基于极限拉深原理的高速智能液压拉深机控制系统的研究
|
摘要
本论文围绕高速智能液压拉深机的控制系统进行了较深入的理论分析和有针对性的实验研究。所提出的基于极限拉深工艺原理的变速、变压边力拉深方式改善了传统拉深工艺中原有的恒力压边方式；在所开发出的监控软件中运用的智能控制方法进一步提高了系统的控制性能，增加了原有液压拉深机的技术含量。本学位论文主要包括以下内容：
首先，提出了采用变速、变压边力进行拉深的极限拉深工艺原理。在详细阐述了国内外液压拉深机发展概况的基础上，通过对金属板料拉深工艺、变形机理进行分析，提出了一种既能够提高板材成形极限，又能在生产中得到广泛应用的新工艺——变速、变压边力拉深，并重点研究了拉深速度、压边力等重要工艺参数对拉深变形的影响。
其次，将PID控制和模糊控制引入高速智能液压拉深机的压边控制中。本文进行了PID控制算法以及模糊控制算法对系统阶跃响应的对比研究，提出采用增量式PID算法能较好地跟踪给定信号，而采用模糊控制算法能进一步提高系统的动态性能。同时，还进行了基于人工神经网络的自适应控制系统的探索性研究。
最后，使用不锈钢板料进行了带法兰边圆筒件拉深试验。通过对超低拉深系数m=0.50下的拉深试验结果分析发现，采用从大到小到稍大的变压边力控制，比传统拉深工艺中原有的恒压边力控制方式更有利于降低拉深系数，提高拉深成形的表面质量，减少工件拉裂的几率。
本论文还应用工业控制计算机研究并开发了液压拉深机监控系统，为液压拉深机的自动化和智能化提供了充分的条件，并以易操作的界面为高速智能液压拉深机创造了更加方便的调试环境。
The thoroughgoing theoretical analysis and the experimental study aimed at the control system of high speed, intelligent controlled hydraulic deep drawing press are carried on in this thesis. Based on the limited drawing technical theory, a way of deep drawing with changeable moving speed and Blank-Holding Force (BHF) is put forward, which improves the traditional mode of drawing with fixed BHF. The control performance is enhanced and the technical content is increased after adopting the intelligent controlling method used in the monitoring and control software, which especially developed for the hydraulic press system. Following substance is included.
First, the limited drawing technical theory used in deep drawing with changeable moving speed and BHF is put forward. Based on the study of deep drawing press' development in the civil and abroad and the distorting theoretical analysis of metal plate, a new technique, that is the way of deep drawing with changeable moving speed and BHF, is put forward. The forming limit can be enhanced and the application can be widely extended by use of this new kind technique. The influence of some important technical parameters such as speed and BHF on deep drawing distortion is also analyzed.
Second, the PID control and fuzzy control method are respectively applied into the BHF control of the hydraulic press. The contrasts to step respond between PID control method and fuzzy control method are studied. The given signs can be precisely followed when PID control method acts, however, the system dynamic performance can be further improved when fuzzy control method has effect. In the meantime, the explorative study of neural network adaptive control system is carried on.
Finally, the experiments on stainless steel cylinder with flange plate work-piece drawing are made. Under the ultra-low drawing coefficient (m=0.5) drawing conditions, the experimental conclusion shows that the BHF control, which reduce from high pressure to low pressure then rises to medium pressure, are more beneficial to reduce the drawing coefficient than the old drawing mode with fixed BHF control. Meanwhile, the surface quality of the work-piece can be improved and the ratio of cracking in the course of drawing can be cut down.
The monitoring and control system with IPC is developed, which provides ample condition for automation and intelligence of the hydraulic press and creates a more convenient and easily operating test environment.
