神经网络优化理论研究及应用
|
摘要
神经网络具有很强的适应于复杂环境和多目标控制要求的自学习能力，能够以任意精度逼近任意非线性函数，在控制领域得到了广泛的应用。
本文针对前向网络中的两种典型模型：BP网络和RBF网络的全局优化进行研究，提出了一套系统的前向网络优化设计方法。设计了一种实用的GA实数编码方案，并对标准遗传算子做出改进，进行网络拓扑结构的优化设计，通过仿真验证，证明该优化设计方法是行之有效的。
板形是衡量冷轧薄带钢的质量指标之一，板形模式识别则是板形控制的关键。本文将优化网络用于板形信号的模式识别，建立了6输入、3输出的识别网络模型，该网络性能在训练过程中始终保持最优，能够达到最佳结构，加快了学习速度和训练精度，可以快速、准确求出板形缺陷的模式信息及数值大小，为后续板形控制调节量的设定提供了可靠依据。
液压弯辊是板形控制系统（AFC）最基本的环节，它的动态特性和稳态性能对于整个AFC系统的性能起着至关重要的作用。将优化网络用于液压弯辊系统的控制中，采用内模控制方案，辨识器和控制器用优化网络来离线设计、在线调整，提高了液压弯辊系统的动态响应速度和稳态跟踪精度，充分发挥了液压弯辊力对板形的调整作用，改善了轧机系统的动态特性。
Neural Network owns strongly self-learning ability, which can adapt to complex conditions and meet multi-target controlling requests. And it is able to approximate arbitrarily any nonlinear function well. As a result, Neural Network is widely applied in controlling field.
In this paper, the central purpose is optimizing multi-layer feed-forward neural network globally. An applied coding scheme of genetic algorithm is proposed and the genetic operators are improved to optimize the BPNN and RBFNN's topology structures. The simulation results indicate that these optimizing methods are effective.
Shape patterns recognition is the key to the shape control. Based on the neural network optimizing theory, the shape signals recognition network with 6-input and 3-output is established. In the training this network's topology structure and mapping function are the most optimal all along. And the learning velocity and precision are improved. With this recognition method shape pattern information and magnitude can be received rapidly and exactly, which can provide reliable data for later shape controlling.
Hydraulic bend roller is the basic segment of AFC system. Its dynamic and static characteristic is important to the whole AFC. For this a neural network IMC strategy is applied. And its NNC and NNI are both established with the optimal neural network. So the hydraulic bend roller system's dynamic response velocity and static tracking precision are advanced. Hydraulic bend roller force can operate better and mill's dynamic characteristic is improved.
本文针对前向网络中的两种典型模型：BP网络和RBF网络的全局优化进行研究，提出了一套系统的前向网络优化设计方法。设计了一种实用的GA实数编码方案，并对标准遗传算子做出改进，进行网络拓扑结构的优化设计，通过仿真验证，证明该优化设计方法是行之有效的。
板形是衡量冷轧薄带钢的质量指标之一，板形模式识别则是板形控制的关键。本文将优化网络用于板形信号的模式识别，建立了6输入、3输出的识别网络模型，该网络性能在训练过程中始终保持最优，能够达到最佳结构，加快了学习速度和训练精度，可以快速、准确求出板形缺陷的模式信息及数值大小，为后续板形控制调节量的设定提供了可靠依据。
液压弯辊是板形控制系统（AFC）最基本的环节，它的动态特性和稳态性能对于整个AFC系统的性能起着至关重要的作用。将优化网络用于液压弯辊系统的控制中，采用内模控制方案，辨识器和控制器用优化网络来离线设计、在线调整，提高了液压弯辊系统的动态响应速度和稳态跟踪精度，充分发挥了液压弯辊力对板形的调整作用，改善了轧机系统的动态特性。
Neural Network owns strongly self-learning ability, which can adapt to complex conditions and meet multi-target controlling requests. And it is able to approximate arbitrarily any nonlinear function well. As a result, Neural Network is widely applied in controlling field.
