基于智能方法的冷轧板形信号分析及板形模式识别
|
摘要
板形模式识别是板形控制系统的重要组成部分，其关键技术包括板形基本模式选择、板形特征参数个数确定和模式识别算法。在国家“十一五”科技支撑计划课题和国家自然科学基金项目的资助下，以非接触式板形仪为基础，对非接触式板形仪信号的消噪处理进行了研究，接着板形模式识别采用人工神经网络技术，以及板形控制策略进行了研究，取得了一定成果。
考虑到非接触式板形仪的板形信号为非直接信号，信号处理的技术要求高，本文采用小波消噪技术进行板形原始信号的预处理。板形模式识别结合轧机板形控制手段的多样化，板形基本模式选择使用1次、2次、3次和4次勒让德多项式，含有3次板形的新模式识别方法，从一定程度上提高了板形模式识别的精度。在板形模式识别算法研究方面，首次采用自适应混沌粒子群算法，通过优化BP网络结构及权值、阈值建立的新的板形模式识别模型，提高了板形识别的精度和速度。仿真结果表明新建网络的抗干扰能力和自学习适应能力强、精度高，为板形控制策略的制定提供了准确有效的板形信息。
以HC轧机为例，基于自适应混沌粒子群BP网络的实测结果，制定了缺陷板形反馈控制策略。基于VC++软件构建了仿真平台，用户可通过可视化界面进行相关参数设置，系统将自动对缺陷板形进行识别，并给出缺陷板形的I值分布及相应的反馈控制策略，实验分析表明该系统的识别精度高，可用于在线板形闭环反馈控制。
Flatness pattern recognition is an important part of flatness control system. The keytechnology includes the choice of flatness basic model, the choice of the number offlatness parameters and pattern recognition algorithms. With the funds from national"Eleventh Five-Year" Technology Support Program and the National Natural ScienceFoundation project, using Artificial neural network that based on non-contact Sheet ShapeMeasurement, author research on flatness pattern recognition and control program,achieving some research findings.
The flatness signal from non-contact Sheet Shape Measurement is not directly signal,it requeses high-lever signal processing technique, and then deals with the noiseelimination using wavelet analysis.In terms of selection of flatness basic models,considering a variety of methods of flatness control of modern mills, a new flatnesspattern recognition method including the cubic flatness is first applied in this paper, whichcan improve accuracy of flatness pattern recognition. The model processes uses the linear,quadratic, cubic, and biquadratic Legendre polynomials as basic patterns. In terms ofpattern recognition algorithms, author chooses self-adaptive chaotic PSO algorithm. Thisalgorithm is used to optimize the structure, the weights and the threshold ofback-propagation(BP) network, which can improve the accuracy and convergence rate ofthe neural networks. Simulation results show that the anti-interference ability andself-learning ability of the ACPSO-BP Network is better. It provides a reliable basis for thedevelopment of flatness control strategy.
For example of the HC mills, developed the feedback control strategy, which is basedon the measured result of ACPSO-BP network. The simulation platform is built based onC++ software. Users can set parameters through visual parametric interface, then thissystem can identify the defect shape automatically and get the I value of the defect shapeand the feedback control strategy clearly. The results show that the system has highaccuracy and can be used for online loop feedback control of flatness after analysis ofexperimental.
