多目标遗传算法及其在船舶型线优化中的应用
|
摘要
最优化问题是一个古老的问题,追求目标的最优化一直是人类的理想。
许多工程优化问题往往性质十分复杂,很难用传统的优化方法来求解。而且实际的工程优化问题中大多数是多目标优化问题,各子目标之间一般都存在互相冲突的现象。多目标与单目标优化问题的本质区别是,前者一般是一组或多组非劣解的集合,而后者只是单个解或一组不连续的解。因此,多目标优化问题的求解变得困难。
自60年代以来,人们对求解多目标优化问题的兴趣日益增加。一种模仿生物进化过程的、被称为“进化算法(Evolutionary Algorithm)”的随机优化技术应运而生,而且在解这类优化问题中显示出了优于传统优化算法的性能。而遗传算法(Genetic Algorithm)是迄今为止进化算法中应用最多、比较成熟、广为人知的算法。由于其在求解复杂优化问题的巨大潜力及其在工业工程领域的成功应用,这种算法受到了广泛的关注。经过几十年的发展,多目标遗传算法已经日趋成熟。
船舶的优化设计是从60年代末期开始逐渐发展起来的一种有效、新的设计方法。船舶优化设计同许多其他工程问题一样属于多目标优化问题。其中船舶型线优化设计通常是以应用数学方法对型线进行光顺,而以船体的布置、水动力和结构性能为目标函数。型线优化设计是一个亟待解决的复杂多目标优化问题。
本文以多目标遗传算法为主要研究内容,将改进的遗传算法应用于求解多目标优化问题并用于船舶优化设计是很有意义的。本文介绍了遗传算法的起源、历程、主要研究方向、遗传算法的基本原理以及改进措施等,编制基于改进的带精英策略的非支配排序遗传算法(NSGAⅡ)的通用多目标优化软件。本文以高速方尾船型优化为例介绍了多目标遗传算法在船型的型线优化设计当中的应用,找到了一种适合NPL 100 Model A船型的数学描述方法,在此基础上,以总阻力系数为优化目标,优化得到阻力性能优良的船舶型线。这说明多目标遗传算法在船型优化中的应用可以大大地提高了船舶优化设计质量、缩短设计周期。
The optimization problem is an ancient question,and to get the optimization solutions has been humanity's ideal.
However,most project optimization problems are so complex that it is very difficult to solve them with the traditional optimization methods.Moreover most real project optimization questions are the multi-objectives optimization problems with more then one targets between which there exist conflicts mutually.The solution of multi-objectives optimization problems is a group or many groups of Non-inferior Set,while the simple target one always has only one solution or one group of non-continual solutions,which is the essential,distinguish between them.So it is much more difficult to solve the multi-objectives optimization problems.
Since the 60s,people's interest on solving the multi-objective optimization questions has increased day by day.And a kind of random optimize search method, called evolutionary algorithm,refer to the evolution law of biological circle was put forward.It is testified that this kind algorithm is of more excellent performance on solving the optimization problems than the traditional ones.And the genetic algorithm contained in the evolutionary algorithm is widely known and mostly used. It has received the widespread attention because of its great potential application in the solution of complex optimization question and actual project optimization problems.After several dozens year development,multi-objective genetic algorithm is already being mature day by day.
Ships' optimization design is a new effective design method which started from the late 60s and was developed gradually.As like as other actual project optimization problems,it is a multi-objectives optimization problem.And we always take hull's arrangement,hydrodynamic force and structure performance as objective function to get the optimization body lines described by mathematics methods.The line optimization design is the complex multi-objective optimization question which urgently waits to be solved
This paper study of multi-objective genetic algorithm and the application in multi-objectives optimization problems and optimization ship design.It is greatly significative.This paper,first of all,introduced the origin of genetic algorithm,the developing course,leading research direction,basic principle and the improved measures etc.and a Multi-Objective Optimization software base on NSGAⅡwas developed.Then,Optimization body lines Design of the high speed displacement transom stern ship was researched.In this paper,author find one mathematics description method of body lines which is suited for NPL 100 Model A,and then take the total resistance coefficient as the optimized goal,the optimization obtains a ship lines with better resistance performance.This proved that the application of multi-objective genetic algorithm can greatly enhance the ship design quality and shorten the design cycle time.