In this paper, the central purpose is optimizing multi-layer feed-forward neural network globally. An applied coding scheme of genetic algorithm is proposed and the genetic operators are improved to optimize the BPNN and RBFNN's topology structures. The simulation results indicate that these optimizing methods are effective.
Shape patterns recognition is the key to the shape control. Based on the neural network optimizing theory, the shape signals recognition network with 6-input and 3-output is established. In the training this network's topology structure and mapping function are the most optimal all along. And the learning velocity and precision are improved. With this recognition method shape pattern information and magnitude can be received rapidly and exactly, which can provide reliable data for later shape controlling.
Hydraulic bend roller is the basic segment of AFC system. Its dynamic and static characteristic is important to the whole AFC. For this a neural network IMC strategy is applied. And its NNC and NNI are both established with the optimal neural network. So the hydraulic bend roller system's dynamic response velocity and static tracking precision are advanced. Hydraulic bend roller force can operate better and mill's dynamic characteristic is improved.
引文
1 D. B. Fogel. An Introduction to Simulated Evolutionary Optimization. IEEE Trans. on Neural Networks, 1994,5(1):3-14
2 X. Yao. A Review of Evolutionary Artificial Neural Networks. Int J Intelligent Systems, 1993,(8):539-567
3 Y. Liu, X. Yao. Evolutionary Design of Artificial Neural Networks with Different Nodes. Proc 1996 IEEE Int Conf Evolutionary Computation. Nagoya, 1996:670-675
4 D. W. Wen, Z. X. Cai. A Composite Structure Learning Method of Neural Networks for Control Application. Proceedings of the Second CW CICIA, Xi'an Jiaotong University Press, 1997:412-416
5 J. W. L. Merrill, R. F. Port. Fractally Configured Neural Networks. Neural networks, 1991,(4):53-60
6 王树清, 等. 先进控制技术及应用. 北京:化学工业出版社, 2001:166
7 P J Antsaklis. Neural Networks in Control Systems. Special Section on Neural Networks for System and Control. IEEE Control System Magazine, 1990:3-5
8 邱东强, 涂亚庆. 神经网络控制的现状与展望. 自动化与仪器仪表, 2001,(5):1-7
9 G. A. Rovithakis, M. A. Chistodoulou. Adaptive Control of Unknowns Plants. IEEE on SMC, 1994,24:400-412
10 王永骥, 涂健. 神经元网络控制. 北京:机械工业出版社, 1999:18-19
11 徐丽娜. 神经网络控制. 哈尔滨:哈尔滨工业大学出版社, 1999:1-3
