基于改进协作优化算法的塔式起重机金属结构一体化设计研究
|
摘要
当今机械产品结构日益复杂，且产品竞争日趋激烈，对于机械产品的性能要求越来越高，因而产品的优化设计也不再局限于零件和部件的优化，而是从更高的层次上提出了整机/系统的整体性能优化设计，而传统的设计优化方法求解这类优化问题存在着诸多的缺陷。在这样的工程背景下，一种新的设计优化方法——多学科设计优化诞生了。
多学科设计优化（Multidisciplinary Design Optimization，简称MDO）是一种通过充分探索和利用系统中相互作用的协同机制来设计复杂系统和子系统的方法论。其目的是利用系统中的协同机制获得系统整体最优解，通过并行设计来缩短设计周期。本文的目的就是研究多学科设计优化算法及其在塔式起重机金属结构设计中的应用。
由于系统内部各个组成学科（子系统）之间存在耦合作用，多学科设计优化存在着系统分析的计算复杂性和各个学科（子系统）之间的通信复杂性。MDO算法就是解决这两个难题的策略和方法。本文对现有的各种MDO算法进行分析、比较，指出基于分解和协调的双层优化算法是MDO算法的发展方向。
协作优化算法（Collaborative Optimizaiton，简称CO）是在一致性约束优化算法（Compatibility Constrained Optimization，简称CCO）的基础上发展起来的，基于分解和协调的一种双层优化算法。子系统优化时暂时不考虑其它子系统的影响，故各个子系统可以相对独立的进行并行优化，设计冲突由系统级优化问题协调。针对标准CO系统级优化问题中K－T条件不满足而引起的计算困难，本文提出了通过采用权因子，将系统级问题无约束化处理的改进方法。该方法在避免上述计算困难的同时，简化了系统级优化问题。
本文的重点是改进协作优化算法在工程设计中的应用研究。传统的塔式起重机的优化设计是对部件（子系统）分别进行的，这样的作法割离了子系统之间的有机的联系，无法保证整机性能的最优性。通过采用改进的协作优化算法对塔式起重机金属结构的塔帽－吊臂－平衡臂进行一体化方案设计，充分考虑了这三个子系统之间的协同机制，得到了具有整体最优性的设计结果。
采用Matlab的图形用户界面（GUI）对塔式起重机的一体化设计进行了系统集成，开发了基于改进协作优化算法的塔式起重机金属结构一体化设计系统（COBTC）。采用该系统对某型号的塔机进行了优化设计，得到了理想的结果。
在论文的最后对全文所作的工作进行总结，并进一步展望了多学科设计优化方法在复杂机械系统设计中的应用前景
At present, mechanic product becomes increasingly complex, and competitions among producers become more fury and rigorous, while the demands for products' performance from costume get much higher. So a product optimization design should not be limited within single part or component. Optimization design should starts from a much higher level, namely, from the view of the whole performance of a whole machine or an integrated system. However, facing such demanding, the traditional methods seem to be obsolete because of many inherent limitations. Under this situation, a new optimization design strategy, Multidisciplinary Design Optimization (MDO), emerged
Multidisciplinary Design Optimization is a methodology for the design of complex engineering systems and subsystems that coherently exploits the synergism of mutually interacting phenomena. The MDO provides a method that can improve product quality using the synergism in the engineering systems and reduce development time through the concurrency of system design. The motivation for this research is the investigation of current MDO algorithms and application of the most promising MDO algorithm to the metal structure preliminary design of tower crane.
Due to the interdisciplinary coupling inherent in the components subsystem of a complex system, two obstacles appear in MDO implementation, they are computational burden for system analysis, the complexity of communications among various subsystems. MDO algorithms address the two issues. Firstly the research presents a survey of basic MDO algorithm and the comparisons among them. And then points out that decomposition and coordination based bi-level algorithm will be the direction of future MDO algorithm.
Collaborative Optimization (CO), originated from the Compatibility Constrained Optimizaiton (CCO), is a bi-level optimization algorithm based on system decomposition and coordination. In CO, when come to a certain subsystem optimization, the influences from other subsystems are ignored temporarily, while a set of local variables take their places in the subsystem design. As for the design conflicts among subsystems, coordination are performed through the solution of the system optimization. In application, the formally formulated CO encounters computational difficulties in most cases, namely in the system-level optimization, the Kuhn-Tucker condition can not be satisfied, in this research, a revised CO was presented, which de-constraints the
system-level optimization through a set of proper weight factors while keep the subsystems' formulations unchanged. This new algorithm can sidestep the above mentioned computational difficulties and simplify the optimization problem.
