微遗传优化与iSIGHT集成技术研究
|
摘要
本文针对工程机械优化设计,在国家基金(NO.50575039)的支持下,研究一类智能优化设计方法(改进微遗传算法),并将其集成到iSIGHT优化设计平台。
工程机械的种类繁多,而且工程机械大多是在野外作业,因此对工程机械的质量、寿命和可靠性等的使用要求,以及生产成本、适应范围和维护费用等经济性要求也越来越高。工程机械的优化设计已经成为我国工程机械设计领域的发展趋势。
当前工程机械优化设计问题具有设计变量多、设计目标函数复杂和约束条件多为非线性的特点。传统基于梯度的优化方法求解上述问题存在困难,使得智能类优化设计方法(比如遗传方法)得到应用。iSIGHT软件可以集成仿真代码并提供优化设计方案,为工程机械优化提供了技术支持。
本课题为克服常规遗传算法计算量大、收敛速度慢等缺点,采用一种改进的微遗传算法(简称μGA-BLX):在交叉操作上采用了BLX-α算子,增强了算法的搜索能力；在变异操作上引入柯西变异算子,增强了种群多样性。并且在算法重新初始化阶段引入算法个体池,减少重新生成个体的盲目性。并将μGA-BLX集成到iSIGHT优化算法中,试图解决工程机械优化问题。本文主要研究工作如下
(1)针对遗传优化的工作量大且收敛速度慢的缺点,经实验选择遗传算子,对微遗传算法进行改进,并通过行星传动变速机构的参数优化和链传动优化验证μGA-BLX的可行性。
(2)针对门式起重机,给出基于iSIGHT的设计流程；通过阐述iSIGHT的Tcl语言结构,研究iSIGHT的二次开发技术,并给出μGA-BLX算法集成的关键技术。
(3)基于上述集成的μGA-BLX算法。在iSIGHT环境下对挖掘机工作装置和装载机变速箱进行优化设计,通过它们优化实验结果来验证本文方法的可行性。
In this paper, with the support of National Fund (NO.50575039), an intelligent optimization mehod (ie, an improved Miacro-Genetic Algorithm) was researched for Engineering Machinical Optimization Design, and was integerated into Optimal Design of iSIGHT.
With many types of Engineering Machine, they mostly operations in field, so the quality of Engineering Machine, lifetime and reliability of servie requirements, production costs, the range of adaption and maintance costs of economiac demands have become more sophisticated. Engineering Machinical Optimization Design has been the trend of the field of Engineering Machine in China.
Most of Engineering Mechaninical optimizations are problems with the characteristics of multi-variables, complex objectives and non-linear constraints. It is so difficult to solve these problems using Traditional Optimization Based on Gradient that Intelligent Algorithms are widely applied, e.g. Genetic Algorithm. ISIGHT can integrate simulational code, provide optimizational algorithms, and provide technical support for Engineering Mechanical Optimization.
In this paper, presenting a improve Micro-Genetic Algorithm (μGA-BLX) to solve mechanical optimizations. TheμGA-BLX overcomes the disadvantages of Standard Genetic Algorithm (SGA), such as high computation, and slow convergence and so on. TheμGA-BLX with the Blend Crossover (BLX-a) and Cauchy Mutation enhanced the search and explore ability of GA. In addition, using the population pool in the restart phase reduced the blindness of regenerate population. TheμGA-BLX was integrated into optimization algorithm of iSIGHT, and was used to solove Engineering Mechaninical optimization. The main contents of this paper are as follows:
(1) To overcome the disadvantages of Genetic Optimization, such as high computation, slow convergency and so on, an improved Micro-Gentic Algorithm was presented with the experimental selection of genetic operators. The results of the Planetary Gear Transmission optimization and Chain optimization showed the feasibility of theμGA-BLX.