12 D. Z. Wang, Z. L. Wang. Identification and Control of Induction Motor Using Artificial Neural Network. Proceedings of the 5th International Conference on Electrical Machines and Systems, Beijing: The International Academic Publishers of World Publishing Corporation, 2001:751-754
13 贺毓辛. 冷轧板带生产. 北京:冶金工业出版社, 1992:2-3
14 王国栋. 板形控制和板形理论. 北京:冶金工业出版社, 1986:200-202
15 Ikuya Hoshino, Masateru Kawai. Observer-Based Multivariable Flatness Control of The Cold Rolling Mill. 12th World Congress IFAC, 1993:149-156
16 刘建昌, 顾树生, 高瀛, 等. 板形最优综合控制算法. 东北大学学报, 1996,17(3): 248-251
17 Daniel E. Williams, Wassim M. Haddad. Nonlinear Control of Roll Moment Distribution to Influence Vehicle Yaw Characteristics. IEEE Trans. Control System Technology, 1995,3(1):110-115
18 Yasunori Katayama. A Neural Fuzzy Control System for Rolling Mills, Steel Technology International, London, 1992:189-196
19 周旭东, 李连诗, 等. 自适应神经元网络板形板厚综合控制. 北京科技大学学报, 1994,16(4):340-345
20 乔俊飞, 柴天佑. 板形控制技术的现状及未来发展. 冶金自动化, 1997,1:11-14
21 陈先霖. 新一代高技术薄带冷轧机的发展趋向. 上海金属, 1995,17(4):1-8
22 赵林, 宋岚, 等. 板形控制技术现状及今后发展方向. 轧钢, 1995,3:49-51
23 涂序彦. 钢铁工业生产过程智能自动化. 1995年中国智能自动化学术会议即智能自动化专业委员会成立大会论文集(上). 天津, 1995:77-82
24 Sean J. Egan et al.. Flatness Modeling and Control for A Continuous Tandem Cold Mill. Iron and Steel Engineer, 1996,73(3):38-41
25 丛爽. 神经网络、模糊系统及其在运动控制中的应用. 合肥: 中国科技大学出版社, 2001:13-23
26 申东日, 冯少辉, 陈义俊. BP网络改进方法概述. 化工自动化及仪表, 2000,27(1): 30-32
27 H Muhlenbein. Limitations of Multi-layer Perception Networks-steps Forwards Genetic Neural Network. Parallel Computing, 1991,14:249-260
28 黎明, 严超华, 刘高航. 遗传算法优化前向神经网络结构和权重矢量. 中国图像图形学报, 1999,4(6):491-495
29 李敏强, 徐博艺, 寇纪淞. 遗传算法与神经网络的结合. 系统工程与理论实践. 1999,2(2):65-69
30 周佩玲, 陶小丽, 傅忠谦, 等. 基于遗传算法的RBF网络及应用. 信号处理, 2001,7(3):269-273
31 王凌. 智能优化算法及其应用. 北京: 清华大学出版社, 施普林格出版社, 2001: 36-37
32 陈国良, 王煦法, 庄镇泉. 遗传算法及其应用. 北京: 人民邮电出版社, 1996:5-14
Randall S. Sexton, Jatinder N. D. Gupta. Comparative Evaluation of Genetic Algorithm and Back Propagation for Training Neural Networks. Information
33 Sciences, 2000,129:45-59
34 D. T. Pham, D. Karaboga. Training Elman and Jordan Networks for System Identication Using Genetic Algorithm. Artificial Intelligence in Engineering, 1999(13):107-117
35 Jasmina Arifovic, Ramazan Gencay. Using Genetic Algorithms to Select Architecture of A Feedforward Artificial Neural Network. Physica A, 2001,289:574-594
36 王佳斌, 王晋隆. 用遗传算法优化前馈神经网络的结构. 黎明职业大学学报, 2000,4(12):29-34
37 张敏, 赵金城. 全局优化神经网络拓扑结构及权值的遗传算法. 大连大学学报,
1999,20(6):9-13