The emphasis of the current research is the study of the application of the revised CO in industrial case. In traditional optimization design of a tower crane, the components of a tower crane is singled out as design objects respectively and the inherent coupling links among the subsystems are ignored completely, which can't result in a optimization design in the system view of point. An integrated preliminary design of a tower crane metal structure, including head, jib and counter-jib, was carried out based on the revised CO, taking fully consideration of the inherent coupling mechanisms among the three components. In this way, system optimal design was reached.
At the same time, an integrated design system, revised Collaborative Optimization Based Tower Crane design optimization system (COBTC), was developed based on the Matlab Graphic User Interface (GUI). A design case was presented to verify the design system and satisfying results reached.
In the end, a summary of the whole research work was given out, together with the outlook on the applications of MDO to the complex mechanical system design
多学科设计优化（Multidisciplinary Design Optimization，简称MDO）是一种通过充分探索和利用系统中相互作用的协同机制来设计复杂系统和子系统的方法论。其目的是利用系统中的协同机制获得系统整体最优解，通过并行设计来缩短设计周期。本文的目的就是研究多学科设计优化算法及其在塔式起重机金属结构设计中的应用。
由于系统内部各个组成学科（子系统）之间存在耦合作用，多学科设计优化存在着系统分析的计算复杂性和各个学科（子系统）之间的通信复杂性。MDO算法就是解决这两个难题的策略和方法。本文对现有的各种MDO算法进行分析、比较，指出基于分解和协调的双层优化算法是MDO算法的发展方向。
协作优化算法（Collaborative Optimizaiton，简称CO）是在一致性约束优化算法（Compatibility Constrained Optimization，简称CCO）的基础上发展起来的，基于分解和协调的一种双层优化算法。子系统优化时暂时不考虑其它子系统的影响，故各个子系统可以相对独立的进行并行优化，设计冲突由系统级优化问题协调。针对标准CO系统级优化问题中K－T条件不满足而引起的计算困难，本文提出了通过采用权因子，将系统级问题无约束化处理的改进方法。该方法在避免上述计算困难的同时，简化了系统级优化问题。
本文的重点是改进协作优化算法在工程设计中的应用研究。传统的塔式起重机的优化设计是对部件（子系统）分别进行的，这样的作法割离了子系统之间的有机的联系，无法保证整机性能的最优性。通过采用改进的协作优化算法对塔式起重机金属结构的塔帽－吊臂－平衡臂进行一体化方案设计，充分考虑了这三个子系统之间的协同机制，得到了具有整体最优性的设计结果。
采用Matlab的图形用户界面（GUI）对塔式起重机的一体化设计进行了系统集成，开发了基于改进协作优化算法的塔式起重机金属结构一体化设计系统（COBTC）。采用该系统对某型号的塔机进行了优化设计，得到了理想的结果。
在论文的最后对全文所作的工作进行总结，并进一步展望了多学科设计优化方法在复杂机械系统设计中的应用前景
At present, mechanic product becomes increasingly complex, and competitions among producers become more fury and rigorous, while the demands for products' performance from costume get much higher. So a product optimization design should not be limited within single part or component. Optimization design should starts from a much higher level, namely, from the view of the whole performance of a whole machine or an integrated system. However, facing such demanding, the traditional methods seem to be obsolete because of many inherent limitations. Under this situation, a new optimization design strategy, Multidisciplinary Design Optimization (MDO), emerged
Multidisciplinary Design Optimization is a methodology for the design of complex engineering systems and subsystems that coherently exploits the synergism of mutually interacting phenomena. The MDO provides a method that can improve product quality using the synergism in the engineering systems and reduce development time through the concurrency of system design. The motivation for this research is the investigation of current MDO algorithms and application of the most promising MDO algorithm to the metal structure preliminary design of tower crane.
