Table Of Content

Lecture Notes in Mechanical Engineering Jürgen Gausemeier Franz Josef Rammig Wilhelm Schäfer Walter Sextro Editors Dependability of Self-optimizing Mechatronic Systems Lecture Notes in Mechanical Engineering Forfurthervolumes: http://www.springer.com/series/11236 AboutthisSeries Lecture Notes in Mechanical Engineering (LNME) publishes the latest develop- mentsinMechanicalEngineering-quickly,informallyandwithhighquality.Orig- inal research reported in proceedings and post-proceedings represents the core of LNME. Also considered for publication are monographs, contributed volumes and lecture notesof exceptionallyhigh quality and interest. Volumespublishedin LNMEembraceallaspects,subfieldsandnewchallengesofmechanicalengineer- ing.Topicsintheseriesinclude: • EngineeringDesign • MachineryandMachineElements • MechanicalStructuresandStressAnalysis • AutomotiveEngineering • EngineTechnology • AerospaceTechnologyandAstronautics • NanotechnologyandMicroengineering • Control,Robotics,Mechatronics • MEMS • TheoreticalandAppliedMechanics • DynamicalSystems,Control • FluidMechanics • EngineeringThermodynamics,HeatandMassTransfer • Manufacturing • PrecisionEngineering,Instrumentation,Measurement • MaterialsEngineering • TribologyandSurfaceTechnology · Jürgen Gausemeier Franz Josef Rammig · Wilhelm Schäfer Walter Sextro Editors Dependability of Self-optimizing Mechatronic Systems ABC Editors JürgenGausemeier WilhelmSchäfer ProductEngineering SoftwareEngineering HeinzNixdorfInstitute HeinzNixdorfInstitute UniversityofPaderborn UniversityofPaderborn Paderborn Paderborn Germany Germany FranzJosefRammig WalterSextro DesignofDistributedEmbeddedSystems ChairofMechatronicsandDynamics HeinzNixdorfInstitute UniversityofPaderborn UniversityofPaderborn Paderborn Paderborn Germany Germany ISSN2195-4356 ISSN2195-4364 (electronic) ISBN978-3-642-53741-7 ISBN978-3-642-53742-4 (eBook) DOI10.1007/978-3-642-53742-4 SpringerHeidelbergNewYorkDordrechtLondon LibraryofCongressControlNumber:2013956528 (cid:2)c Springer-VerlagBerlinHeidelberg2014 Thisworkissubjecttocopyright.AllrightsarereservedbythePublisher,whetherthewholeorpartof thematerialisconcerned,specificallytherightsoftranslation,reprinting,reuseofillustrations,recitation, broadcasting,reproductiononmicrofilmsorinanyotherphysicalway,andtransmissionorinformation storageandretrieval,electronicadaptation,computersoftware,orbysimilarordissimilarmethodology nowknownorhereafterdeveloped.Exemptedfromthislegalreservationarebriefexcerptsinconnection with reviews or scholarly analysis or material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’slocation,initscurrentversion,andpermissionforusemustalwaysbeobtainedfromSpringer. PermissionsforusemaybeobtainedthroughRightsLinkattheCopyrightClearanceCenter.Violations areliabletoprosecutionundertherespectiveCopyrightLaw. Theuseofgeneraldescriptivenames,registerednames,trademarks,servicemarks,etc.inthispublication doesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevant protectivelawsandregulationsandthereforefreeforgeneraluse. Whiletheadviceandinformationinthisbookarebelievedtobetrueandaccurateatthedateofpub- lication,neithertheauthorsnortheeditorsnorthepublishercanacceptanylegalresponsibilityforany errorsoromissionsthatmaybemade.Thepublishermakesnowarranty,expressorimplied,withrespect tothematerialcontainedherein. Printedonacid-freepaper SpringerispartofSpringerScience+BusinessMedia(www.springer.com) Preface Intelligenttechnicalsystemsofferthetemptingpossibilitytocreateproductsthatare farmorecapablethantoday’smechatronicsystems.However,thesystem-inherent intelligencecomeswithincreasedsystemcomplexity.Thisleadstoanincreasedrisk forsystem-inherentfaults,whichmighthavedevastatingeffectssuchasacomplete breakdown or might even compromise the safety of users. Yet at the same time the new capabilities can be used to actively compensate for faults. Among these intelligent technical systems are self-optimizing systems, which are able to adapt theirbehaviorbasedonuserdemands,thecurrentsituationorthesystemstateitself. WithintheCollaborativeResearchCenter(CRC)614"Self-OptimizingConcepts andStructuresin MechanicalEngineering"attheUniversityofPaderborn,we de- voted our research to the developmentof these self-optimizing systems. They are basedonmechatronics,whichisformedbyclosecollaborationofmechanical,elec- trical and software engineering. To make a system intelligent, the software engineering aspects gain further significance and are enhanced by advanced control engineeringandmathematicalmethods.Oneresultofourworkistheaccompanying book "Design Methodology for Intelligent Technical Systems", which introduces ourproposeddevelopmentprocessforself-optimizingsystems.Thisprocessneeds be accompaniedby dedicated methodsto assure or increase system dependability asearlyaspossibleinordertoavoidcostlydesignchanges.Wefoundthatthisaug- menting process is so important yet also challenging that we decided to separate all related worksinto this book.It was written by the membersof the CRC 614’s working group"Safety and Stability" and coordinatedby Tobias Meyer, to whom I’dliketoexpressmysincerestgratitude. The results of the CRC 614 and the book at hand present a major milestone towardsthedevelopmentofdependableintelligentsystems.Wehopethatitinspires youtocreatetomorrow’sproducts! YourCRC-Team Paderborn, Prof.Dr.