Ultrasonic Guided Waves Based Delay And Sum Damage Imaging Method By Multi-Band Temporal Filtering Technique

Ultrasonic Lamb waves have the advantages of long propagation distance and sensitivity to damage,and are often used for non-destructive evaluation(NDE)and structural health monitoring(SHM)of plate structures.Research on damage localization imaging method based on Lamb wave has become a research hotspot in the field of SHM and NDE.However,the dispersion characteristics of Lamb waves cause stretching and distortion during propagation,which will affect the extraction of transit time with respect to damage scattering information,resulting in inaccurate positioning and imaging;the effects of dispersion on the incident signal and the aliasing of the damage scatter signal also makes the interpretation of the damage feature information more difficult.Moreover,the existing damage imaging method based on time domain analysis is difficult to effectively utilize the effective damage information contained in all frequency components in the excitation bandwidth,so the imaging accuracy is affected.To overcome,this research work proposes an ultrasonic Lamb wave scattering signal analysis method based on multi-band time domain filtering,which uses wavelet analysis to separate multiple narrowband signals in the excitation band to suppress signal dispersion characteristics,and then constructs a time domain filtering window function to separate the damage scattering information in each narrowband signal.Frequency domain/space delay superimposed damage imaging is performed on each narrowband signal to further avoid the influence of dispersion on damage imaging;finally,the imaging results of each frequency component are comprehensively analyzed(logical' AND')to further improve the accuracy and signal-to-noise ratio of damage imaging.The finite element numerical model of ultrasonic Lamb wave drive,propagation and damage scattering in the prefabricated aluminum plate is established.Based on the numerical simulation data,the proposed multi-band time domain filtering ultrasonic signal processing method and delayed superimposed damage imaging algorithm are used to guide the establishment of experimental research optimization.The experimental platform of ultrasonic Lamb wave damage detection was constructed.The damage imaging experiment was carried out on the aluminum plate with double damage,and the effectiveness of the proposed damage detection method for double damage detection was further verified,which provided theoretical guidance for practical engineering applications.The specific research work of this paper is as follows:Firstly,the research background and significance of the ultrasonic Lamb wave damage detection technology of engineering structure are briefly described.The research progress of the related ultrasonic Lamb wave signal dispersion characteristics processing and damage imaging method is reviewed.Then,based on the basic theory of Lamb waves,the propagation characteristics of Lamb waves in the plate are clarified.Based on this,the signal processing methods and approaches of multi-band time domain filtering used in this research work are proposed.Wavelet transform is used to extract multiple narrowband signal components in the excitation signal bandwidth to eliminate the dispersion effect caused by the wideband signal,and the adaptive time filter window function is used to extract the scattered signal containing the damage feature information.Thirdly,the array beamforming damage imaging method based on Delay and Sum(DAS)is introduced.The phase or spatial delay is applied to each narrowband signal based on the dispersion relationship of Lamb waves to compensate the dispersion characteristics,thereby improving the damage location and the accuracy of the imaging,and further use the logical "AND" operation to comprehensively analyze each narrow-band signal to obtain high-signal-to-noise ratio damage imaging results.Fourth,relying on the COMSOL finite element analysis platform,a three-dimensional simulation model of the damaged aluminum plate was constructed.The time step and mesh size in the numerical simulation process are optimized by relevant theoretical analysis.The influence of Rayleigh damped boundary reflection on the identification of damage scattering signals is designed.The multi-band time domain filtering method is used to separate the damage characteristic information.Scattering signals verify the effectiveness of the frequency domain delay superposition array beamforming damage imaging method.The results show that the phase and spatial delay superposition has better time-delay superposition and better damage accuracy,which can achieve accurate positioning of damage.The imaging results of the respective narrow-band signals are logically summed to further improve the damage imaging accuracy and signal-to-noise ratio,and the preset damage is clearly indicated,which provides a basis for carrying out the damage detection experimental research.Fifthly,according to the above-mentioned signal processing methods and approaches of multiband time domain filtering,an experimental platform for ultrasonic Lamb wave damage detection is constructed.Ultrasonic Lamb wave damage imaging is performed on the aluminum plate with double damage preset,and the method is verified.The effectiveness of double injury.The results show that the multi-band time domain filtering method can effectively extract and separate the double-damage scattering signal without reference signal,and combine the array beamforming delay superposition imaging method to detect the damage,and the logic of the narrow-band signal imaging results and can effectively use wideband signals to obtain high signal-to-noise ratio and high-precision damage imaging results.The method of imaging damage imaging based on ultrasonic Lamb wave proposed in this thesis can provide method and technical guidance for accurately detecting damage of plate and shell structure in engineering practice.The last part summarizes the research work of this thesis and points out the further research direction.