(2) ISIGHT Optimization Design Platform for Gantry Crane is studied, introduced process of iSIGHT Design, and described the TCL structure of iSIGHT. The second development of iSIGHT and the key technologies of Algorithm Integration are studied.
(3) IntegratedμGA-BLX based on iSIGHT. UsingμGA-BLX to optimize excavator and loader gearbox in the iSIGHT, and analyzed the results to verify feasibility of the proposed algorithm.
工程机械的种类繁多,而且工程机械大多是在野外作业,因此对工程机械的质量、寿命和可靠性等的使用要求,以及生产成本、适应范围和维护费用等经济性要求也越来越高。工程机械的优化设计已经成为我国工程机械设计领域的发展趋势。
当前工程机械优化设计问题具有设计变量多、设计目标函数复杂和约束条件多为非线性的特点。传统基于梯度的优化方法求解上述问题存在困难,使得智能类优化设计方法(比如遗传方法)得到应用。iSIGHT软件可以集成仿真代码并提供优化设计方案,为工程机械优化提供了技术支持。
本课题为克服常规遗传算法计算量大、收敛速度慢等缺点,采用一种改进的微遗传算法(简称μGA-BLX):在交叉操作上采用了BLX-α算子,增强了算法的搜索能力；在变异操作上引入柯西变异算子,增强了种群多样性。并且在算法重新初始化阶段引入算法个体池,减少重新生成个体的盲目性。并将μGA-BLX集成到iSIGHT优化算法中,试图解决工程机械优化问题。本文主要研究工作如下
(1)针对遗传优化的工作量大且收敛速度慢的缺点,经实验选择遗传算子,对微遗传算法进行改进,并通过行星传动变速机构的参数优化和链传动优化验证μGA-BLX的可行性。
(2)针对门式起重机,给出基于iSIGHT的设计流程；通过阐述iSIGHT的Tcl语言结构,研究iSIGHT的二次开发技术,并给出μGA-BLX算法集成的关键技术。
(3)基于上述集成的μGA-BLX算法。在iSIGHT环境下对挖掘机工作装置和装载机变速箱进行优化设计,通过它们优化实验结果来验证本文方法的可行性。
In this paper, with the support of National Fund (NO.50575039), an intelligent optimization mehod (ie, an improved Miacro-Genetic Algorithm) was researched for Engineering Machinical Optimization Design, and was integerated into Optimal Design of iSIGHT.
With many types of Engineering Machine, they mostly operations in field, so the quality of Engineering Machine, lifetime and reliability of servie requirements, production costs, the range of adaption and maintance costs of economiac demands have become more sophisticated. Engineering Machinical Optimization Design has been the trend of the field of Engineering Machine in China.
Most of Engineering Mechaninical optimizations are problems with the characteristics of multi-variables, complex objectives and non-linear constraints. It is so difficult to solve these problems using Traditional Optimization Based on Gradient that Intelligent Algorithms are widely applied, e.g. Genetic Algorithm. ISIGHT can integrate simulational code, provide optimizational algorithms, and provide technical support for Engineering Mechanical Optimization.
In this paper, presenting a improve Micro-Genetic Algorithm (μGA-BLX) to solve mechanical optimizations. TheμGA-BLX overcomes the disadvantages of Standard Genetic Algorithm (SGA), such as high computation, and slow convergence and so on. TheμGA-BLX with the Blend Crossover (BLX-a) and Cauchy Mutation enhanced the search and explore ability of GA. In addition, using the population pool in the restart phase reduced the blindness of regenerate population. TheμGA-BLX was integrated into optimization algorithm of iSIGHT, and was used to solove Engineering Mechaninical optimization. The main contents of this paper are as follows:
(1) To overcome the disadvantages of Genetic Optimization, such as high computation, slow convergency and so on, an improved Micro-Gentic Algorithm was presented with the experimental selection of genetic operators. The results of the Planetary Gear Transmission optimization and Chain optimization showed the feasibility of theμGA-BLX.
