Research On Active And Passive Damage Detection Method Using Elastic Wave Theory

Due to the structural defects,fatigue effect,material aging and external environmental load,cracking is one of the main failure forms in metal structures and concrete structures.To avoid the hazardous incidents caused by crack expansion,cracks must be detected promptly and necessary remedial measures should be taken.Structural health monitoring shows important theoretical and engineering application prospect for structural safety maintenance,such as damage detection,damage location,damage degree assessment and structural residual life prediction.Presently,the detection of surface breaking cracks in the concrete structures and fatigue cracks in metal structures mainly rely on the ultrasonic method,through the linear and non-linear characteristics of the guided wave in the structure.However,due to the attenuation of the guided wave propagation and the sound transmission effect of closed fatigue crack,traditional non-destructive testing is difficult to detect the deep cracks in concretes and closed fatigue cracks in metals.In this paper,the method of combining surface wave and time reversal theory is presented to study the active detection of surface crack in concretes;additionally,the passive detection of fatigue crack in metal plates is also researched based on Lamb wave method and Green's function reconstruction from diffused field.The work of this thesis is divided into two parts:1.A method of surface breaking crack detection of concrete structures based on surface wave and time reversal theory is presented.The finite element numerical simulation is used to study the time reversal process of surface wave excitation and reception in concrete beam.The damage index is proposed according to the correlation coefficient between the reconstructed signals obtained by the surface wave time reversal process and the original excitation signal.The numerical simulation results show that the deeper the surface crack,the greater the distortion between the reconstructed signal and the original signal,and accordingly greater the damage index,which can effectively identify the surface cracks of the concrete structure.2.A method of fatigue crack detection in metallic plate by reconstructing the Green's function from environmental noise is proposed,and the validity and feasibility of the passive detection method with limited ambient noise sources is verified.Fatigue cracks usually generate nonlinear effects,where different wave amplitudes and frequency composition would cause different nonlinear response.This study also undertakes the analysis about the capacity of the proposed approach to identify cracking under different noise amplitudes and frequency ranges.Experimental investigations are conducted in an aluminum plate to assess the cross-correlations of sensing paths and finally detect the prefabricated fatigue crack.Damage index is proposed according to the variation between cross-correlations obtained from pristine crack closed state and the crack opening-closure state while sufficient noise amplitude is used to generate nonlinearity.A probability distribution map of damage is calculated based on damage indices.The fatigue crack introduced in the aluminum plate is successfully identified and orientated,which verifies that fatigue crack can be detected by reconstructing Green's functions from imperfect diffuse field where ambient noise sources locally distributed.