Xin-SheYangandSlawomirKoziel(Eds.) ComputationalOptimizationandApplicationsinEngineeringandIndustry StudiesinComputationalIntelligence,Volume359 Editor-in-Chief Prof.JanuszKacprzyk SystemsResearchInstitute PolishAcademyofSciences ul.Newelska6 01-447Warsaw Poland E-mail:[email protected] Furthervolumesofthisseriescanbefoundonour Vol.349.VitorR.Carvalho homepage:springer.com ModelingIntentioninEmail,2011 ISBN978-3-642-19955-4 Vol.339.HalinaKwasnicka,LakhmiC.Jain(Eds.) Vol.350.ThanasisDaradoumis,SantiCaballe´, InnovationsinIntelligentImageAnalysis,2010 AngelA.Juan,andFatosXhafa(Eds.) ISBN978-3-642-17933-4 Technology-EnhancedSystemsandToolsforCollaborative Vol.340.HeinrichHussmann,GerritMeixner,and LearningScaffolding,2011 DetlefZuehlke(Eds.) ISBN978-3-642-19813-7 Model-DrivenDevelopmentofAdvancedUserInterfaces,2011 Vol.351.NgocThanhNguyen,BogdanTrawin´ski,and ISBN978-3-642-14561-2 JasonJ.Jung(Eds.) Vol.341.StéphaneDoncieux,NicolasBredeche,and NewChallengesforIntelligentInformationandDatabase Jean-BaptisteMouret(Eds.) Systems,2011 NewHorizonsinEvolutionaryRobotics,2011 ISBN978-3-642-19952-3 ISBN978-3-642-18271-6 Vol.352.NikBessisandFatosXhafa(Eds.) Vol.342.FedericoMontesinoPouzols,DiegoR.Lopez,and NextGenerationDataTechnologiesforCollective AngelBarrigaBarros ComputationalIntelligence,2011 MiningandControlofNetworkTrafficbyComputational ISBN978-3-642-20343-5 Intelligence,2011 ISBN978-3-642-18083-5 Vol.353.IgorAizenberg Complex-ValuedNeuralNetworkswithMulti-Valued Vol.343.KuroshMadani,AntónioDouradoCorreia, Neurons,2011 AgostinhoRosa,andJoaquimFilipe(Eds.) ISBN978-3-642-20352-7 ComputationalIntelligence,2011 ISBN978-3-642-20205-6 Vol.354.LjupcoKocarevandShiguoLian(Eds.) Chaos-BasedCryptography,2011 Vol.344.AtillaElc¸i,MamadouTadiouKoné,and ISBN978-3-642-20541-5 MehmetA.Orgun(Eds.) SemanticAgentSystems,2011 Vol.355.YanMengandYaochuJin(Eds.) ISBN978-3-642-18307-2 Bio-InspiredSelf-OrganizingRoboticSystems,2011 ISBN978-3-642-20759-4 Vol.345.ShiYu,Léon-CharlesTranchevent, BartDeMoor,andYvesMoreau Vol.356.SlawomirKozielandXin-SheYang(Eds.) Kernel-basedDataFusionforMachineLearning,2011 ComputationalOptimization,MethodsandAlgorithms,2011 ISBN978-3-642-19405-4 ISBN978-3-642-20858-4 Vol.346.WeisiLin,DachengTao,JanuszKacprzyk,ZhuLi, Vol.357.NadiaNedjah,LeandroSantosCoelho, EbroulIzquierdo,andHaohongWang(Eds.) VivianaCoccoMariani,andLuizadeMacedoMourelle(Eds.) MultimediaAnalysis,ProcessingandCommunications,2011 InnovativeComputingMethodsandTheirApplicationsto ISBN978-3-642-19550-1 EngineeringProblems,2011 ISBN978-3-642-20957-4 Vol.347.SvenHelmer,AlexandraPoulovassilis,and FatosXhafa Vol.358.NorbertJankowski,W(cid:2)lodzis(cid:2)lawDuch,and ReasoninginEvent-BasedDistributedSystems,2011 KrzysztofGra¸bczewski(Eds.) ISBN978-3-642-19723-9 Meta-LearninginComputationalIntelligence,2011 ISBN978-3-642-20979-6 Vol.348.BeniaminoMurgante,GiuseppeBorruso,and AlessandraLapucci(Eds.) Vol.359.Xin-SheYangandSlawomirKoziel(Eds.) Geocomputation,SustainabilityandEnvironmental ComputationalOptimizationandApplicationsin Planning,2011 EngineeringandIndustry,2011 ISBN978-3-642-19732-1 ISBN978-3-642-20985-7 Xin-SheYang and Slawomir Koziel (Eds.) Computational Optimization and Applications in Engineering and Industry 123 Editors Xin-SheYang SlawomirKoziel MathematicsandScientificComputing, EngineeringOptimization&ModelingCenter, NationalPhysicalLaboratory, SchoolofScienceandEngineering, Teddington, ReykjavikUniversity,Menntavegur1, MiddlesexTW110LW,UK 101Reykjavik,Iceland E-mail:[email protected] E-mail:[email protected] ISBN 978-3-642-20985-7 e-ISBN 978-3-642-20986-4 DOI 10.1007/978-3-642-20986-4 Studiesin Computational Intelligence ISSN1860-949X Library of Congress Control Number:2011927164 (cid:2)c 2011 Springer-VerlagBerlin Heidelberg Thisworkissubjecttocopyright.Allrightsarereserved,whetherthewholeorpart of the material is concerned, specifically therights of translation, reprinting,reuse ofillustrations, recitation,broadcasting, reproductiononmicrofilm orinanyother way, and storage in data banks. Duplication of this publication or parts thereof is permittedonlyundertheprovisionsoftheGermanCopyrightLawofSeptember9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution undertheGerman Copyright Law. The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Typeset&CoverDesign:ScientificPublishing ServicesPvt. Ltd., Chennai, India. Printed on acid-free paper 9 8 7 6 5 4 3 2 1 springer.com Preface Computersimulationaswellasnumericalmodellingandoptimizationarebe- coming a commonplace in contemporary engineering and science. The com- plexityofthesystemsconsideredinvariousfieldshasbeenconstantlygrowing over the years and the theoretical models offer more and more