Due to the interdisciplinary coupling inherent in the components subsystem of a complex system, two obstacles appear in MDO implementation, they are computational burden for system analysis, the complexity of communications among various subsystems. MDO algorithms address the two issues. Firstly the research presents a survey of basic MDO algorithm and the comparisons among them. And then points out that decomposition and coordination based bi-level algorithm will be the direction of future MDO algorithm.
Collaborative Optimization (CO), originated from the Compatibility Constrained Optimizaiton (CCO), is a bi-level optimization algorithm based on system decomposition and coordination. In CO, when come to a certain subsystem optimization, the influences from other subsystems are ignored temporarily, while a set of local variables take their places in the subsystem design. As for the design conflicts among subsystems, coordination are performed through the solution of the system optimization. In application, the formally formulated CO encounters computational difficulties in most cases, namely in the system-level optimization, the Kuhn-Tucker condition can not be satisfied, in this research, a revised CO was presented, which de-constraints the
system-level optimization through a set of proper weight factors while keep the subsystems' formulations unchanged. This new algorithm can sidestep the above mentioned computational difficulties and simplify the optimization problem.
The emphasis of the current research is the study of the application of the revised CO in industrial case. In traditional optimization design of a tower crane, the components of a tower crane is singled out as design objects respectively and the inherent coupling links among the subsystems are ignored completely, which can't result in a optimization design in the system view of point. An integrated preliminary design of a tower crane metal structure, including head, jib and counter-jib, was carried out based on the revised CO, taking fully consideration of the inherent coupling mechanisms among the three components. In this way, system optimal design was reached.
At the same time, an integrated design system, revised Collaborative Optimization Based Tower Crane design optimization system (COBTC), was developed based on the Matlab Graphic User Interface (GUI). A design case was presented to verify the design system and satisfying results reached.
In the end, a summary of the whole research work was given out, together with the outlook on the applications of MDO to the complex mechanical system design
引文
[1]冯培恩等.机械广义优化设计的理论框架.中国机械工程.2000.第11期：126-128
[2]董仲元等.近年来国际设计方法学研究的发展.机械设计.1993.No.6：3-6
[3]程耿东等.我国机械优化研究与应用的综述和展望.机械强度.1995.第17卷.N0.2：68-74
[4]李守林.我国塔式起重机的现状与发展.建筑机械化.2000（6）：9-11
[5]罗文龙.发展具有中国自主产权的中小型塔机.建筑机械化.2002（3）：7-9
[6]Kevin.F.Hulme，The Design of A Simulation-Based Framework for the Development of Solution Approaches in Multidisciplinary Design Optimization.[Ph.D. Dissertation].State University of New York at Buffalo.2000
[7]余雄庆.多学科设计优化算法及其在飞机设计中的应用研究.南京.南京航空航天大学.博士学位论文，1999
[8]熊光楞主编.并行工程的理论和实践.北京.清华大学出版社.2000.5.
[9]Yu X Q, Stelmack M A, Batill S M. An Application of the Concurrent Subspace Design (CSD) to the preliminary design of a low-reynolds number UAV [R].AIAA 98-4917, Missouri, AIAA, 1998
[10]蔡显新.整体叶轮形状优化、保形一体化设计研究.南京.南京航空航天大学.博士学位论文，1999
[11]李伟剑.基于MEMS技术的微型飞机的多学科设计优化.西安.西北工业大学，硕士学位论文.2001
[12]王健.导弹总体多学科设计优化技术.战术导弹技术.2000（4）：1-7
[13]陈琪锋，李晓斌，戴金海.导弹总体参数优化设计的合作协同进化MDO算法.国防科技大学学报.2001.Vol.23 No.5：9-11
[14] Li Weiji，Hu Yu. A New Effective Multidisciplinary Design Optimization Algorithm. ICAS 2002 Congress, Canada, 2002
[15]李响，李为吉.基于序列响应面方法的协同优化算法.西北工业大学学报.2003.Vol.21 No.1：79-82
[16]余雄庆.基于Agent模型的多学科设计优化方法.机械科学与技术.2002. Vol.21 No.5：844-851
[17]郭健.多学科设计优化技术研究. 西安.西北工业大学，硕士学位论文.2001
[18]余雄庆，丁运亮.多学科设计优化算法及其在飞行器设计中应用.航空学报.2000. Vol.21 No.1：1-6
[19]Sobieszanski-Sobieski J, Haftka T. Multidisciplinary aerospace design
optimizaiton: Survey of Resent Development[R].AIAA 96-0711.Nevada:AIAA.1996
[20]Sobieszanski-Sobieski J. A step from hierarchic to nonhierarchic systems[R].NASA-CP-3031.Part I, Virginia: NASA, 1989
[21] Kroo I, Altus S, Braun R, et al. Multidisciplinary optimization methods for aircraft preliminary design[R].AIAA-94-4325,FL:AIAA,1994.