(2) ISIGHT Optimization Design Platform for Gantry Crane is studied, introduced process of iSIGHT Design, and described the TCL structure of iSIGHT. The second development of iSIGHT and the key technologies of Algorithm Integration are studied.
(3) IntegratedμGA-BLX based on iSIGHT. UsingμGA-BLX to optimize excavator and loader gearbox in the iSIGHT, and analyzed the results to verify feasibility of the proposed algorithm.
引文
[1]陈法军等.机械优化设计遗传算法[M].北京：机械工业出版社,2005.
[2]董立立等.机械优化设计理论方法研究综述[J].机床与液压设计,2010,38(15)：115-1]9.
[3]梁尚明,殷国富.现代机械优化设计方法[M].北京：化学工业出版社,2005.
[4]雷德明,严新平.多目标智能优化算法及其应用[M].北京：科学出版社,2009.
[5]王凌.智能优化算法及其应用[M].北京：清华大学出版社,2000.
[6]吴军.混和进化算法及其工程应用研究[D].西安：西北工业大学出版社,2007.
[7]冯忠绪等.工程机械理论[M].北京：人民交通出版社,2004.
[8]章成器.工程机械优化设计引论[J].工程机械,1982.
[9]吴庆鸣,何小新等.工程机械设计[M].武汉：武汉大学出版社,2006.
[10]王世明,杨为民.国外工程机械新技术新结构和发展趋势[J].国外工程机械新技术新结构和发展趋势,2004,35(1)：61-66.
[11]黄综益,李兴华.工程机械技术发展展望[J].建筑机械化,2000,21(1)：4.
[12]管荣根.工程机械的绿色设计[J].工程机械,2001,35(10)：26-27.
[13]秦东晨,陈江义,胡滨生等.机械结构优化设计的综述与展望[J].中国科技信息,2005(9)：2.
[14]陈育仪.工程机械优化设计[M].北京：中国铁道出版社,1987.
[15]王科社.机械优化设计[M].北京：国防工业出版社,2006.
[16]张运节.机械优化设计综述与展望[J].科技信息,2009：2.
[17]JHHolland. Adaptation in naural and artificial systems[D]. Englewood Cliffs: University of Michigan]975.
[18]黄豪中.柴油均质压燃(HCCI)发动机燃烧过程数值模拟和实验研究[D].天津：天津大学,2007.
[19]PardalosPanosM, RomeijinHEdwin, TuyHoang. Recent developments and trends in global optimization[J]. Journal of Computational and Applied Mathemmatics,2000(124): 204-228.
[20]DeJongK. An analysis of the behavior of a class of genetic adaptive systems[D]. Michigan:University of Michigan,1975.
[21]KrishnakumarK. Micro-genetic algorithms for stationary and nonStationary function optimization[Z]. Philadelphia, PA, USA:1989.
[22]XuYG, LiuGR, WuZP. A novel hybrid genetic algorithm using local optimizer based on heuristic pattern move [J]. Applied Artificial Intelligence,2001,15(7): 601-613.
[23]LiuGR, HanX, OhyoshiT. Computational inverse techniques for material characterization using dynamic response [J].秋田大学学术情报,2002,10(1): 26-33.
[24]LiuGR, MaHJ. Material characterization of composite laminate using dynamic response and real parameter coded micro genetic algorithm[J]. Engineering with Computer,2003(20):295-308.
[25]LiuGR, MaWB, HanX. Inversion of loading time history using displacement response of composite laminates:three-dimensional cases [[J]. Acta Mechanical,.2002, 157(1-4):223-234.
[26]方海鹏,冯恩民,俞红梅.改进的遗传模拟退火算法及在换热网络综合中的应用.[J].应用基础与工程科学学报,,1998,6(1)：1-6.
[27]李明星.基于遗传算法的的门座起重机变幅系统优化设计[D].武汉：武汉理工大学,2006.