考虑到非接触式板形仪的板形信号为非直接信号，信号处理的技术要求高，本文采用小波消噪技术进行板形原始信号的预处理。板形模式识别结合轧机板形控制手段的多样化，板形基本模式选择使用1次、2次、3次和4次勒让德多项式，含有3次板形的新模式识别方法，从一定程度上提高了板形模式识别的精度。在板形模式识别算法研究方面，首次采用自适应混沌粒子群算法，通过优化BP网络结构及权值、阈值建立的新的板形模式识别模型，提高了板形识别的精度和速度。仿真结果表明新建网络的抗干扰能力和自学习适应能力强、精度高，为板形控制策略的制定提供了准确有效的板形信息。
以HC轧机为例，基于自适应混沌粒子群BP网络的实测结果，制定了缺陷板形反馈控制策略。基于VC++软件构建了仿真平台，用户可通过可视化界面进行相关参数设置，系统将自动对缺陷板形进行识别，并给出缺陷板形的I值分布及相应的反馈控制策略，实验分析表明该系统的识别精度高，可用于在线板形闭环反馈控制。
Flatness pattern recognition is an important part of flatness control system. The keytechnology includes the choice of flatness basic model, the choice of the number offlatness parameters and pattern recognition algorithms. With the funds from national"Eleventh Five-Year" Technology Support Program and the National Natural ScienceFoundation project, using Artificial neural network that based on non-contact Sheet ShapeMeasurement, author research on flatness pattern recognition and control program,achieving some research findings.
The flatness signal from non-contact Sheet Shape Measurement is not directly signal,it requeses high-lever signal processing technique, and then deals with the noiseelimination using wavelet analysis.In terms of selection of flatness basic models,considering a variety of methods of flatness control of modern mills, a new flatnesspattern recognition method including the cubic flatness is first applied in this paper, whichcan improve accuracy of flatness pattern recognition. The model processes uses the linear,quadratic, cubic, and biquadratic Legendre polynomials as basic patterns. In terms ofpattern recognition algorithms, author chooses self-adaptive chaotic PSO algorithm. Thisalgorithm is used to optimize the structure, the weights and the threshold ofback-propagation(BP) network, which can improve the accuracy and convergence rate ofthe neural networks. Simulation results show that the anti-interference ability andself-learning ability of the ACPSO-BP Network is better. It provides a reliable basis for thedevelopment of flatness control strategy.
For example of the HC mills, developed the feedback control strategy, which is basedon the measured result of ACPSO-BP network. The simulation platform is built based onC++ software. Users can set parameters through visual parametric interface, then thissystem can identify the defect shape automatically and get the I value of the defect shapeand the feedback control strategy clearly. The results show that the system has highaccuracy and can be used for online loop feedback control of flatness after analysis ofexperimental.
引文
[1]黄敏.基于小波分析和神经网络的板形模式方法[J].模式识别与人工智能, 2005, 27(3):27-31.
[2]李楠.板形模式识别与控制的智能方法研究[D].秦皇岛:燕山大学机械电子工程学科硕士学位论文,2006: 1-21.
[3] A.G. Carlstedt, O.Keijser. Modern Approach to Flatness Measurement and Control in ColdRolling[J]. Iron and Steel Engineer, 1991, (4): 34-39.
[4]邸洪双,张晓峰,刘相华.冷轧薄带板形检测信号正交多项式分解及数学模型[J].钢铁,1995, 3(9) 33-36.
[5]周旭东,王国栋.冷轧板形正交多项式分解模型[J].钢铁, 1997, 32(8): 46-47.
[6] Ikuya Hoshino, Masateru Kawai, Misao Kekubo, et al. Observer-Based Multivariable FlatnessControl of the Cold Rolling Mill[C].12thWorld International Federation of Automatic Control.Evolutionary programming springier, 1993, (6): 149-156.
[7]刘进.冷轧带钢板形缺陷表达式回归及数学模型[J].轧钢, 1996, (5): 5-9.
[8] Peng Yan, Liu Hongmin. A Neural Network Recognition Method of Shape Pattern[J]. Journal ofIron and Steel Research International, 2001, 8(1): 16-20.
[9]张秀玲.冷带轧机板形智能识别与智能控制研究[D].秦皇岛:燕山大学机械设计及理论学科硕士学位论文,2002: 20-45.
[10]单修迎,刘宏民,贾春玉.含有三次板形的新型板形模式识别方法[J].钢铁研究学报, 2010,45(8): 56-60.
[11]王益群,尹国芳,孙旭光.板形信号模式识别方法的研究[J].机械工程学报, 2003, 39(8):91-94.