accurate de- scription of the physical phenomena, structures and devices. However, most of these models are far too complicated to be handled through analytical solutions;computer simulationis requiredfor a majority of real-worldappli- cations not only to evaluate the model but also to exploit it in the design process. Advanced state-of-the-art commercial simulation software packages are available and used in everyday design work in mechanical engineering, civilengineering,aerospaceindustry,electricalengineering,andmanyothers. Computational optimization has become an essential and, in many cases, critical component of the design process. In almost all applications in engi- neering and industry it is necessary to maximize performance and efficiency while minimizing the cost, size, weight, or energy consumption at the same time. This is usually a complex task that involves manipulation of available design parameters in order to find satisfactory values of one or more objec- tives that are evaluated through often computationally expensive computer simulation. In many cases, complex constraints have to be satisfied in the optimization process. Thereareseveralfactorsduetowhichthesearchforoptimaldesigncanbe complicatedevenfurther.Oneofthemisthepresenceofuncertaintiesthatis common for most real-world systems. In particular, material properties and geometryofthemanufactureddevicemaydifferfromtheirnominalvaluesasa resultoffabricationtolerances.Therefore,theoptimizationprocessmayseek for the robust design which ensures the highest probability of satisfying the performancerequirementsunderthepresenceofuncertaintiesratherthanjust for the optimal design. Many optimization problems are nonlinear and NP- hard,thatis,thesolutiontimeforfindingoptimaldesigngrowsexponentially with the problem size. In some cases the designer may face multiple local optima and the global search procedures are necessary. On the other hand, many practical problems have multiple and competing objectives where the VI Preface best design is obtained through a decision-making process based on a set of Pareto-optimalsolutions. The dependence of contemporary engineering design on computer simula- tions introduces additional difficulties to optimization. Growing demand for accuracy and ever-increasing complexity of structures and systems result in the simulation process being more and more time consuming. In many engi- neeringfields,theevaluationofasingledesigncantakeaslongasseveraldays or even weeks so that straightforwardapproaches by employing high-fidelity simulator directly in the optimization loop are prohibitive. Interestingly, the increasing computational power of today’s computers does not alleviate this problem because the availability of faster computers is compromised by the tendency of simulating more and more complex structures and systems with higher and higher accuracy. On the other hand, simulation-based objective functions are inherently noisy, which makes the optimization process even more difficult. Still, simulation-driven design becomes a must for growing number of areas, which creates a need for robust and efficient optimization methodologiesthat canyield satisfactorydesigns evenat the presence of an- alyticallyintractableobjectivesandlimited computationalresources.Inpar- ticular, any technique that improves the efficiency of simulators or reduces the function evaluation count is crucially important. Surrogate-based and knowledge-based optimization uses certain approximations to the objective soastoreducethecostofobjectiveevaluations.Theapproximationsareoften local,whilethequalityofapproximationsisevolvingastheiterationsproceed. Extensive research conducted in the area of computational optimization and modeling resulted in many techniques that alleviate the difficulties of traditionaldesign optimizationmethodologies.Many of these techniques ad- dress particular issues, such as multiple local optima, multiple objectives or handling computationally expensive cost functions. Substantial progress has been observedin the developmentof derivative-freeoptimization techniques, the use of adjoint sensitivities, as well as methods exploiting surrogatemod- els, both function-approximation- and physically-based. Thisbookiscontributedfromworldwideexpertswhoareworkinginthese excitingareas,andeachchapterispracticallyself-contained.Thisbookstrives to review and discuss the latest developments concerning optimization and modellingwithafocusonapplicationsforsolvingreal-worldproblemsinvar- iousdisciplinesofengineeringandscience,includingaerodynamics,oilindus- try, gas and water transport, microwave engineering, structural engineering, navigation, civil engineering, and