[28]唐飞,滕弘飞.一种改进的遗传算法及其在布局优化在的应用[J].软件学报,1999,10(10)： 7.
[29]朱延广,梅珊,赵雯,et al.支持复杂产品总体优化设计的多算法协作优化框架研究[J].系统仿真学,2007,19(11)：2417-2420.
[30]马静敏.基于iSIGHT的电子汽车衡的多学科设计优化[D].青岛：山东科技大学,2005.
[31]耿建光.工程系统过程集成和设计优化--iSIGHT中的过程集成与设计自动化[J].军民两用技术与产品,2002,″″(11).
[32]SLPadula, JJKorte, HJDunn, et al. Multidisciplinary Optimization BranchExperience Using iSIGHT Software[M]. Hampton Virginia:National Aeronautics and Space Administration Langley Research Center 1999.
[33]Chen J Z, Wu Y, Gence C, et al. Integrated electrical and thermal analysis of integrated power electronics modules using iSIGHT[C]. In:Applied Power Electronics Conference and Exposition,2001. APEC 2001. Sixteenth Annual IEEE,2001, 2:1002-1006 vol.2.
[34]Chen X, Yu X, Ji B:Study of Crankshaft Strength Based on iSIGHT Platform and DOE Methods[C]. In:Measuring Technology and Mechatronics Automation (ICMTMA),2010 International Conference on,2010,3:548-551.
[35]杨铭.基于iSIGHT的船型生成及耐波性优化研究[D].武汉：武汉理工大学,2010.
[36]程成,须文波,冷文浩.基于iSIGHT平台DOE方法的螺旋桨敞水性能优化设计[J].计算机工程与应用,2007,28(6)：5.
[37]KochPN, EvansJP, PowellD. Interdigitation for effective design space exploration using iSIGHT[J]. Journal of Structural and Multidisciplinary Optimiaztion,2002, 23(2).
[38]李金从,邓家(?).多学科优化集成设计框架[J].现代制造工程,2003(3)：77-79.
[39]LDDavis,MMitchell. Handbook of genetic algorithms[M]:Van Nostrand Reinhold, 1991.
[40]JHHolland. Genetic algorithms.1992 Available from: http://www.fortunecity.com/emachines/ell/86/algo.html.
[41]AHWright. Genetic algorithms for real parameter optimization[Z]1991.
[42]董颖,刘欢杰,许宝栋,et al.一种基于实数编码的改进遗传算法[J].东北大学学报,2006,26(4)：321-323.
[43]周明,孙树栋.遗传算法原理及应用[M].北京：国防工业出版社,1999.
[44]雷英杰,张善文.Matlab遗传算法工具箱及应用[M].西安：西安电子科技大学出版社,2005.
[45]DeJongKA. An analysis of the behavior of a class os genetic adaptive systems[D]. Michigan:University of Michigan,1975.
[46]GoldbergDE, DebK. A comparative analysis os selection shemes used in genetic algorithm[J]. Rawling G. Foundation of genetic algorithms,1991.
[47]潘正君,康立山.演化算法[M].北京：清华大学出版社,1998.
[48]史彦军.算法布局的协同差异演化方法与应用研究[D].大连：大连理工大学,2006.
[49]王凌,郑大钟.基于Cauchy和Gauss分布状态发生器的模拟退火算法[J].清华大学学报,2000,40(9)：109-112.
[50]李斌,朱如鹏.基于遗传算法的行星齿轮传动优化及Matlab实现[J].机械设计,2004(10)：29-31.
[51]濮良贵.机械设计[M].北京：高等教育出版社,2006.
[52]刘颖,马剑,彭雪鹏.链传动的优化设计[J].辽宁科技学院学报,2006,7(4)：27-28.
[53]孟兆明,常德功.机械最优设计技术及其应用[M].北京：机械工业出版,2008.
[54]曹玉泉.门式起重机优化设计模型构建研究[J].长安大学学报,2003,20(4)：55-56.