[12] Zhao Xiao-Yan. Flatness pattern recognition based on a binary tree hierarchical BP model[J].Joural of Science and Technology Beijing, 2009, 31(2): 261-265.
[13]刘建,王益群,孙楠.基于粒子群理论的板形模糊模式识别方法[J].机械工程学报, 2008,44(1): 173-178.
[14] J. Kennedy, R. C. Eberhart. Particle swarm optimization[C]. IEEE International Conference onNeural Networks, Perth, Australia, 1995: 1942-1948.
[15] Saravaman, N. Waagen, D.Eiben. Genetic algorithm and particle swarm optimization[J].Evolutionary programming springer, 1998, (7): 611-616.
[16] W Z Lu, H Y Fan, A Y T Leung. Analysis of pollutant levels in central Hong Kong applying neuralnetwork method with particle swarm optimization[J]. Environmental Monitoring and Assessment,2002, 79: 217-230.
[17] R.C.Eberart, Y.H.Shi. Tracking and optimizing dynamic systems with particle swarms[C]. IEEEInternational Conference on Evolutionary Computation, 2001: 94-100.
[18] R.Eberhart, J. Kennedy. A new optimizer using particle swarm theory[C]. The sixth internationalsymposium on micro machine and human science, Nagoya,Japan, 1995: 39-43.
[19] Jacob Robinson, Yahya Rahmat, Rahmat-Samii, et al. Particle swarm optimization inelectromagnetics[C]. IEEE Transaction on antennas and propagation, 2004, 52(2): 397-497.
[20]任海鹏,刘丁,郑岗.一种基于遗传算法的板形模式识别方法[J].重型机械, 2002, (3): 9-12.
[21]张材,谭建平.基于遗传算法-反向传播模型的板形模式识别[J].中南大学学报, 2006, 37(2):294-299.
[22]张秀玲,张志强. DHNN优化设计新方法及在板形模式识别的应用[J].智能系统学报, 2008,3(3): 250-253.
[23] Jia Chunyu, Shan Xiuying, Liu Hongmin, et al. Fuzzy neural model for flatness patternrecognition[J]. Journal of Iron and Steel Research, International, 2008, 15(6): 33-38.
[24]张秀玲,陈丽萍,季颖,等.基于径向函数神经网络的板形模式识别研究[J].工业仪表与自动化装置, 2009, 3(3): 7-9.
[25]冯晓华,马坚,郑岗.基于模糊距离的RBF神经网络板形模式识别[J].西安工业大学学报,2006, 26(5):427-430.
[26]张清东,黄纶伟,周晓敏.宽带钢轧机板形控制技术比较研究[J].北京科技大学学报, 2000,22(2): 177-181.
[27]许健勇.薄板冷轧厚度与板形高精度控制技术[J].钢铁, 2002, 37(1): 73-78.
[28] Claire, Nappez, etal. Control of strip flatness in cold rolling [J]. Iron and Steel Engineer, 1997, 4:42-45.
[29]刘宏民,连家创,段振勇.带材轧制前后张力横向分布的研究[J].重型机械, 1989, (4): 12-20.
[30]李本利.液压涨形轧辊凸度的弹性有限元分析[J].锻压机械, 1993, (5): 24-26.
[31] A.Tomizawa, T.Masui.可变凸度轧辊的最新应用[C].第六届国际轧钢会议文集, 1995: 67-74.
[32]武红林.现代板形调节机构[J].轻合金加工技术, 1996, 24(I): 35-40.
[33] B.Korch, B.Straub, L.Wirtz.四辊冷轧可逆轧机使用VC轧辊的经验合生产效果[J].国外钢铁,1996, (II): 47-51.
[34]吴家观. HC轧机的研究及应用概况[J].武汉钢铁学院学报, 1993, 16(3): 34-36.
[35]白振华,李兴东,杨杰. VC轧辊凸度的形成及板形控制能力的研究[J].上海金属, 2002,24(5): 25-27.