首先，提出了采用变速、变压边力进行拉深的极限拉深工艺原理。在详细阐述了国内外液压拉深机发展概况的基础上，通过对金属板料拉深工艺、变形机理进行分析，提出了一种既能够提高板材成形极限，又能在生产中得到广泛应用的新工艺——变速、变压边力拉深，并重点研究了拉深速度、压边力等重要工艺参数对拉深变形的影响。
其次，将PID控制和模糊控制引入高速智能液压拉深机的压边控制中。本文进行了PID控制算法以及模糊控制算法对系统阶跃响应的对比研究，提出采用增量式PID算法能较好地跟踪给定信号，而采用模糊控制算法能进一步提高系统的动态性能。同时，还进行了基于人工神经网络的自适应控制系统的探索性研究。
最后，使用不锈钢板料进行了带法兰边圆筒件拉深试验。通过对超低拉深系数m=0.50下的拉深试验结果分析发现，采用从大到小到稍大的变压边力控制，比传统拉深工艺中原有的恒压边力控制方式更有利于降低拉深系数，提高拉深成形的表面质量，减少工件拉裂的几率。
本论文还应用工业控制计算机研究并开发了液压拉深机监控系统，为液压拉深机的自动化和智能化提供了充分的条件，并以易操作的界面为高速智能液压拉深机创造了更加方便的调试环境。
The thoroughgoing theoretical analysis and the experimental study aimed at the control system of high speed, intelligent controlled hydraulic deep drawing press are carried on in this thesis. Based on the limited drawing technical theory, a way of deep drawing with changeable moving speed and Blank-Holding Force (BHF) is put forward, which improves the traditional mode of drawing with fixed BHF. The control performance is enhanced and the technical content is increased after adopting the intelligent controlling method used in the monitoring and control software, which especially developed for the hydraulic press system. Following substance is included.
First, the limited drawing technical theory used in deep drawing with changeable moving speed and BHF is put forward. Based on the study of deep drawing press' development in the civil and abroad and the distorting theoretical analysis of metal plate, a new technique, that is the way of deep drawing with changeable moving speed and BHF, is put forward. The forming limit can be enhanced and the application can be widely extended by use of this new kind technique. The influence of some important technical parameters such as speed and BHF on deep drawing distortion is also analyzed.
Second, the PID control and fuzzy control method are respectively applied into the BHF control of the hydraulic press. The contrasts to step respond between PID control method and fuzzy control method are studied. The given signs can be precisely followed when PID control method acts, however, the system dynamic performance can be further improved when fuzzy control method has effect. In the meantime, the explorative study of neural network adaptive control system is carried on.
Finally, the experiments on stainless steel cylinder with flange plate work-piece drawing are made. Under the ultra-low drawing coefficient (m=0.5) drawing conditions, the experimental conclusion shows that the BHF control, which reduce from high pressure to low pressure then rises to medium pressure, are more beneficial to reduce the drawing coefficient than the old drawing mode with fixed BHF control. Meanwhile, the surface quality of the work-piece can be improved and the ratio of cracking in the course of drawing can be cut down.
The monitoring and control system with IPC is developed, which provides ample condition for automation and intelligence of the hydraulic press and creates a more convenient and easily operating test environment.
引文
[1]朱钒．面向大规模定制模式下的高速智能液压拉深机关键技术的研究．华中理工大学博士后研究工作报告，2000．
[2]范宏才，侍慕超，张昶盛．现代锻压机械．机械工业出版社．1994．
[3]朱钒，张志，张道富等．国内外液压机技术现状与发展趋势．机床与液压，2000 No．1
[4]李硕本．冲压工艺学．北京：机械工业出版社，1982
[5]Yossifon S, et al. On the Acceptable Blankholder force Range in the deep Drawing process, J. Mater. process. Techn. 1992, 33
[6]何丹农，陶宏之等．研究拉深成形过程中压边力值的合理确定．金属成形工艺，1999 No．5
[7]马力，张志等．极限拉深原理与极限拉深液压机研究．机械开发，2000 年增刊
[8]林建亚，何存兴．液压元件．机械工业出版社．1988．
[9]刘长年．液压伺服系统的分析与设计．科学出版社，1985．
[10]路甬祥，胡大绂．电液比例控制技术．第1版．机械工业出版社，1988．
[11]黎启柏．电液比例控制与数字控制系统．机械工业出版社，1997．
[12]王春行．液压伺服控制系统．机械工业出版社，1996．
[13]葛宜远．流体传动及控制技术的新成就．液压与气动，1998 No．2
[14]杨勇，黄文梅．一种新的电液伺服系统神经网络自适应控制，湖南大学学报，1998 No．10
[15]李运华，王占林．神经网络控制及其在液压伺服系统中的应用．机床与液压，1996 No．3
[16]熊宇飞，陆元章．多变量电液位置伺服系统的时序建模．机床与液压，1987 No．1
[17]Fu K.S. Learning Control Systems and Intelligent Control System. An Intersection of AI/AC IEEE Trans. On AC 1971.16:70-72
[18]Saridis, G.N. Towards The realization of intelligent Control. Proc.of IEEE. 1979. (67)8:4-11
[19]Saridis,G.N. Knowledge,Implementation,Structure of Intelligent Control System. Proc.of IEEE Int. Sump. On Intelligent Control. 1987.