[55]卢雄伟.8m3矿用掘机工作装置优化设计及结构强度分析[D].长春：吉林大学,2009.
[56]丁素芳.装载机系统的可靠性分析及其变速箱的模糊可靠性优化设计研究[D].广西：广西大学,2006.
[2]董立立等.机械优化设计理论方法研究综述[J].机床与液压设计,2010,38(15)：115-1]9.
[3]梁尚明,殷国富.现代机械优化设计方法[M].北京：化学工业出版社,2005.
[4]雷德明,严新平.多目标智能优化算法及其应用[M].北京：科学出版社,2009.
[5]王凌.智能优化算法及其应用[M].北京：清华大学出版社,2000.
[6]吴军.混和进化算法及其工程应用研究[D].西安：西北工业大学出版社,2007.
[7]冯忠绪等.工程机械理论[M].北京：人民交通出版社,2004.
[8]章成器.工程机械优化设计引论[J].工程机械,1982.
[9]吴庆鸣,何小新等.工程机械设计[M].武汉：武汉大学出版社,2006.
[10]王世明,杨为民.国外工程机械新技术新结构和发展趋势[J].国外工程机械新技术新结构和发展趋势,2004,35(1)：61-66.
[11]黄综益,李兴华.工程机械技术发展展望[J].建筑机械化,2000,21(1)：4.
[12]管荣根.工程机械的绿色设计[J].工程机械,2001,35(10)：26-27.
[13]秦东晨,陈江义,胡滨生等.机械结构优化设计的综述与展望[J].中国科技信息,2005(9)：2.
[14]陈育仪.工程机械优化设计[M].北京：中国铁道出版社,1987.
[15]王科社.机械优化设计[M].北京：国防工业出版社,2006.
[16]张运节.机械优化设计综述与展望[J].科技信息,2009：2.
[17]JHHolland. Adaptation in naural and artificial systems[D]. Englewood Cliffs: University of Michigan]975.
[18]黄豪中.柴油均质压燃(HCCI)发动机燃烧过程数值模拟和实验研究[D].天津：天津大学,2007.
[19]PardalosPanosM, RomeijinHEdwin, TuyHoang. Recent developments and trends in global optimization[J]. Journal of Computational and Applied Mathemmatics,2000(124): 204-228.
[20]DeJongK. An analysis of the behavior of a class of genetic adaptive systems[D]. Michigan:University of Michigan,1975.
[21]KrishnakumarK. Micro-genetic algorithms for stationary and nonStationary function optimization[Z]. Philadelphia, PA, USA:1989.
[22]XuYG, LiuGR, WuZP. A novel hybrid genetic algorithm using local optimizer based on heuristic pattern move [J]. Applied Artificial Intelligence,2001,15(7): 601-613.
[23]LiuGR, HanX, OhyoshiT. Computational inverse techniques for material characterization using dynamic response [J].秋田大学学术情报,2002,10(1): 26-33.
[24]LiuGR, MaHJ. Material characterization of composite laminate using dynamic response and real parameter coded micro genetic algorithm[J]. Engineering with Computer,2003(20):295-308.
[25]LiuGR, MaWB, HanX. Inversion of loading time history using displacement response of composite laminates:three-dimensional cases [[J]. Acta Mechanical,.2002, 157(1-4):223-234.
[26]方海鹏,冯恩民,俞红梅.改进的遗传模拟退火算法及在换热网络综合中的应用.[J].应用基础与工程科学学报,,1998,6(1)：1-6.
[27]李明星.基于遗传算法的的门座起重机变幅系统优化设计[D].武汉：武汉理工大学,2006.
[28]唐飞,滕弘飞.一种改进的遗传算法及其在布局优化在的应用[J].软件学报,1999,10(10)： 7.
[29]朱延广,梅珊,赵雯,et al.支持复杂产品总体优化设计的多算法协作优化框架研究[J].系统仿真学,2007,19(11)：2417-2420.