[36] W. Bald. CVC Technology for Cold Rolling Mills-Plant Examples[J]. Iron and Steel Eng, 1988,(5): 24-28.
[37]陶红勇,王京,陆秀志.神经网络在板形控制中的应用[J].轧钢, 2003, (4): 10-12.
[38]胡小平,毛征宇,胡燕平.基于人工神经网络的一种板形反馈控制[J].制造业自动化, 2001,23(3): 40-14,45.
[39]周晓敏,张清东,王长松.基于Hopfield神经网络的板形预测控制模型[J].上海金属, 2007,29(2): 44-47.
[40]王粉花,孙一康,王新平.基于免疫算法的模糊神经网络在板厚板形控制中的应用[J].信息与控制, 2004, 33(4): 504-507.
[41]王智洁.基于心电和脑电数据融合的人体精神疲劳检测研究[D].镇江:江苏大学检测技术与自动化装置学科硕士学位论文,2011: 1-82.
[42]王芳.小波分析在信号去噪中的应用研究[D].西安:西华大学电力电子与电力工程学科硕士学位论文,2009: 2-12.
[43]韦力强.基于小波变换的信号去噪研究[D].长沙:湖南大学电工理论与新技术学科硕士学位论文,2007: 3-30.
[44]王国栋.板形控制和板形理论[M].北京:冶金工业出版社, 2000: 50-166.
[45]邵国强.冷连轧机板形智能控制系统的研究[D].沈阳:东北大学控制理论与控制工程学科硕士学位论文,2008: 5-18.
[46]邹恩,李祥飞,张泰山.混沌控制综述[J].株洲工学院学报, 2002, 16(4): 16-18.
[47]周会锋.板形识别.预测和控制仿真的智能方法研究[D].秦皇岛:燕山大学机械设计理论学科硕士学位论文,2006: 1-82.
[48]黄美灵.群智能算法在智能交通中的研究与应用[D].重庆:重庆交通大学桥梁与隧道工程学科硕士学位论文,2010: 1-82.
[49] Jia Chunyu, Shan Xiuying, Liu Hongmin, etc. Fuzzy neural model for flatness patternrecognition[J]. Journal of Iron and Steel Research International, 2008, 15(6): 33-38.
[50]张泽旭.神经网络控制与MATLAB仿真[M].哈尔滨:哈尔滨工业大学出版社, 2011:673-768.
[51]周蕾.粒子群算法的改进及其人工神经网络中的应用[D].西安:西安电子科技大学控制理论与控制学科硕士学位论文,2010: 1-82.
[52]何宁.江苏电视台播出部信息管理系统设计[D].南京:东南大学测试计量技术及仪器,2002:1-82.
[53]李美领.冷轧板形在线模式识别及反馈控制策略研究[D].秦皇岛:燕山大学机械设计及理论学科硕士学位论文,2010: 1-82.
[54]彭艳,刘宏民,段婷婷. HC非对称弯辊非对称横移板形控制方法:中国,CN200910227863.1[P].2009-06-02.
[55]徐乐江.板带冷轧板形控制与机型选择[M].北京:冶金工业出版社, 2007: 229-238.
[2]李楠.板形模式识别与控制的智能方法研究[D].秦皇岛:燕山大学机械电子工程学科硕士学位论文,2006: 1-21.
[3] A.G. Carlstedt, O.Keijser. Modern Approach to Flatness Measurement and Control in ColdRolling[J]. Iron and Steel Engineer, 1991, (4): 34-39.
[4]邸洪双,张晓峰,刘相华.冷轧薄带板形检测信号正交多项式分解及数学模型[J].钢铁,1995, 3(9) 33-36.
[5]周旭东,王国栋.冷轧板形正交多项式分解模型[J].钢铁, 1997, 32(8): 46-47.
[6] Ikuya Hoshino, Masateru Kawai, Misao Kekubo, et al. Observer-Based Multivariable FlatnessControl of the Cold Rolling Mill[C].12thWorld International Federation of Automatic Control.Evolutionary programming springier, 1993, (6): 149-156.