[22]Kodiyalam S, Sobieszczanski-Sobieski J. Multidisciplinary design optimization-some formal methods, framework requirements, and application to vehicle design, Int.J. Vehicle Design(Special Issue).p3-22
[23]Ilan Kroo, Valerie Manning. Collaborative Optimization: Status and Directions.AIAA-2000-4721, 8th AIAA/NASA/ISSMO Symposium on Multidisciplinary Analysis and Optimization, Long Bench, CA
[24] Robert Braun, Peter Gage, Ilan Kroo et al. Implementation and Performance Issues in Collaborative Optimization. AIAA Paper. AIAA-96-4017.
[25]Natalia M. Alexandrov, Robbert Michael Lewis. Analytical and Computational Properties of Distributed Approaches to MDO.AIAA Paper. AIAA-2000-4718.
[26]Natalia M. Alexandrov, Robbert Michael Lewis. Analytical and Computational Properties Aspects of Collaborative Optimization. AIAA Technical Report. April. 2001
[27]张国瑞，丁怡等编.塔式起重机.北京，机械工业出版社.1989.3
[28]胡宗武，顾迪民编著.起重机设计计算.北京.北京科学出版社.1989.8
[29]陆念力，兰朋，李以申.塔式起重机双吊点水平臂内力计算的精确通用公式.建筑机械。1997（4）：3：5
[30]马鹏飞.塔式起重机水平吊臂双吊点位置优化.工程机械.1998（5）：15-17
[31]郭岗，李桂芳.塔机平衡臂吊点位置的确定及程序设计.建筑机械技术与管理.1999（5）：14-15
[32]王积永，宋世军等.塔式起重机平衡重的确定及起升特性的调整.工程机械.2000（2）22-23
[33]中华人民共和国国家标准《塔式起重机设计规范》GB/T13752-92,中国标准出版社.1993
[34]Mark R. Rawlings, Richard J. Balling. Collaborative Optimization with Disciplinary Conceptual Design. AIAA Paper. AIAA-98-4919
[35]J. Bennett, P. Fenyes, W. Haering et al. Issues in Industrial Multidisciplinary Optimizaiton. AIAA Paper. AIAA-98-4727.
[36]龚剑，朱亮编著.Matlab 5.x 入门与提高.清华大学出版社.2000.5
[37]施晓红，周佳编著.精通GUI图形界面编程.北京大学出版社.2003.1
[38]何强，何英编著.Matlab扩展编程.清华大学出版社.2002.5
重庆大学硕士学位论文 1 绪论
[2]董仲元等.近年来国际设计方法学研究的发展.机械设计.1993.No.6：3-6
[3]程耿东等.我国机械优化研究与应用的综述和展望.机械强度.1995.第17卷.N0.2：68-74
[4]李守林.我国塔式起重机的现状与发展.建筑机械化.2000（6）：9-11
[5]罗文龙.发展具有中国自主产权的中小型塔机.建筑机械化.2002（3）：7-9
[6]Kevin.F.Hulme，The Design of A Simulation-Based Framework for the Development of Solution Approaches in Multidisciplinary Design Optimization.[Ph.D. Dissertation].State University of New York at Buffalo.2000
[7]余雄庆.多学科设计优化算法及其在飞机设计中的应用研究.南京.南京航空航天大学.博士学位论文，1999
[8]熊光楞主编.并行工程的理论和实践.北京.清华大学出版社.2000.5.