[20]李宁，张乃尧．采用不同的模糊推理方法时典型模糊控制器的结构分析．模糊系统与数学，1998 Vol．12 No，3
[21]刘永红．神经网络理论的发展与前沿问题．信息与控制，1998 Vol．28 No．1
[22]Hebb D O. The Organization of Behavior, New York; Wiley, 1949
[23]Rosenblatt F. The Perceptron: A Probabilistic Model for Information storage and Organization in the Brain. Psychological Review, 1958, (65): 386～408
[24]Holland J H. Genetic Algorithms and the Optimal Allocations of Trials, SIAM Journal of Computing, 1973, (2): 88～105
[25]Holland J H. Adaptation in Natural and Artificial Systems, Ann Arbor: The University of Michigan Press, 1975
[26]Hopfield J J. Neural Networks and Physical Systems with Emergent Collective Computational Abilities, Proc. Natl. Acad. Sci., USA, 1982, (79): 2254～2558
[27]张立明．人工神经网络的模型及其应用．上海：复旦大学出版社，1992．
[28]袁南儿，王万良，苏宏业．计算机新型控制策略及其应用．北京：清华大学出版社，1998．152～153
[29]毛宗源，姚尹武，胡尚信．变结构神经网络．控制理论与应用，1999 Vol．16 No．1
[30]康景峰，张凤传等．QCS003型液压试验台微机控制及数据采集系统．实验技术与管理，1996 Vol．13 No．1
[31]陈众，方璐，李楠．VC环境下小型工业监控软件的开发．计算机自动测量与控制，2000．8(5)
[32]张乃镇，汤峰，刘建业．采用Visual C++开发的导弹伺服机构测控系统．军用测控技术．2000．8(3)
[33]蔡自兴．智能控制—基础与应用．北京：国防工业出版社．1998
[34]吕冬，丁柯，何丹农等．基于神经网络的拉深力智能化预测系统．中国有色金属学报，2000，Vol．10，No．3
[35]罗亚军，张永清，赵军．板材拉深成形智能化控制过程中摩擦系数的识别．摩擦与润滑，2001，No．2
[36]R C Eberhart R W Dobbbins. Neural Network. Pc tools, Academic Press, 1990.
[37]张立明．人工神经网络的模型及其应用．上海：复旦大学出版社．1994．
[38]Myllykosky Pand Larkiola J. Development of prediction model for mechanical properties of batch annealed thin steel strip by using artificial neural network modeling [J]. J Mater Proc Technol, 1996, 60: 399-404.