38 A. Blanco, M. Delgado, M. C. Pegalajar. A Genetic Algorithm to Obtain the Optimal Recurrent Neural Network. International Journal of Approximate Reasoning, 2000, 23:67-83
39 李兵, 谢剑英. 遗传算法的自适应代沟的替代策略研究. 控制理论与应用, 2001, 18(1):41-44
40 Shouzhi Li, Minyuan Li, Yongxiang Pan. Genetic Annealing Algorithm and It's Convergence Analysis. Control Theory &Appliance, 2002,19(3):376-380
41 B. Runqiang, C. Zengqiang, Yuan. Zhuzhi. Improved Crossover Strategy of Genetic Algorithms and Analysis of Its Performance. Proceedings of the 3rd World Congress on Intelligent Control and Automation, Jun 28-July 2,2000:516-520
42 陈长征, 王楠. 遗传算法中交叉和变异概率选择的自适应方法及作用机理. 控制理论与应用,2002,19(1):41-43
43 Srinivas M, Patnark L M. Adaptive Probabilities of Crossover and Mutation in GA. IEEE Trans Syst, and Cybernetics, 1994,24(4):656-667
44 Steve A. Billings, Guang L. Zheng. Radial Basis Function Network Configuration Using Genetic Algorithms. Neural Networks, 1995,8(6):877-890
45 徐悦, 柴天佑, 毛志忠. 四辊可逆冷轧机板形控制系统的研究. 冶金自动化, 1995, 1:8-12
46 Ikuya Hoshino, Masateru Kawai, Misao Kokubo et al. Observer-Based Multivariable Flatness Control of the Cold Rolling Mill. IFAC, 1992:149
47 连家创, 刘宏民. 板厚板形控制. 北京:兵器工业出版社, 1996:41-42
48 张秀玲. 冷带轧机板形智能识别与智能控制研究. [燕山大学工学博士学位论文]. 2002:13-14
49 华建新, 周泽雁. 冷轧带钢板形缺陷的多项式回归及数学模型. 钢铁, 1992，27(3): 27-31
50 邸洪双, 张晓风, 刘相华等. 冷轧薄带板形检测信号正交多项式分解及数学模型.钢铁, 1995,30(9):33-36
51 杨景明, 王洪瑞, 宋维公. 冷轧薄板板形的数据处理方法研究. 东北重型机械学院学报, 1994,18(2):131-135
52 张云鹏, 王长松, 张清东. 基于效应函数的冷轧板形闭环控制策略. 北京科技大学学报, 1999,21(2):195-197
53 张清东, 陈先霖, 徐金梧. 板形缺陷模式识别方法的研究. 钢铁, 1996,31(增刊): 57-60
54 乔俊飞. UC轧机板形建模与控制方法的研究. [东北大学工学博士学位论文]. 1998:36-54
55 王国栋, 刘相华, 等. 金属轧制过程人工智能优化. 北京: 冶金工业出版社, 2000: 21-23
56 周晓敏, 张清东, 王长松,等. 智能控制和预测控制在冷轧板形自动控制中的应用.上海金属, 2000,22(4):40-43
57 M. Y. Rafiq, G. Bugmann, D. J. Esterbrook. Neural Network Design for Engineering Appliance. Computers and Structure, 2001,79: 1541-1552
58 R. Ciocana, P. Petulescub, D. J. Rothd. The Use of The Neural Networks in The Recognition of The Austenitic Steel Types. NDT&International, 2000,33: 85-89
59 R. S. Guha, Y. C. Hsiehb. A Neural Network Based Model for Abnormal Pattern Recognition of Control Charts. Computers & Industrial Engineering, 1999,36: 97-108
60 乔俊飞, 郭戈, 柴天佑. 板形模式识别的一种模糊识别方法. 钢铁, 1998,33(6): 36-40
61 乔俊飞, 郭戈, 柴天佑. 神经网络在板形检测中的应用. 中国有色金属学报, 1998, 8(9): 551-556
62 Sung-Bae Cho. Pattern Recognition with Neural Networks Combined By Genetic Algorithm. Fuzzy Sets and Systems, 1999,103:339-347
