Abstract:As the fatigue problem of orthotropic steel plates becomes increasingly severe in medium- and long-span bridges, structural health monitoring of the plates that could identify cracks timely is in great demand. Considering that orthotropic steel plates are of multiple structural elements and complex geometries, it is challenging to locate the cracks accurately. To this end, a novel acoustic emission (AE) source location method based on empirical wavelet transform and long short-term memory neural network was developed for crack monitoring of orthotropic steel plates. Empirical wavelet transform adaptively decomposed prominent components contained in the raw AE signal, of which the one with distinct dispersive modes was selected as the characteristic component to more precisely determine the arrival time through Akiake information criteria. The long short-term memory model established a more exact relationship between the arrival time differences of AE signals at multiple sensors and the source coordinates. With the help of empirical wavelet transform and long short-term memory model, it significantly eliminated the influence of dispersion, reflection, scattering, multiple propagation paths and noise, and thus guaranteed higher crack location accuracy. Through field experiment on a long-span suspension bridge, where cracks were simulated by artificial AE sources, a comprehensive comparison was carried out among traditional time-of-arrival method, delta-T mapping method, improved delta-T mapping method based on Akiake information criteria and Gaussian process, a counterpart of the proposed method based on empirical wavelet transform and Gaussian process, as well as the proposed method based on empirical wavelet transform and long short-term memory neural network. The location errors and their variations of the proposed method were found far less than those of the other existing methods. The location accuracy could be further improved by increasing the grid resolution of pencil lead break pre-tests. The results demonstrated the feasibility and superiority of the proposed crack location method for field applications on large-scale complex structures like orthotropic steel plates.

A three-stage approach based on 1D-CNNs for AE source localisation on historic fibrous plaster ceilings

Ultrasonic Array Tomography-Oriented Subsurface Crack Recognition and Cross-Section Image Reconstruction of Reinforced Concrete Structure Using Deep Neural Networks

Data level fusion of acoustic emission sensors using deep learning

Automated crack identification in structures using acoustic waveforms and deep learning

Acoustic Emission Characterization of Damage Patterns in Composite Scarf Joints Using Unsupervised Learning

All-phase fast Fourier transform and multiple cross-correlation analysis based on Geiger iteration for acoustic emission sources localization in complex metallic structures

A novel acoustic emission source location method for crack monitoring of orthotropic steel plates

A Multi-channel Input Framework for Structure Displacement Response Prediction Using Convolutional Neural Network

A Deep Learning Method for Damage Prognostics of Fiber-Reinforced Composite Laminates Using Acoustic Emission

Acoustic emission-based damage localization using wavelet-assisted deep learning

Using Acoustic Emission Technique for Structural Health Monitoring of Laminate Composite: A Novel CNN-LSTM Framework

Experimental Study on Monitoring Damage Progression of Basalt-FRP Reinforced Concrete Slabs Using Acoustic Emission and Machine Learning

Localizing damage on stainless steel structures using acoustic emission signals and weighted ensemble regression-based convolutional neural network

False Ceiling Deterioration Detection and Mapping Using a Deep Learning Framework and the Teleoperated Reconfigurable 'Falcon' Robot

Two-dimensional acoustic emission source localization on layered engineered wood by machine learning: a case study of laminated veneer lumber plate structure

Damage monitoring of masonry structures using the acoustic emission technique – From tensile and shear characterization tests to shear wall tests

The research of acoustic emission techniques for non destructive testing and health monitoring on civil engineering structures

Experimental and Numerical Investigation of Acoustic Emission Source Localization Using an Enhanced Guided Wave Phased Array Method

Predicting local material thickness from steady-state ultrasonic wavefield measurements using a convolutional neural network

Classification of cracking sources of different engineering media via machine learning

Deep learning-based monitoring of surface residual stress and efficient sensing of AE for laser shock peening