[30]马静敏.基于iSIGHT的电子汽车衡的多学科设计优化[D].青岛：山东科技大学,2005.
[31]耿建光.工程系统过程集成和设计优化--iSIGHT中的过程集成与设计自动化[J].军民两用技术与产品,2002,″″(11).
[32]SLPadula, JJKorte, HJDunn, et al. Multidisciplinary Optimization BranchExperience Using iSIGHT Software[M]. Hampton Virginia:National Aeronautics and Space Administration Langley Research Center 1999.
[33]Chen J Z, Wu Y, Gence C, et al. Integrated electrical and thermal analysis of integrated power electronics modules using iSIGHT[C]. In:Applied Power Electronics Conference and Exposition,2001. APEC 2001. Sixteenth Annual IEEE,2001, 2:1002-1006 vol.2.
[34]Chen X, Yu X, Ji B:Study of Crankshaft Strength Based on iSIGHT Platform and DOE Methods[C]. In:Measuring Technology and Mechatronics Automation (ICMTMA),2010 International Conference on,2010,3:548-551.
[35]杨铭.基于iSIGHT的船型生成及耐波性优化研究[D].武汉：武汉理工大学,2010.
[36]程成,须文波,冷文浩.基于iSIGHT平台DOE方法的螺旋桨敞水性能优化设计[J].计算机工程与应用,2007,28(6)：5.
[37]KochPN, EvansJP, PowellD. Interdigitation for effective design space exploration using iSIGHT[J]. Journal of Structural and Multidisciplinary Optimiaztion,2002, 23(2).
[38]李金从,邓家(?).多学科优化集成设计框架[J].现代制造工程,2003(3)：77-79.
[39]LDDavis,MMitchell. Handbook of genetic algorithms[M]:Van Nostrand Reinhold, 1991.
[40]JHHolland. Genetic algorithms.1992 Available from: http://www.fortunecity.com/emachines/ell/86/algo.html.
[41]AHWright. Genetic algorithms for real parameter optimization[Z]1991.
[42]董颖,刘欢杰,许宝栋,et al.一种基于实数编码的改进遗传算法[J].东北大学学报,2006,26(4)：321-323.
[43]周明,孙树栋.遗传算法原理及应用[M].北京：国防工业出版社,1999.
[44]雷英杰,张善文.Matlab遗传算法工具箱及应用[M].西安：西安电子科技大学出版社,2005.
[45]DeJongKA. An analysis of the behavior of a class os genetic adaptive systems[D]. Michigan:University of Michigan,1975.
[46]GoldbergDE, DebK. A comparative analysis os selection shemes used in genetic algorithm[J]. Rawling G. Foundation of genetic algorithms,1991.
[47]潘正君,康立山.演化算法[M].北京：清华大学出版社,1998.
[48]史彦军.算法布局的协同差异演化方法与应用研究[D].大连：大连理工大学,2006.
[49]王凌,郑大钟.基于Cauchy和Gauss分布状态发生器的模拟退火算法[J].清华大学学报,2000,40(9)：109-112.
[50]李斌,朱如鹏.基于遗传算法的行星齿轮传动优化及Matlab实现[J].机械设计,2004(10)：29-31.
[51]濮良贵.机械设计[M].北京：高等教育出版社,2006.
[52]刘颖,马剑,彭雪鹏.链传动的优化设计[J].辽宁科技学院学报,2006,7(4)：27-28.
[53]孟兆明,常德功.机械最优设计技术及其应用[M].北京：机械工业出版,2008.
[54]曹玉泉.门式起重机优化设计模型构建研究[J].长安大学学报,2003,20(4)：55-56.
[55]卢雄伟.8m3矿用掘机工作装置优化设计及结构强度分析[D].长春：吉林大学,2009.
[56]丁素芳.装载机系统的可靠性分析及其变速箱的模糊可靠性优化设计研究[D].广西：广西大学,2006.