[7]刘进.冷轧带钢板形缺陷表达式回归及数学模型[J].轧钢, 1996, (5): 5-9.
[8] Peng Yan, Liu Hongmin. A Neural Network Recognition Method of Shape Pattern[J]. Journal ofIron and Steel Research International, 2001, 8(1): 16-20.
[9]张秀玲.冷带轧机板形智能识别与智能控制研究[D].秦皇岛:燕山大学机械设计及理论学科硕士学位论文,2002: 20-45.
[10]单修迎,刘宏民,贾春玉.含有三次板形的新型板形模式识别方法[J].钢铁研究学报, 2010,45(8): 56-60.
[11]王益群,尹国芳,孙旭光.板形信号模式识别方法的研究[J].机械工程学报, 2003, 39(8):91-94.
[12] Zhao Xiao-Yan. Flatness pattern recognition based on a binary tree hierarchical BP model[J].Joural of Science and Technology Beijing, 2009, 31(2): 261-265.
[13]刘建,王益群,孙楠.基于粒子群理论的板形模糊模式识别方法[J].机械工程学报, 2008,44(1): 173-178.
[14] J. Kennedy, R. C. Eberhart. Particle swarm optimization[C]. IEEE International Conference onNeural Networks, Perth, Australia, 1995: 1942-1948.
[15] Saravaman, N. Waagen, D.Eiben. Genetic algorithm and particle swarm optimization[J].Evolutionary programming springer, 1998, (7): 611-616.
[16] W Z Lu, H Y Fan, A Y T Leung. Analysis of pollutant levels in central Hong Kong applying neuralnetwork method with particle swarm optimization[J]. Environmental Monitoring and Assessment,2002, 79: 217-230.
[17] R.C.Eberart, Y.H.Shi. Tracking and optimizing dynamic systems with particle swarms[C]. IEEEInternational Conference on Evolutionary Computation, 2001: 94-100.
[18] R.Eberhart, J. Kennedy. A new optimizer using particle swarm theory[C]. The sixth internationalsymposium on micro machine and human science, Nagoya,Japan, 1995: 39-43.
[19] Jacob Robinson, Yahya Rahmat, Rahmat-Samii, et al. Particle swarm optimization inelectromagnetics[C]. IEEE Transaction on antennas and propagation, 2004, 52(2): 397-497.
[20]任海鹏,刘丁,郑岗.一种基于遗传算法的板形模式识别方法[J].重型机械, 2002, (3): 9-12.
[21]张材,谭建平.基于遗传算法-反向传播模型的板形模式识别[J].中南大学学报, 2006, 37(2):294-299.
[22]张秀玲,张志强. DHNN优化设计新方法及在板形模式识别的应用[J].智能系统学报, 2008,3(3): 250-253.
[23] Jia Chunyu, Shan Xiuying, Liu Hongmin, et al. Fuzzy neural model for flatness patternrecognition[J]. Journal of Iron and Steel Research, International, 2008, 15(6): 33-38.
[24]张秀玲,陈丽萍,季颖,等.基于径向函数神经网络的板形模式识别研究[J].工业仪表与自动化装置, 2009, 3(3): 7-9.
[25]冯晓华,马坚,郑岗.基于模糊距离的RBF神经网络板形模式识别[J].西安工业大学学报,2006, 26(5):427-430.
[26]张清东,黄纶伟,周晓敏.宽带钢轧机板形控制技术比较研究[J].北京科技大学学报, 2000,22(2): 177-181.
[27]许健勇.薄板冷轧厚度与板形高精度控制技术[J].钢铁, 2002, 37(1): 73-78.
[28] Claire, Nappez, etal. Control of strip flatness in cold rolling [J]. Iron and Steel Engineer, 1997, 4:42-45.
[29]刘宏民,连家创,段振勇.带材轧制前后张力横向分布的研究[J].重型机械, 1989, (4): 12-20.