63 张秀玲, 刘宏民. 变结构神经网络在板形信号模式识别方面的应用. 钢铁研究学报, 2001,13(2):62-65
64 程卫国, 冯峰, 王雪梅, 等. MATLAB 5.3精要编程及高级应用. 机械工业出版社, 2000: 325-374
65 彭艳. 基于条元法的HC冷轧机板形预设定控制理论研究及工业应用. [燕山大学工学博士学位论文]. 2000:11-24
66 乔俊飞, 郭戈, 柴天佑. 带材板形自适应控制. 东北大学学报, 1998,19(4):338-341
67 Mason J D. The Measurement and Control of Flatness. Proc. of 5th Int. Rolling Conf., 1990:205-209
68 A G Carkstedt, O Keijser. Modern Approach to Flatness Measurement and Control in Cold Rollin. Iron and Steel Engineer, 1991,68(4):34-37
69 乔俊飞, 孙雅明, 毛鹏. 一种基于神经网络的内模控制方法及其应用. 天津大学学报, 2000,33(1):25-28
70 张秀玲, 刘宏民. 神经网络模型参考自适应控制及其在带材板形控制系统中的应用. 机械工程学报, 2001,37(9):83-87
71 张秀玲, 刘宏民. 液压弯辊系统的智能内模控制. 电工技术学报, 2002,17(1):91-95
72 廖明, 吴宁, 谢品芳. 神经网络内模控制算法的研究. 电气传动自动化, 1998,20(4): 24-28
2 X. Yao. A Review of Evolutionary Artificial Neural Networks. Int J Intelligent Systems, 1993,(8):539-567
3 Y. Liu, X. Yao. Evolutionary Design of Artificial Neural Networks with Different Nodes. Proc 1996 IEEE Int Conf Evolutionary Computation. Nagoya, 1996:670-675
4 D. W. Wen, Z. X. Cai. A Composite Structure Learning Method of Neural Networks for Control Application. Proceedings of the Second CW CICIA, Xi'an Jiaotong University Press, 1997:412-416
5 J. W. L. Merrill, R. F. Port. Fractally Configured Neural Networks. Neural networks, 1991,(4):53-60
6 王树清, 等. 先进控制技术及应用. 北京:化学工业出版社, 2001:166
7 P J Antsaklis. Neural Networks in Control Systems. Special Section on Neural Networks for System and Control. IEEE Control System Magazine, 1990:3-5
8 邱东强, 涂亚庆. 神经网络控制的现状与展望. 自动化与仪器仪表, 2001,(5):1-7
9 G. A. Rovithakis, M. A. Chistodoulou. Adaptive Control of Unknowns Plants. IEEE on SMC, 1994,24:400-412
10 王永骥, 涂健. 神经元网络控制. 北京:机械工业出版社, 1999:18-19
11 徐丽娜. 神经网络控制. 哈尔滨:哈尔滨工业大学出版社, 1999:1-3
12 D. Z. Wang, Z. L. Wang. Identification and Control of Induction Motor Using Artificial Neural Network. Proceedings of the 5th International Conference on Electrical Machines and Systems, Beijing: The International Academic Publishers of World Publishing Corporation, 2001:751-754
13 贺毓辛. 冷轧板带生产. 北京:冶金工业出版社, 1992:2-3
14 王国栋. 板形控制和板形理论. 北京:冶金工业出版社, 1986:200-202
15 Ikuya Hoshino, Masateru Kawai. Observer-Based Multivariable Flatness Control of The Cold Rolling Mill. 12th World Congress IFAC, 1993:149-156
16 刘建昌, 顾树生, 高瀛, 等. 板形最优综合控制算法. 东北大学学报, 1996,17(3): 248-251
17 Daniel E. Williams, Wassim M. Haddad. Nonlinear Control of Roll Moment Distribution to Influence Vehicle Yaw Characteristics. IEEE Trans. Control System Technology, 1995,3(1):110-115