others. We would like to thank our editors, Drs Thomas Ditzinger and Holger Schaepe, and staff at Springer for their help and professionalism. Last but not least, we thank our families for their help and support. Xin-She Yang Slawomir Koziel 2011 List of Contributors Editors Xin-She Yang Mathematics and Scientific Computing, National Physical Laboratory, Teddington, Middlesex TW11 0LW, UK. ([email protected]) Slawomir Koziel Engineering Optimization & Modeling Center, School of Science and Engineering, Reykjavik University, Menntavegur 1, 101 Reykjavik, Iceland. ([email protected]) Contributors Amir Hossein Alavi School of Civil Engineering, Iran University of Science and Technology, Tehran, Iran. (ah [email protected]) I. Aydogdu DepartmentofEngineeringSciences,MiddleEastTechnicalUniversity,Ankara, Turkey. ([email protected]) Lindsay Chapman Materials Processing and Performance,National Physical Laboratory, Teddington, Middlesex TW11 0LW, UK. ([email protected]) Ana Cort´es Departament d’Arquitectura de Computadors i Sistemes Operatius, Escola d’Enginyeria,Universitat Auto`noma de Barcelona, 08193 Bellaterra (Barcelona), Spain. ([email protected]) David Echeverr´ıa Ciaurri DepartmentofEnergyResourcesEngineering,StanfordUniversity,Stanford, CA 94305,USA. ([email protected]) VIII List of Contributors Mo´nica Denham Facultad de Ingenier´ıa, Universidad Nacional de R´ıo Negro, Sede Andina, San Carlos de Bariloche,Argentina. Pia Domschke DepartmentofMathematics,TechnischeUniversit¨atDarmstadt,Dolivostr.15, D-64293 Darmstadt, Germany. ([email protected]) Louis J. Durlofsky DepartmentofEnergyResourcesEngineering,StanfordUniversity,Stanford, CA 94305-2220,USA. ([email protected]) Amir Hossein Gandomi College of Civil Engineering, Tafresh University, Tafresh, Iran. ([email protected]) Z.W. Geem Information Technology Program, iGlobal University, Annandale, Virginia, USA. ([email protected]) Golnar Gharooni-fard Computer Department of Islamic Azad University, Mashhad Branch, Young Researchers Club, Iran. ([email protected]) O. Hasancebi Department ofCivil Engineering,Middle EastTechnicalUniversity,Ankara, Turkey. ([email protected]) Momin Jamil Harman International,Automotive Division, Becker-G¨oringStr. 16, D-76307 Karlsbad, Germany. ([email protected]) Oliver Kolb DepartmentofMathematics,TechnischeUniversit¨atDarmstadt,Dolivostr.15, D-64293 Darmstadt, Germany. ([email protected]) Slawomir Koziel Engineering Optimization & Modeling Center, School of Science and Engineering, Reykjavik University, Menntavegur 1, 101 Reykjavik, Iceland. ([email protected]) List of Contributors IX Jens Lang DepartmentofMathematics,TechnischeUniversit¨atDarmstadt,Dolivostr.15, D-64293 Darmstadt, Germany. Center of Smart Interfaces, Technische Universit¨at Darmstadt, Petersenstr.30, D-64287 Darmstadt, Germany. Graduate School Computational Engineering,Technische Universit¨at Darmstadt, Dolivostr. 15, D-64293 Darmstadt, Germany. ([email protected]) Leifur Leifsson Engineering Optimization & Modeling Center, School of Science and Engineering, Reykjavik University, Menntavegur 1, 101 Reykjavik, Iceland. ([email protected]) Tom`as Margalef Departament d’Arquitectura de Computadors i Sistemes Operatius, Escola d’Enginyeria,Universitat Auto`noma de Barcelona, 08193 Bellaterra (Barcelona), Spain. ([email protected]) Clare Matthews Mathematics and Scientific Computing, National Physical Laboratory, Teddington, Middlesex TW11 0LW, UK. ([email protected]) Fahime Moein-darbari Computer Department of Islamic Azad University, Mashhad Branch, Young Researchers Club, Iran. ([email protected]) Tapan Mukerji DepartmentofEnergyResourcesEngineering,StanfordUniversity,Stanford, CA 94305-2220,USA. ([email protected]) Stanislav Ogurtsov Engineering Optimization & Modeling Center, School of Science and Engineering, Reykjavik University, Menntavegur 1, 101 Reykjavik, Iceland. ([email protected]) Simon Roberts National PhysicalLaboratory,Teddington, Middlesex TW11 0LW, UK. M.P. Saka Department of Engineering Sciences, Middle East Technical University, Ankara, Turkey. ([email protected]) X List of Contributors Kerstin Wendt Departament d’Arquitectura de Computadors i Sistemes Operatius, Escola d’Enginyeria,Universitat Auto`noma de Barcelona, 08193 Bellaterra (Barcelona), Spain. ([email protected]) Louise Wright Mathematics and Scientific Computing, National Physical Laboratory, Teddington, Middlesex TW11 0LW, UK. ([email protected]) Xin-She Yang Mathematics and Scientific Computing, National Physical Laboratory, Teddington, Middlesex TW11 0LW, UK. ([email protected]) Hans-Ju¨rgen Zepernick Blekinge Institute of Technology, SE-37179Karlskrona,Sweden. ([email protected])