Probabilistic structural health monitoring and reliability analysis

Abstract

Thisthesisinvestigatestheapplicationofprobabilisticmethodstosolveproblemsintheareasofstructuraldamagedetection,crackdetectionandreliabilityanalysisofstructuresinthepresenceofuncertainties.Intheareaofstructuraldamagedetection,anartificialneuralnetwork(ANN)basedmethodthatfollowsthepatternrecognitionapproachisproposedwhichconsidersthedesignoftheANNarchitecture.Thepro-posedstructuraldamagedetectionmethodisverifiedbytheIASC-ASCEbenchmarkstructure.Twotypesofpatternfeatures,modalparametersandRitzvectors,areexaminedinthebenchmarkstudy. Forthedetectionofcracksinbeam-typestructures,atwo-stagemethodologyisproposed.Inthefirststage,di erentclassesofmodelsareemployedtomodelbeamswithdi erentnumbersofcracks.TheBayesianmodelclassselectionmethodisem-ployedtoselectthemostplausibleclassofmodelsbasedonthesetofmeasureddynamicdatatoidentifythenumberofcracks.Inthesecondstage,theposterior(updated)probabilitydensityfunctionsofthecracklocationsandextentsarecal-culatedfollowingtheBayesianstatisticalframework.Theproposedmethodologyisverifiedbybothcomputersimulationsandexperiments. Intheareaofstructuralreliabilityanalysis,theapplicationofimportancesamplingtoestimatethefirst-passageprobabilityofsingle-degree-of-freedomelasto-plasticsys-temsthataresubjectedtowhitenoiseexcitationsisinvestigated.Acomputationallyeÿcientmethodtofindthecriticalexcitationispresented.Thecharacteristicsofthecriticalexcitationareexploredandtheeÿciencyoftheresultingimportancesamplingstrategyiscriticallyassessed.Approximatecriticalexcitations,whicharereferredtoassuboptimalexcitations,aredevelopedandappliedtoimportancesampling.

Date of Award	2 Oct 2007
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Heung Fai LAM (Supervisor) & Siu Kui AU (Co-supervisor)