许多工程优化问题往往性质十分复杂,很难用传统的优化方法来求解。而且实际的工程优化问题中大多数是多目标优化问题,各子目标之间一般都存在互相冲突的现象。多目标与单目标优化问题的本质区别是,前者一般是一组或多组非劣解的集合,而后者只是单个解或一组不连续的解。因此,多目标优化问题的求解变得困难。
自60年代以来,人们对求解多目标优化问题的兴趣日益增加。一种模仿生物进化过程的、被称为“进化算法(Evolutionary Algorithm)”的随机优化技术应运而生,而且在解这类优化问题中显示出了优于传统优化算法的性能。而遗传算法(Genetic Algorithm)是迄今为止进化算法中应用最多、比较成熟、广为人知的算法。由于其在求解复杂优化问题的巨大潜力及其在工业工程领域的成功应用,这种算法受到了广泛的关注。经过几十年的发展,多目标遗传算法已经日趋成熟。
船舶的优化设计是从60年代末期开始逐渐发展起来的一种有效、新的设计方法。船舶优化设计同许多其他工程问题一样属于多目标优化问题。其中船舶型线优化设计通常是以应用数学方法对型线进行光顺,而以船体的布置、水动力和结构性能为目标函数。型线优化设计是一个亟待解决的复杂多目标优化问题。
本文以多目标遗传算法为主要研究内容,将改进的遗传算法应用于求解多目标优化问题并用于船舶优化设计是很有意义的。本文介绍了遗传算法的起源、历程、主要研究方向、遗传算法的基本原理以及改进措施等,编制基于改进的带精英策略的非支配排序遗传算法(NSGAⅡ)的通用多目标优化软件。本文以高速方尾船型优化为例介绍了多目标遗传算法在船型的型线优化设计当中的应用,找到了一种适合NPL 100 Model A船型的数学描述方法,在此基础上,以总阻力系数为优化目标,优化得到阻力性能优良的船舶型线。这说明多目标遗传算法在船型优化中的应用可以大大地提高了船舶优化设计质量、缩短设计周期。
The optimization problem is an ancient question,and to get the optimization solutions has been humanity's ideal.
However,most project optimization problems are so complex that it is very difficult to solve them with the traditional optimization methods.Moreover most real project optimization questions are the multi-objectives optimization problems with more then one targets between which there exist conflicts mutually.The solution of multi-objectives optimization problems is a group or many groups of Non-inferior Set,while the simple target one always has only one solution or one group of non-continual solutions,which is the essential,distinguish between them.So it is much more difficult to solve the multi-objectives optimization problems.
Since the 60s,people's interest on solving the multi-objective optimization questions has increased day by day.And a kind of random optimize search method, called evolutionary algorithm,refer to the evolution law of biological circle was put forward.It is testified that this kind algorithm is of more excellent performance on solving the optimization problems than the traditional ones.And the genetic algorithm contained in the evolutionary algorithm is widely known and mostly used. It has received the widespread attention because of its great potential application in the solution of complex optimization question and actual project optimization problems.After several dozens year development,multi-objective genetic algorithm is already being mature day by day.
Ships' optimization design is a new effective design method which started from the late 60s and was developed gradually.As like as other actual project optimization problems,it is a multi-objectives optimization problem.And we always take hull's arrangement,hydrodynamic force and structure performance as objective function to get the optimization body lines described by mathematics methods.The line optimization design is the complex multi-objective optimization question which urgently waits to be solved
This paper study of multi-objective genetic algorithm and the application in multi-objectives optimization problems and optimization ship design.It is greatly significative.This paper,first of all,introduced the origin of genetic algorithm,the developing course,leading research direction,basic principle and the improved measures etc.and a Multi-Objective Optimization software base on NSGAⅡwas developed.Then,Optimization body lines Design of the high speed displacement transom stern ship was researched.In this paper,author find one mathematics description method of body lines which is suited for NPL 100 Model A,and then take the total resistance coefficient as the optimized goal,the optimization obtains a ship lines with better resistance performance.This proved that the application of multi-objective genetic algorithm can greatly enhance the ship design quality and shorten the design cycle time.