18 Yasunori Katayama. A Neural Fuzzy Control System for Rolling Mills, Steel Technology International, London, 1992:189-196
19 周旭东, 李连诗, 等. 自适应神经元网络板形板厚综合控制. 北京科技大学学报, 1994,16(4):340-345
20 乔俊飞, 柴天佑. 板形控制技术的现状及未来发展. 冶金自动化, 1997,1:11-14
21 陈先霖. 新一代高技术薄带冷轧机的发展趋向. 上海金属, 1995,17(4):1-8
22 赵林, 宋岚, 等. 板形控制技术现状及今后发展方向. 轧钢, 1995,3:49-51
23 涂序彦. 钢铁工业生产过程智能自动化. 1995年中国智能自动化学术会议即智能自动化专业委员会成立大会论文集(上). 天津, 1995:77-82
24 Sean J. Egan et al.. Flatness Modeling and Control for A Continuous Tandem Cold Mill. Iron and Steel Engineer, 1996,73(3):38-41
25 丛爽. 神经网络、模糊系统及其在运动控制中的应用. 合肥: 中国科技大学出版社, 2001:13-23
26 申东日, 冯少辉, 陈义俊. BP网络改进方法概述. 化工自动化及仪表, 2000,27(1): 30-32
27 H Muhlenbein. Limitations of Multi-layer Perception Networks-steps Forwards Genetic Neural Network. Parallel Computing, 1991,14:249-260
28 黎明, 严超华, 刘高航. 遗传算法优化前向神经网络结构和权重矢量. 中国图像图形学报, 1999,4(6):491-495
29 李敏强, 徐博艺, 寇纪淞. 遗传算法与神经网络的结合. 系统工程与理论实践. 1999,2(2):65-69
30 周佩玲, 陶小丽, 傅忠谦, 等. 基于遗传算法的RBF网络及应用. 信号处理, 2001,7(3):269-273
31 王凌. 智能优化算法及其应用. 北京: 清华大学出版社, 施普林格出版社, 2001: 36-37
32 陈国良, 王煦法, 庄镇泉. 遗传算法及其应用. 北京: 人民邮电出版社, 1996:5-14
Randall S. Sexton, Jatinder N. D. Gupta. Comparative Evaluation of Genetic Algorithm and Back Propagation for Training Neural Networks. Information
33 Sciences, 2000,129:45-59
34 D. T. Pham, D. Karaboga. Training Elman and Jordan Networks for System Identication Using Genetic Algorithm. Artificial Intelligence in Engineering, 1999(13):107-117
35 Jasmina Arifovic, Ramazan Gencay. Using Genetic Algorithms to Select Architecture of A Feedforward Artificial Neural Network. Physica A, 2001,289:574-594
36 王佳斌, 王晋隆. 用遗传算法优化前馈神经网络的结构. 黎明职业大学学报, 2000,4(12):29-34
37 张敏, 赵金城. 全局优化神经网络拓扑结构及权值的遗传算法. 大连大学学报,
1999,20(6):9-13
38 A. Blanco, M. Delgado, M. C. Pegalajar. A Genetic Algorithm to Obtain the Optimal Recurrent Neural Network. International Journal of Approximate Reasoning, 2000, 23:67-83
39 李兵, 谢剑英. 遗传算法的自适应代沟的替代策略研究. 控制理论与应用, 2001, 18(1):41-44
40 Shouzhi Li, Minyuan Li, Yongxiang Pan. Genetic Annealing Algorithm and It's Convergence Analysis. Control Theory &Appliance, 2002,19(3):376-380
41 B. Runqiang, C. Zengqiang, Yuan. Zhuzhi. Improved Crossover Strategy of Genetic Algorithms and Analysis of Its Performance. Proceedings of the 3rd World Congress on Intelligent Control and Automation, Jun 28-July 2,2000:516-520
42 陈长征, 王楠. 遗传算法中交叉和变异概率选择的自适应方法及作用机理. 控制理论与应用,2002,19(1):41-43
