Structuralhealthmonitoring(SHM)isarelevanttopicforcivilsystemsandinvolvesthemonitoring,dataprocessingandinterpretationtoevaluatetheconditionofastructure,inordertodetectdamage.Inrealstructures,twoormoresitesortypesofdamagecanbepresentatthesametime.Ithasbeenshownthatonekindofdamagedconditioncaninterferewiththedetectionofanotherkindofdamage,leadingtoanincorrectassessmentaboutthestructurecondition.Identifyingcombineddamageonstructuresstillrepresentsachallengeforconditionmonitoring,becausethereliableidentificationofacombineddamagedconditionisadifficulttask.Thus,thisworkpresentsafusionofmethodologies,whereasinglewavelet-packetandtheempiricalmodedecomposition(EMD)methodarecombinedwithartificialneuralnetworks(ANNs)fortheautomatedandonlineidentification-locationofsingleormultiple-combineddamageinascaledmodelofafive-baytruss-typestructure.Resultsshowedthattheproposedmethodologyisveryefficientandreliableforidentifyingandlocatingthethreekindsofdamage,aswellastheircombinations.Therefore,thismethodologycouldbeappliedtodetection-locationofdamageinrealtruss-typestructures,whichwouldhelptoimprovethecharacteristicsandlifespanofrealstructures.