引文
[1]胡毓达.实用多目标最优化[M].上海:上海科学技术出版社,1990
[2]金鸿章.王科俊.何琳.遗传算法理论及其在船舶横摇运动控制中的应用.哈尔滨工程大学出版社.P2-5
[3]金鸿章.王科俊.何琳.遗传算法理论及其在船舶横摇运动控制中的应用.哈尔滨工程大学出版社.PIO
[4]赵瑞.多目标遗传算法应用的研究:[硕士学位论文].天津:天津大学理学院.2005年6月
[5]Rosenbeg R S,Simulation of genetic Populations with biochemical Properties.PhD.thesis,University of Michigan,Ann Habor,Michigan,1967
[6]Cheng,R.and M.Gen,A survey of genetic multi-objective optimizations Techinical report,Ashikaga Institute of Technology,1998
[7]Ishibuchi,H.and T.Murata,A multi-objective genetic local search algorithm and its application to flowshop scheduling,IEEE Transactions on Systems,Man and Cybernetics,1998,28(3):392403
[8]Murata,T,H.Ishibuchi,and H.Tanaka,Multi-objective genetic algorithm and its application to flowshop scheduling,Computers and Industrial Engineering,1996,30(4):957968
[9]Fonseca C M,Flemin P J.An overview of evolutionary algorithms in multi-objective optimization.Evolutionary Computation,1995,3(1):1-16
[10]Srinivas N,Deb K.Multi-objective Optimization Using Non-dominated Sorting in Genetic Algorithms.Evolutionary Computation,1994,2(3):221-48
[11]Rudolph G.Evolutionary search under partially ordered sets[Z].Technical Report No CI -67/99,Dortmund:Department of Computer Science/LSll,University of Dortmund,Germany,1999
[12]Zitzler E,Deb K,Thiele L Comparison of multi-objective evolutionary algorithms:Empirical results[J].Evolutionary Computation,Journal,1999,8(2):173-196
[13]Deb K,Aarawal S,Pratap A,et al.A fast elitist non-dominated sorting genetic algorithm for multi-objective optimization:NSGA-II[A].Proc of the Parallel Problem Solving from Nature V1 Con[C].Paria,2000.849-858
[14]朱建才.多目标优化方法库的开发与应用研究:[硕士学位论文].西北工业大学.2006年3月
[15]郑丽君.基于遗传算法的多目标优化与决策方法研究:[硕士学位论文].国防科学技术大学.2003年11月
[16]梦红云.多目标进化算法及其应用研究:[博士学位论文].西安:西安电子科技大学.2005年9月
[17]苏勇彦.单目标、多目标优化进化算法及其应用:[硕士学位论文].武汉:武汉理工大学.2007年4月
[18]Homaifar A,Lai S&Qi X.Constrained optimization via genetic algorithm.Simulationo,1994,62(4),242-254
[19]Jonies J&Houck C.On the use of non-stationary penalty functions to solve nonlinear constrained optimization problems with gas.In:Michalcwicz Z,Schaffer J D,Schwefel H P Fogel D B,&Kitano H Eds,Proceedings of first IEEE International Conference on Evolutionary Comutation,Piscataway,NJ:IEEE Press,1994.579-584
[20]J.D.Schaffer,Multiple Objective Optimization with Vector Evaluated Genetic Algorithms,First International Conference on Genetic Algorithms,1985
[21]金鸿章.王科俊.何琳.遗传算法理论及其在船舶横摇运动控制中的应用.哈尔滨:哈尔滨工程大学出版社.P6
[22]Eckart Zitzle,Lgthar Thiele,Multiobjective Optimization using Evolutionary Algorithms.A Comparative Case study,Lecture Notes in Computer Science,1998