[39]卢险峰．冲压工艺模具学．第1版．机械工业出版社，1998．
[40]中川威雄 阿部帮雄 林丰．板料冲压加工．天津科学技术出版社，1982．
[41]姜海峰等．变压边力拉深的原理及实验系统．塑性工程学报．1999(3)，Vol．6，NO．1：31
[42]梁炳文等．弹塑性稳定理论．国防工业出版社，1983
[43]汪大年主编．金属塑性成性原理．机械工业出版社，1982
[44]M．B．斯德洛日夫[俄]．金属压力加工原理．机械工业出版社．1980
[45]中川威雄等．冲压加工技术手册[日]．轻工业出版社，1988．
[46]π．A．肖夫曼[俄]．冷压原理．机械工业出版社，1957．
[47]姚文季．金属板料的压延性能研究．南京航空学院学报，1964，2．
[48]熊志卿，熊焰．拉深极限研究．塑性工程学报．1997．6，Vol．4，No．2
[49]邓明，彭成允，李渝．拉延速度对拉延件破裂的影响．锻压机械，2001(4)．
[50]刘莉莉．拉深系数与冲压力关系的分析．河南职技师院学报，2000(6)，Vol．28，No．2
[51]丁松聚．冷冲模设计[M]．北京：机械工业出版社，1997．5，141-161．
[52]吕雪山，王先进．薄板成形与制造[M]．北京：中国物资出版社．1993．
[53]王东哲，吕东等．板料成形中压边力的控制理论研究．锻压技术，2000年4期
[54]卢险峰．冲压工艺模具学．第1版．机械工业出版社，1998．
[55]刘天骥编译．冲压件缺陷及消除方法．重庆大学出版社，1988．
[56]翁克索夫，约翰逊，工藤英明．金属塑性变形理论．机械工业出版社，1992．
[57]Katsuhiko Ogata著．卢伯英，于海勋等译．现代控制工程(第三版)．电子工业出版社，2000
[58]L. A. Zadeh. Fuzzy sets. Information and Control. 1965, 8: 338-353.
[59]Alfons Kemper, Guido Moerkotte. Object-Oriented Database Management. Prentics-Hall Internatial Inc. 1994.
[60]D N Chorafas, H Steinmahn. Object-Oriented Databases. Prentics-Hall Internatial Inc. 1994.
[61]王东哲，娄臻亮，张永清．薄板拉深工艺中压边力闭环控制的研究现状．机械工程师，2000．7
[62]朱钒，吴晓晖，李庆亮 等．高速智能型液压拉深机计算机控制系统的设计．机械开发，1999(4)．
[63]J.F. Allen. Maintaining knowledge about temporal intervals. CACM. 1993, 26:11.
[64]J.Han, Y.Fu. Discovery of multiple-level association rules from large databases. In Proc. Of the 21th Int. Conf. Of VLDB. 1995.
[65]湛从昌，蔡倩．液压元件及系统计算机辅助监测与故障诊断．机床与液压．1997，6．
[66]康伟 郑正奇．Windows下实时数据采集的实现．计算机应用研究，2001(3)．
[67]谢为民，高晖，邵大文．液压机监控系统的研究．机床与液压．1996，2：8-9．
[68]Dong Lv. Physical Simulation Based Intelligent System For The Prediction of Sheet Metal Drawing Capability. ACTA METALLUR GICA SINICA, Vol. 13, No.2.1999.
[69]L A Dobrzanski, W Sitek. Application of a neural network in modeling of hardenability of constructional steels. J. Mater. Proc.Tech. 1998, 78: 59～66.
[70]Osakada K, Yang G B. Neural networks for process planning of cold forging[J]. Annals of the CIRP, 1991, 40(1): 243～246.