-Ing.JürgenGausemeier October2013 SpeakeroftheCollaborativeResearchCenter614 Acknowledgements ThisbookwaswrittenwithinthecontextoftheCollaborativeResearchCenter614 "Self-OptimizingConceptsandStructuresin MechanicalEngineering",whichhas beenfundedbytheDeutscheForschungsgemeinschaft(DFG). Contents 1 IntroductiontoSelf-optimizationandDependability .............. 1 1.1 Self-optimizingMechatronicSystems ....................... 3 1.1.1 OperatorControllerModule(OCM) ................ 6 1.1.2 BasicProceduresforSelf-optimization............... 7 1.2 Challenges.............................................. 12 1.2.1 DefinitionoftheTermDependability ................ 13 1.2.2 DependabilityasaChallenge....................... 13 1.2.3 EnsuringDependabilityduringtheDevelopmentProcess 14 1.3 ApplicationsofSelf-optimizingSystems..................... 16 1.3.1 RailTechnology–TheRailCabSystem.............. 16 1.3.2 MiniatureRobotBeBot ........................... 20 1.3.3 X-by-WireTestVehicle ........................... 21 1.4 StructureofThisBook.................................... 22 References.................................................... 22 2 DevelopmentofSelf-optimizingSystems ......................... 25 2.1 Domain-SpanningConceptualDesign ....................... 27 2.2 Domain-SpecificDesignandDevelopment................... 31 References.................................................... 36 3 MethodsofImprovingtheDependability ofSelf-optimizingSystems ..................................... 37 3.1 ConceptualDesignPhase.................................. 37 3.1.1 Early Probabilistic Reliability Analysis of an Advanced Mechatronic System Based on Its PrincipleSolution ................................ 38 3.1.2 EarlyDesignoftheMulti-LevelDependabilityConcept 47 3.2 DesignandDevelopment.................................. 54 3.2.1 Increasingthe DependabilityofSelf-optimizing SystemsduringOperationUsingtheMulti-Level DependabilityConcept............................ 55 X Contents 3.2.2 IterativeLearningofStochasticDisturbanceProfiles ... 62 3.2.3 MutationTestingofElectronicComponentDesign..... 69 3.2.4 OptimalControlwithUncertainty................... 76 3.2.5 BehaviorPlanning................................ 84 3.2.6 ComputationofRobustParetoPoints................ 89 3.2.7 Behavior-BasedAdaptationofDifferingModel Parameters ...................................... 95 3.2.8 AnalysisofSelf-healingOperations ................. 102 3.2.9 SafePlanning.................................... 113 3.2.10 VerificationforInteractingMechatronicSystems withMotionProfiles .............................. 119 3.2.11 Dependability-OrientedMultiobjectiveOptimization... 128 3.2.12 Self-healinginOperatingSystems .................. 133 3.2.13 Self-healingviaDynamicReconfiguration............ 141 3.2.14 OnlineModelChecking ........................... 147 3.2.15 Virtualization.................................... 152 3.3 MethodologyfortheSelectionofDependabilityMethodsfor theDevelopmentofSelf-optimizingSystems ................. 158 References.................................................... 162 4 CaseStudy ................................................... 173 4.1 SelectingSuitableMethodsUsingtheMethodology ........... 174 4.2 DevelopmentoftheActiveGuidanceModule................. 178 4.2.1 Early Probabilistic Reliability Analysis of an AdvancedMechatronic Systems Based on Its PrincipleSolution ................................ 178 4.2.2 EarlyDesignoftheMulti-LevelDependabilityConcept 181 4.3 DevelopmentoftheRailCabVehicle ........................ 182 4.3.1 VerificationwithMotionProfilesforInteracting MechatronicSystems ............................. 182 4.3.2 AnalysisofSelf-healingOperations ................. 183 4.3.3 SafePlanning.................................... 184 4.3.4 BehaviorPlanning................................ 185 4.3.5 Behavior-BasedAdaptationofDifferingModel Parameters ...................................... 186 4.3.6 Virtualization.................................... 187 References.................................................... 188 5 ConclusionandOutlook ....................................... 189 Reference..................................................... 190 Index ............................................................ 191 List of Contributors Prof.Dr.-Ing.habil.WilhelmDangelmaier HeinzNixdorfInstitute,UniversityofPaderborn,BusinessComputing,especially CIM,Fuerstenallee11,D-33102Paderborn,Germany, e-mail:[email protected] Prof.Dr.MichaelDellnitz ChairofAppliedMathematics,UniversityofPaderborn,WarburgerStraße100, 33098Paderborn,Germany, e-mail:[email protected] RafalDorociak HeinzNixdorfInstitute,UniversityofPaderborn,ProductEngineering, Fuerstenallee11,D-33102Paderborn,Germany, e-mail:[email protected] KathrinFlaßkamp ComputationalDynamicsandOptimalControl,UniversityofPaderborn,Warburger Straße100,33098Paderborn,Germany, e-mail:[email protected] Prof.Dr.-Ing.JuergenGausemeier HeinzNixdorfInstitute,UniversityofPaderborn,ProductEngineering,Fuerste- nallee11,D-33102Paderborn,Germany, e-mail:[email protected] StefanGroesbrink HeinzNixdorfInstitute,UniversityofPaderborn,DesignofDistributedEmbedded Systems,Fuerstenallee11,33102Paderborn,Germany, e-mail:[email protected] PhilipHartmann HeinzNixdorfInstitute,UniversityofPaderborn,BusinessComputing,especially CIM,Fuerstenallee11,D-33102Paderborn,Germany, e-mail:[email protected]