[30]李本利.液压涨形轧辊凸度的弹性有限元分析[J].锻压机械, 1993, (5): 24-26.
[31] A.Tomizawa, T.Masui.可变凸度轧辊的最新应用[C].第六届国际轧钢会议文集, 1995: 67-74.
[32]武红林.现代板形调节机构[J].轻合金加工技术, 1996, 24(I): 35-40.
[33] B.Korch, B.Straub, L.Wirtz.四辊冷轧可逆轧机使用VC轧辊的经验合生产效果[J].国外钢铁,1996, (II): 47-51.
[34]吴家观. HC轧机的研究及应用概况[J].武汉钢铁学院学报, 1993, 16(3): 34-36.
[35]白振华,李兴东,杨杰. VC轧辊凸度的形成及板形控制能力的研究[J].上海金属, 2002,24(5): 25-27.
[36] W. Bald. CVC Technology for Cold Rolling Mills-Plant Examples[J]. Iron and Steel Eng, 1988,(5): 24-28.
[37]陶红勇,王京,陆秀志.神经网络在板形控制中的应用[J].轧钢, 2003, (4): 10-12.
[38]胡小平,毛征宇,胡燕平.基于人工神经网络的一种板形反馈控制[J].制造业自动化, 2001,23(3): 40-14,45.
[39]周晓敏,张清东,王长松.基于Hopfield神经网络的板形预测控制模型[J].上海金属, 2007,29(2): 44-47.
[40]王粉花,孙一康,王新平.基于免疫算法的模糊神经网络在板厚板形控制中的应用[J].信息与控制, 2004, 33(4): 504-507.
[41]王智洁.基于心电和脑电数据融合的人体精神疲劳检测研究[D].镇江:江苏大学检测技术与自动化装置学科硕士学位论文,2011: 1-82.
[42]王芳.小波分析在信号去噪中的应用研究[D].西安:西华大学电力电子与电力工程学科硕士学位论文,2009: 2-12.
[43]韦力强.基于小波变换的信号去噪研究[D].长沙:湖南大学电工理论与新技术学科硕士学位论文,2007: 3-30.
[44]王国栋.板形控制和板形理论[M].北京:冶金工业出版社, 2000: 50-166.
[45]邵国强.冷连轧机板形智能控制系统的研究[D].沈阳:东北大学控制理论与控制工程学科硕士学位论文,2008: 5-18.
[46]邹恩,李祥飞,张泰山.混沌控制综述[J].株洲工学院学报, 2002, 16(4): 16-18.
[47]周会锋.板形识别.预测和控制仿真的智能方法研究[D].秦皇岛:燕山大学机械设计理论学科硕士学位论文,2006: 1-82.
[48]黄美灵.群智能算法在智能交通中的研究与应用[D].重庆:重庆交通大学桥梁与隧道工程学科硕士学位论文,2010: 1-82.
[49] Jia Chunyu, Shan Xiuying, Liu Hongmin, etc. Fuzzy neural model for flatness patternrecognition[J]. Journal of Iron and Steel Research International, 2008, 15(6): 33-38.
[50]张泽旭.神经网络控制与MATLAB仿真[M].哈尔滨:哈尔滨工业大学出版社, 2011:673-768.
[51]周蕾.粒子群算法的改进及其人工神经网络中的应用[D].西安:西安电子科技大学控制理论与控制学科硕士学位论文,2010: 1-82.
[52]何宁.江苏电视台播出部信息管理系统设计[D].南京:东南大学测试计量技术及仪器,2002:1-82.
[53]李美领.冷轧板形在线模式识别及反馈控制策略研究[D].秦皇岛:燕山大学机械设计及理论学科硕士学位论文,2010: 1-82.
[54]彭艳,刘宏民,段婷婷. HC非对称弯辊非对称横移板形控制方法:中国,CN200910227863.1[P].2009-06-02.
[55]徐乐江.板带冷轧板形控制与机型选择[M].北京:冶金工业出版社, 2007: 229-238.