[71]Lightbody G. Direct Neuron Model Reference Adaptive Control. IEE Pro-Control Theory Appl, 1995, 142(1):68～73
[72]何述东等．多层前向神经网络结构的研究进展．控制理论与应用．Vol．15，No．3，1998．
[2]范宏才，侍慕超，张昶盛．现代锻压机械．机械工业出版社．1994．
[3]朱钒，张志，张道富等．国内外液压机技术现状与发展趋势．机床与液压，2000 No．1
[4]李硕本．冲压工艺学．北京：机械工业出版社，1982
[5]Yossifon S, et al. On the Acceptable Blankholder force Range in the deep Drawing process, J. Mater. process. Techn. 1992, 33
[6]何丹农，陶宏之等．研究拉深成形过程中压边力值的合理确定．金属成形工艺，1999 No．5
[7]马力，张志等．极限拉深原理与极限拉深液压机研究．机械开发，2000 年增刊
[8]林建亚，何存兴．液压元件．机械工业出版社．1988．
[9]刘长年．液压伺服系统的分析与设计．科学出版社，1985．
[10]路甬祥，胡大绂．电液比例控制技术．第1版．机械工业出版社，1988．
[11]黎启柏．电液比例控制与数字控制系统．机械工业出版社，1997．
[12]王春行．液压伺服控制系统．机械工业出版社，1996．
[13]葛宜远．流体传动及控制技术的新成就．液压与气动，1998 No．2
[14]杨勇，黄文梅．一种新的电液伺服系统神经网络自适应控制，湖南大学学报，1998 No．10
[15]李运华，王占林．神经网络控制及其在液压伺服系统中的应用．机床与液压，1996 No．3
[16]熊宇飞，陆元章．多变量电液位置伺服系统的时序建模．机床与液压，1987 No．1
[17]Fu K.S. Learning Control Systems and Intelligent Control System. An Intersection of AI/AC IEEE Trans. On AC 1971.16:70-72
[18]Saridis, G.N. Towards The realization of intelligent Control. Proc.of IEEE. 1979. (67)8:4-11
[19]Saridis,G.N. Knowledge,Implementation,Structure of Intelligent Control System. Proc.of IEEE Int. Sump. On Intelligent Control. 1987.
[20]李宁，张乃尧．采用不同的模糊推理方法时典型模糊控制器的结构分析．模糊系统与数学，1998 Vol．12 No，3
[21]刘永红．神经网络理论的发展与前沿问题．信息与控制，1998 Vol．28 No．1
[22]Hebb D O. The Organization of Behavior, New York; Wiley, 1949
[23]Rosenblatt F. The Perceptron: A Probabilistic Model for Information storage and Organization in the Brain. Psychological Review, 1958, (65): 386～408
[24]Holland J H. Genetic Algorithms and the Optimal Allocations of Trials, SIAM Journal of Computing, 1973, (2): 88～105
[25]Holland J H. Adaptation in Natural and Artificial Systems, Ann Arbor: The University of Michigan Press, 1975
[26]Hopfield J J. Neural Networks and Physical Systems with Emergent Collective Computational Abilities, Proc. Natl. Acad. Sci., USA, 1982, (79): 2254～2558
[27]张立明．人工神经网络的模型及其应用．上海：复旦大学出版社，1992．
[28]袁南儿，王万良，苏宏业．计算机新型控制策略及其应用．北京：清华大学出版社，1998．152～153
[29]毛宗源，姚尹武，胡尚信．变结构神经网络．控制理论与应用，1999 Vol．16 No．1
[30]康景峰，张凤传等．QCS003型液压试验台微机控制及数据采集系统．实验技术与管理，1996 Vol．13 No．1
[31]陈众，方璐，李楠．VC环境下小型工业监控软件的开发．计算机自动测量与控制，2000．8(5)
[32]张乃镇，汤峰，刘建业．采用Visual C++开发的导弹伺服机构测控系统．军用测控技术．2000．8(3)
[33]蔡自兴．智能控制—基础与应用．北京：国防工业出版社．1998
[34]吕冬，丁柯，何丹农等．基于神经网络的拉深力智能化预测系统．中国有色金属学报，2000，Vol．10，No．3
[35]罗亚军，张永清，赵军．板材拉深成形智能化控制过程中摩擦系数的识别．摩擦与润滑，2001，No．2
[36]R C Eberhart R W Dobbbins. Neural Network. Pc tools, Academic Press, 1990.
[37]张立明．人工神经网络的模型及其应用．上海：复旦大学出版社．1994．
[38]Myllykosky Pand Larkiola J. Development of prediction model for mechanical properties of batch annealed thin steel strip by using artificial neural network modeling [J]. J Mater Proc Technol, 1996, 60: 399-404.
