化控制研究[J].自动化学报,1999,25(6):800-803
[97]刘国海,张浩,戴先中.神经网络逆系统在电机变频调速系统中的应用[J].电工技术学报,2003,18 (3) :67-71
[98]桑保华,薛晓中.多变量解耦控制方法[J].火力与指挥控制,2007,32(11):245-248
[99]常玉清,李玉朝,吕哲,王福利.基于两级神经网络的发酵过程多变量前馈解耦控制[J].东北大学学报,2007,28(7):925-928
[100]张晓婕.多变量时变系统CARMA模型近似解耦法[J].中国计量学院学报,2004,15(4):284-286
[101]王晰,李少远.多模型分层递阶自适应前馈解耦控制器[J].控制与决策,2005,20(1):17-22
[102]柴天佑.多变量自适应解耦控制及应用[M ].北京:科学出版社,2001
[103]张杰,杨翠娥.电加热锅炉系统神经网络PID解耦控制器[J].系统工程理论与实践,2002,22(1):123-127
[104]周涌,陈庆伟.基于动态神经网络解耦线性化的内模控制[J].南京理工大学学报,2004,28(6):566-570
[105]梁伟平,张健宇.基于神经网络的除氧器水位解耦控制[J].华北电力大学学报,2005,32(1):66-68
[106]何关钰.限制解耦控制的充分必要条件[J].控制理论与应用,1998,15(1):125-129
[107]杨辉,王金章.多变量解耦模糊控制器的研究[J].控制与决策,1988,3 (1):17-21
[108] Czogalaand E,Zimmermann H J.Some aspects of synthesis of probabilistic fuzzy controllers[J]. Fuzzy Sets and Systems,1984,13(2): 169-178
[109] Kiszka J B,Gupta M M,Trojan G M.Multivariable fuzzy controller under godel a implicat ion[J].Fuzzy Sets and Systems,1990,34(3):301-321
[110] Walichiewicz L.Decomposition of linguistic rule in the design of a multidimensional fuzzy control algorithm[J].Cybernetics SystRes,1984,2 ( 3) :185-193
[111]任新宇,樊思齐.航空发动机多变量自学习模糊解耦控制[J].推进技术,2004,25(6):535-537
[112]邱焕耀,毛宗源.采用模糊控制的感应电动机解耦变结构控制系统的研究[J].自动化学报,1998,24(3):391-394
[113]林莉军.宝钢热镀锌机组锌层重量控制模型的应用[J].宝钢技术,2007,4: 33?36
[114]王宸煜,王敏,董溯攀.人工神经网络在镀锌钢板点焊性能估测中的应用[J].上海交通大学学报,2001,35(3):420?423
[115] Sanchez A P, Blanco I D, Vega A A C, et al. Virtual Sensor Design for Coating Thickness Estimation in a Hot Dip Galvanising Line Based on Interpolated SOM Local Models[C]. IEEE Industrial Electronics Society, 2002 28th Annual Conference of the Industrial Electronics Society, New York: IEEE, 2002, 1584?1589
[116]姚林,阳建宏,何飞,等.基于核偏最小二乘的锌层重量预测模型[J].控制工程,2008,15(2):154?157
[117]姚林,阳建宏,徐金梧,等.基于偏最小二乘回归模型的带钢热镀锌质量监控方法[J].北京科技大学学报,2007,29(6):627-631