[9]Yu X Q, Stelmack M A, Batill S M. An Application of the Concurrent Subspace Design (CSD) to the preliminary design of a low-reynolds number UAV [R].AIAA 98-4917, Missouri, AIAA, 1998
[10]蔡显新.整体叶轮形状优化、保形一体化设计研究.南京.南京航空航天大学.博士学位论文，1999
[11]李伟剑.基于MEMS技术的微型飞机的多学科设计优化.西安.西北工业大学，硕士学位论文.2001
[12]王健.导弹总体多学科设计优化技术.战术导弹技术.2000（4）：1-7
[13]陈琪锋，李晓斌，戴金海.导弹总体参数优化设计的合作协同进化MDO算法.国防科技大学学报.2001.Vol.23 No.5：9-11
[14] Li Weiji，Hu Yu. A New Effective Multidisciplinary Design Optimization Algorithm. ICAS 2002 Congress, Canada, 2002
[15]李响，李为吉.基于序列响应面方法的协同优化算法.西北工业大学学报.2003.Vol.21 No.1：79-82
[16]余雄庆.基于Agent模型的多学科设计优化方法.机械科学与技术.2002. Vol.21 No.5：844-851
[17]郭健.多学科设计优化技术研究. 西安.西北工业大学，硕士学位论文.2001
[18]余雄庆，丁运亮.多学科设计优化算法及其在飞行器设计中应用.航空学报.2000. Vol.21 No.1：1-6
[19]Sobieszanski-Sobieski J, Haftka T. Multidisciplinary aerospace design
optimizaiton: Survey of Resent Development[R].AIAA 96-0711.Nevada:AIAA.1996
[20]Sobieszanski-Sobieski J. A step from hierarchic to nonhierarchic systems[R].NASA-CP-3031.Part I, Virginia: NASA, 1989
[21] Kroo I, Altus S, Braun R, et al. Multidisciplinary optimization methods for aircraft preliminary design[R].AIAA-94-4325,FL:AIAA,1994.
[22]Kodiyalam S, Sobieszczanski-Sobieski J. Multidisciplinary design optimization-some formal methods, framework requirements, and application to vehicle design, Int.J. Vehicle Design(Special Issue).p3-22
[23]Ilan Kroo, Valerie Manning. Collaborative Optimization: Status and Directions.AIAA-2000-4721, 8th AIAA/NASA/ISSMO Symposium on Multidisciplinary Analysis and Optimization, Long Bench, CA
[24] Robert Braun, Peter Gage, Ilan Kroo et al. Implementation and Performance Issues in Collaborative Optimization. AIAA Paper. AIAA-96-4017.
[25]Natalia M. Alexandrov, Robbert Michael Lewis. Analytical and Computational Properties of Distributed Approaches to MDO.AIAA Paper. AIAA-2000-4718.
[26]Natalia M. Alexandrov, Robbert Michael Lewis. Analytical and Computational Properties Aspects of Collaborative Optimization. AIAA Technical Report. April. 2001
[27]张国瑞，丁怡等编.塔式起重机.北京，机械工业出版社.1989.3
[28]胡宗武，顾迪民编著.起重机设计计算.北京.北京科学出版社.1989.8
[29]陆念力，兰朋，李以申.塔式起重机双吊点水平臂内力计算的精确通用公式.建筑机械。1997（4）：3：5
[30]马鹏飞.塔式起重机水平吊臂双吊点位置优化.工程机械.1998（5）：15-17
[31]郭岗，李桂芳.塔机平衡臂吊点位置的确定及程序设计.建筑机械技术与管理.1999（5）：14-15
[32]王积永，宋世军等.塔式起重机平衡重的确定及起升特性的调整.工程机械.2000（2）22-23
[33]中华人民共和国国家标准《塔式起重机设计规范》GB/T13752-92,中国标准出版社.1993
[34]Mark R. Rawlings, Richard J. Balling. Collaborative Optimization with Disciplinary Conceptual Design. AIAA Paper. AIAA-98-4919
[35]J. Bennett, P. Fenyes, W. Haering et al. Issues in Industrial Multidisciplinary Optimizaiton. AIAA Paper. AIAA-98-4727.
[36]龚剑，朱亮编著.Matlab 5.x 入门与提高.清华大学出版社.2000.5
[37]施晓红，周佳编著.精通GUI图形界面编程.北京大学出版社.2003.1
[38]何强，何英编著.Matlab扩展编程.清华大学出版社.2002.5
重庆大学硕士学位论文 1 绪论