[39]卢险峰．冲压工艺模具学．第1版．机械工业出版社，1998．
[40]中川威雄 阿部帮雄 林丰．板料冲压加工．天津科学技术出版社，1982．
[41]姜海峰等．变压边力拉深的原理及实验系统．塑性工程学报．1999(3)，Vol．6，NO．1：31
[42]梁炳文等．弹塑性稳定理论．国防工业出版社，1983
[43]汪大年主编．金属塑性成性原理．机械工业出版社，1982
[44]M．B．斯德洛日夫[俄]．金属压力加工原理．机械工业出版社．1980
[45]中川威雄等．冲压加工技术手册[日]．轻工业出版社，1988．
[46]π．A．肖夫曼[俄]．冷压原理．机械工业出版社，1957．
[47]姚文季．金属板料的压延性能研究．南京航空学院学报，1964，2．
[48]熊志卿，熊焰．拉深极限研究．塑性工程学报．1997．6，Vol．4，No．2
[49]邓明，彭成允，李渝．拉延速度对拉延件破裂的影响．锻压机械，2001(4)．
[50]刘莉莉．拉深系数与冲压力关系的分析．河南职技师院学报，2000(6)，Vol．28，No．2
[51]丁松聚．冷冲模设计[M]．北京：机械工业出版社，1997．5，141-161．
[52]吕雪山，王先进．薄板成形与制造[M]．北京：中国物资出版社．1993．
[53]王东哲，吕东等．板料成形中压边力的控制理论研究．锻压技术，2000年4期
[54]卢险峰．冲压工艺模具学．第1版．机械工业出版社，1998．
[55]刘天骥编译．冲压件缺陷及消除方法．重庆大学出版社，1988．
[56]翁克索夫，约翰逊，工藤英明．金属塑性变形理论．机械工业出版社，1992．
[57]Katsuhiko Ogata著．卢伯英，于海勋等译．现代控制工程(第三版)．电子工业出版社，2000
[58]L. A. Zadeh. Fuzzy sets. Information and Control. 1965, 8: 338-353.
[59]Alfons Kemper, Guido Moerkotte. Object-Oriented Database Management. Prentics-Hall Internatial Inc. 1994.
[60]D N Chorafas, H Steinmahn. Object-Oriented Databases. Prentics-Hall Internatial Inc. 1994.
[61]王东哲，娄臻亮，张永清．薄板拉深工艺中压边力闭环控制的研究现状．机械工程师，2000．7
[62]朱钒，吴晓晖，李庆亮 等．高速智能型液压拉深机计算机控制系统的设计．机械开发，1999(4)．
[63]J.F. Allen. Maintaining knowledge about temporal intervals. CACM. 1993, 26:11.
[64]J.Han, Y.Fu. Discovery of multiple-level association rules from large databases. In Proc. Of the 21th Int. Conf. Of VLDB. 1995.
[65]湛从昌，蔡倩．液压元件及系统计算机辅助监测与故障诊断．机床与液压．1997，6．
[66]康伟 郑正奇．Windows下实时数据采集的实现．计算机应用研究，2001(3)．
[67]谢为民，高晖，邵大文．液压机监控系统的研究．机床与液压．1996，2：8-9．
[68]Dong Lv. Physical Simulation Based Intelligent System For The Prediction of Sheet Metal Drawing Capability. ACTA METALLUR GICA SINICA, Vol. 13, No.2.1999.
[69]L A Dobrzanski, W Sitek. Application of a neural network in modeling of hardenability of constructional steels. J. Mater. Proc.Tech. 1998, 78: 59～66.
[70]Osakada K, Yang G B. Neural networks for process planning of cold forging[J]. Annals of the CIRP, 1991, 40(1): 243～246.
[71]Lightbody G. Direct Neuron Model Reference Adaptive Control. IEE Pro-Control Theory Appl, 1995, 142(1):68～73
[72]何述东等．多层前向神经网络结构的研究进展．控制理论与应用．Vol．15，No．3，1998．