[23]潘正君.康立山.陈毓平.遗传算法[M].北京:清华大学出版社,1998
[24]高媛.非支配排序遗传算法(NSGA)的研究与应用.[硕士学位论文]江苏:浙江大学.2006年3月
[25]王达.遗传算法用于多目标过程优化综合的研究:[硕士学位论文].山东:青岛科技大学.2005年4月
[26]S.Kirpartrick,C.D.Gelatt Jr.,M.P.Vechhi,Optimization by Simulated Annealing,Science.,Number4598,May 1983
[27]Joanna Lis,A.E.Eiben,Multi-Sexual Genetic Algorithm for Multiobjective Optimization,International Conference on Evolutionary Computation,1996
[28]Masahiro Tanaka,Hikaru Watanabe,Yasuyuki Furukawa,GA-based Decision Support System for Multi-Criteria Optimization,IEEE International Conference on Systems,Man and Cybernetics-2,Intelligent Systems for the 21 st Century,1995
[29]柳存根.裘泳铭.姚震球.林杰人.遗传进化算法在船舶初步设计中的应用.上海交通大学学报.2000年1月
[30]冯志强.程智斌.非线性规划在船型方案设计中的应用.船舶.2004年2
[31]陈宾康.函数合成法生成型线与光顺.中国造船,1992(1)
[32]陈宾康.赵成壁.一种生成高速方尾排水型船舶型线的数学方法.武汉造船.1996(1)
[33]邓宝林.高速排水型方尾船舶的型线优化设计研究.[硕士学位论文].武汉:武汉理工大学.2007年5月
[34]第六机械工业部第七研究院第七0二研究所.高速园般排水船型设计参考资料.译文79-005,1969.10
[35]黄德波.赵连恩.朱念昌.高速方尾船理论优化船型设计.哈尔滨船舶工程学院学报p17-26.1985.2
[36]邹劲.史冬岩等.计算机辅助船舶设计.哈尔滨工程大学出版社.2002.7
[37]Baba E.An application of wave pattern analysis to ship form improvement.J SNA,Japan,1972,32
[38]王能超.数值分析简明教程.高等教育出版社.1984.12
[39]钱能.C++程序设计教程.清华大学出版社。2004.08
[40]黄维通.Visual C++面向对象与可视化程序设计(第二版).清华大学出版社.2004.07
[41]冯恩德.席龙飞.船舶设计原理.大连海运学院出版社.1990.08.
[42]李世谟.船舶阻力.人民交通出版社.1989.02
[43]张则松.LNG运输船总体设计与型线优化研究..[硕士学位论文].大连:大连理工大学.2004年3月
[2]金鸿章.王科俊.何琳.遗传算法理论及其在船舶横摇运动控制中的应用.哈尔滨工程大学出版社.P2-5
[3]金鸿章.王科俊.何琳.遗传算法理论及其在船舶横摇运动控制中的应用.哈尔滨工程大学出版社.PIO
[4]赵瑞.多目标遗传算法应用的研究:[硕士学位论文].天津:天津大学理学院.2005年6月
[5]Rosenbeg R S,Simulation of genetic Populations with biochemical Properties.PhD.thesis,University of Michigan,Ann Habor,Michigan,1967
[6]Cheng,R.and M.Gen,A survey of genetic multi-objective optimizations Techinical report,Ashikaga Institute of Technology,1998
[7]Ishibuchi,H.and T.Murata,A multi-objective genetic local search algorithm and its application to flowshop scheduling,IEEE Transactions on Systems,Man and Cybernetics,1998,28(3):392403
[8]Murata,T,H.Ishibuchi,and H.Tanaka,Multi-objective genetic algorithm and its application to flowshop scheduling,Computers and Industrial Engineering,1996,30(4):957968
[9]Fonseca C M,Flemin P J.An overview of evolutionary algorithms in multi-objective optimization.Evolutionary Computation,1995,3(1):1-16
[10]Srinivas N,Deb K.Multi-objective Optimization Using Non-dominated Sorting in Genetic Algorithms.Evolutionary Computation,1994,2(3):221-48
[11]Rudolph G.Evolutionary search under partially ordered sets[Z].Technical Report No CI -67/99,Dortmund:Department of Computer Science/LSll,University of Dortmund,Germany,1999