43 Srinivas M, Patnark L M. Adaptive Probabilities of Crossover and Mutation in GA. IEEE Trans Syst, and Cybernetics, 1994,24(4):656-667
44 Steve A. Billings, Guang L. Zheng. Radial Basis Function Network Configuration Using Genetic Algorithms. Neural Networks, 1995,8(6):877-890
45 徐悦, 柴天佑, 毛志忠. 四辊可逆冷轧机板形控制系统的研究. 冶金自动化, 1995, 1:8-12
46 Ikuya Hoshino, Masateru Kawai, Misao Kokubo et al. Observer-Based Multivariable Flatness Control of the Cold Rolling Mill. IFAC, 1992:149
47 连家创, 刘宏民. 板厚板形控制. 北京:兵器工业出版社, 1996:41-42
48 张秀玲. 冷带轧机板形智能识别与智能控制研究. [燕山大学工学博士学位论文]. 2002:13-14
49 华建新, 周泽雁. 冷轧带钢板形缺陷的多项式回归及数学模型. 钢铁, 1992，27(3): 27-31
50 邸洪双, 张晓风, 刘相华等. 冷轧薄带板形检测信号正交多项式分解及数学模型.钢铁, 1995,30(9):33-36
51 杨景明, 王洪瑞, 宋维公. 冷轧薄板板形的数据处理方法研究. 东北重型机械学院学报, 1994,18(2):131-135
52 张云鹏, 王长松, 张清东. 基于效应函数的冷轧板形闭环控制策略. 北京科技大学学报, 1999,21(2):195-197
53 张清东, 陈先霖, 徐金梧. 板形缺陷模式识别方法的研究. 钢铁, 1996,31(增刊): 57-60
54 乔俊飞. UC轧机板形建模与控制方法的研究. [东北大学工学博士学位论文]. 1998:36-54
55 王国栋, 刘相华, 等. 金属轧制过程人工智能优化. 北京: 冶金工业出版社, 2000: 21-23
56 周晓敏, 张清东, 王长松,等. 智能控制和预测控制在冷轧板形自动控制中的应用.上海金属, 2000,22(4):40-43
57 M. Y. Rafiq, G. Bugmann, D. J. Esterbrook. Neural Network Design for Engineering Appliance. Computers and Structure, 2001,79: 1541-1552
58 R. Ciocana, P. Petulescub, D. J. Rothd. The Use of The Neural Networks in The Recognition of The Austenitic Steel Types. NDT&International, 2000,33: 85-89
59 R. S. Guha, Y. C. Hsiehb. A Neural Network Based Model for Abnormal Pattern Recognition of Control Charts. Computers & Industrial Engineering, 1999,36: 97-108
60 乔俊飞, 郭戈, 柴天佑. 板形模式识别的一种模糊识别方法. 钢铁, 1998,33(6): 36-40
61 乔俊飞, 郭戈, 柴天佑. 神经网络在板形检测中的应用. 中国有色金属学报, 1998, 8(9): 551-556
62 Sung-Bae Cho. Pattern Recognition with Neural Networks Combined By Genetic Algorithm. Fuzzy Sets and Systems, 1999,103:339-347
63 张秀玲, 刘宏民. 变结构神经网络在板形信号模式识别方面的应用. 钢铁研究学报, 2001,13(2):62-65
64 程卫国, 冯峰, 王雪梅, 等. MATLAB 5.3精要编程及高级应用. 机械工业出版社, 2000: 325-374
65 彭艳. 基于条元法的HC冷轧机板形预设定控制理论研究及工业应用. [燕山大学工学博士学位论文]. 2000:11-24
66 乔俊飞, 郭戈, 柴天佑. 带材板形自适应控制. 东北大学学报, 1998,19(4):338-341
67 Mason J D. The Measurement and Control of Flatness. Proc. of 5th Int. Rolling Conf., 1990:205-209
68 A G Carkstedt, O Keijser. Modern Approach to Flatness Measurement and Control in Cold Rollin. Iron and Steel Engineer, 1991,68(4):34-37
69 乔俊飞, 孙雅明, 毛鹏. 一种基于神经网络的内模控制方法及其应用. 天津大学学报, 2000,33(1):25-28
70 张秀玲, 刘宏民. 神经网络模型参考自适应控制及其在带材板形控制系统中的应用. 机械工程学报, 2001,37(9):83-87
71 张秀玲, 刘宏民. 液压弯辊系统的智能内模控制. 电工技术学报, 2002,17(1):91-95
72 廖明, 吴宁, 谢品芳. 神经网络内模控制算法的研究. 电气传动自动化, 1998,20(4): 24-28