Lamb wave damage severity estimation using ensemble-based machine learning method with separate model network

Abstract

Lamb wave-based damage estimation have great potential for structural health monitoring. However, designing a generalizable model that predicts accurate and reliable damage quantification result is still a practice challenge due to complex behavior of waves with different damage severities. In the recent years, machine learning (ML) algorithms have been proven to be an efficient tool to analyze damage-modulated Lamb wave signals. In this study, ensemble-based ML algorithms are employed to develop a generalizable crack quantification model for thin metallic plates. For this, the scattering of Lamb wave signals due to different configuration of crack dimension and orientation is extensively studied. Various finite element simulations signals, representing distinct crack severities in term of crack length, penetration and orientation are acquired. Realizing that both temporal and spectral information of signal is extremely important to damage quantification, three time-frequency (TF) based damage sensitive indices namely energy concentration, TF flux and coefficient of energy variance are proposed. These damage features are extracted by employing smoothed-pseudo Wigner–Ville distribution. After that data augmentation technique based on the spline-based interpolation is applied to enhance the size of the dataset. Eventually, these fully developed damage dataset is deployed to train ensemble-based models. Here we propose separate model network, in which different models are trained and then link together to predict new and unseen datasets. The performance of the proposed model is demonstrated by two cases: first simulated data incorporated with high artificial noises are employed to test the model and in the second scenario, experimental data in raw form are used. Results indicate that the proposed model has the potential to develop a general model that yields reliable answer for crack quantification.

LSTM-Based Autoencoder with Maximal Overlap Discrete Wavelet Transforms Using Lamb Wave for Anomaly Detection in Composites

“…Secondly, unlike linear Time-Frequency (TF) distributions, which provide only an approximate energy distribution in the TF domain, SPWVD accurately represents the true TF energy distribution of the signal. This precise estimation enables the identification and assessment of damage-related features with high accuracy [29].…”

From data to insight, enhancing structural health monitoring using physics-informed machine learning and advanced data collection methods

“…For most of the literature, supervised learners are used for determining the damage size, width, penetration or orientation. For example, in the article [87] simulationbased datasets with all the parameters of a crack (length, depth and orientation) were simultaneously used to extract the damage indices. Considering all these features made the datasets heavily skewed and imbalanced and therefore it would have been difficult for a single ML model to precisely predict the severity of the crack.…”

Elastic Wave Mechanics in Damaged Metallic Plates

“…The increase in the RMSD value suggests better damage detection at higher frequencies because damage detection is wavelength-dependent. It is worthy of note that damage has a higher chance of detection if its size is larger than the probing wavelength [14]. From the linear fitting of the observation data, an empirical regression model with an R 2 of about 0.97 was deduced and expressed in Equation (22).…”

“…Many studies have been performed in detecting unfilled damage, with limited studies on detecting filled damage using GWUT. In [14], the authors studied the relationship between the scattering of damage signals and different severities of damage using the S 0 Lamb wave mode of the pitch-catch method. The severities considered were notch length, orientation and notion depth with respect to the plate thickness.…”