[12]Zitzler E,Deb K,Thiele L Comparison of multi-objective evolutionary algorithms:Empirical results[J].Evolutionary Computation,Journal,1999,8(2):173-196
[13]Deb K,Aarawal S,Pratap A,et al.A fast elitist non-dominated sorting genetic algorithm for multi-objective optimization:NSGA-II[A].Proc of the Parallel Problem Solving from Nature V1 Con[C].Paria,2000.849-858
[14]朱建才.多目标优化方法库的开发与应用研究:[硕士学位论文].西北工业大学.2006年3月
[15]郑丽君.基于遗传算法的多目标优化与决策方法研究:[硕士学位论文].国防科学技术大学.2003年11月
[16]梦红云.多目标进化算法及其应用研究:[博士学位论文].西安:西安电子科技大学.2005年9月
[17]苏勇彦.单目标、多目标优化进化算法及其应用:[硕士学位论文].武汉:武汉理工大学.2007年4月
[18]Homaifar A,Lai S&Qi X.Constrained optimization via genetic algorithm.Simulationo,1994,62(4),242-254
[19]Jonies J&Houck C.On the use of non-stationary penalty functions to solve nonlinear constrained optimization problems with gas.In:Michalcwicz Z,Schaffer J D,Schwefel H P Fogel D B,&Kitano H Eds,Proceedings of first IEEE International Conference on Evolutionary Comutation,Piscataway,NJ:IEEE Press,1994.579-584
[20]J.D.Schaffer,Multiple Objective Optimization with Vector Evaluated Genetic Algorithms,First International Conference on Genetic Algorithms,1985
[21]金鸿章.王科俊.何琳.遗传算法理论及其在船舶横摇运动控制中的应用.哈尔滨:哈尔滨工程大学出版社.P6
[22]Eckart Zitzle,Lgthar Thiele,Multiobjective Optimization using Evolutionary Algorithms.A Comparative Case study,Lecture Notes in Computer Science,1998
[23]潘正君.康立山.陈毓平.遗传算法[M].北京:清华大学出版社,1998
[24]高媛.非支配排序遗传算法(NSGA)的研究与应用.[硕士学位论文]江苏:浙江大学.2006年3月
[25]王达.遗传算法用于多目标过程优化综合的研究:[硕士学位论文].山东:青岛科技大学.2005年4月
[26]S.Kirpartrick,C.D.Gelatt Jr.,M.P.Vechhi,Optimization by Simulated Annealing,Science.,Number4598,May 1983
[27]Joanna Lis,A.E.Eiben,Multi-Sexual Genetic Algorithm for Multiobjective Optimization,International Conference on Evolutionary Computation,1996
[28]Masahiro Tanaka,Hikaru Watanabe,Yasuyuki Furukawa,GA-based Decision Support System for Multi-Criteria Optimization,IEEE International Conference on Systems,Man and Cybernetics-2,Intelligent Systems for the 21 st Century,1995
[29]柳存根.裘泳铭.姚震球.林杰人.遗传进化算法在船舶初步设计中的应用.上海交通大学学报.2000年1月
[30]冯志强.程智斌.非线性规划在船型方案设计中的应用.船舶.2004年2
[31]陈宾康.函数合成法生成型线与光顺.中国造船,1992(1)
[32]陈宾康.赵成壁.一种生成高速方尾排水型船舶型线的数学方法.武汉造船.1996(1)
[33]邓宝林.高速排水型方尾船舶的型线优化设计研究.[硕士学位论文].武汉:武汉理工大学.2007年5月
[34]第六机械工业部第七研究院第七0二研究所.高速园般排水船型设计参考资料.译文79-005,1969.10
[35]黄德波.赵连恩.朱念昌.高速方尾船理论优化船型设计.哈尔滨船舶工程学院学报p17-26.1985.2
[36]邹劲.史冬岩等.计算机辅助船舶设计.哈尔滨工程大学出版社.2002.7
[37]Baba E.An application of wave pattern analysis to ship form improvement.J SNA,Japan,1972,32
[38]王能超.数值分析简明教程.高等教育出版社.1984.12
[39]钱能.C++程序设计教程.清华大学出版社。2004.08
[40]黄维通.Visual C++面向对象与可视化程序设计(第二版).清华大学出版社.2004.07
[41]冯恩德.席龙飞.船舶设计原理.大连海运学院出版社.1990.08.
[42]李世谟.船舶阻力.人民交通出版社.1989.02
[43]张则松.LNG运输船总体设计与型线优化研究..[硕士学位论文].大连:大连理工大学.2004年3月