Abstract

Tracking algorithms constitute an efficient numerical technique for modelling fracture in quasi-brittle materials. They succeed in representing localized cracks in the numerical model without mesh-induced directional bias. Currently available tracking algorithms have an important limitation: cracking originates either from the boundary of the discretized domain or from predefined “crack-root” elements and then propagates along one orientation. This paper aims to circumvent this drawback by proposing a novel tracking algorithm that can simulate cracking starting at any point of the mesh and propagating along one or two orientations. This enhancement allows the simulation of structural case-studies experiencing multiple cracking. The proposed approach is validated through the simulation of a benchmark example and an experimentally tested structural frame under in-plane loading. Mesh-bias independency of the numerical solution, computational cost and predicted collapse mechanisms with and without the tracking algorithm are discussed.

Abstract

This research study employs Convolutional Neural Network (CNN) for the purpose of identifying damage related to cracks in aluminum plates. Successfully identifying features associated with damage is a pivotal step within the damage identification process. However, the process of identifying these features is known to be arduous and complex in its implementation. CNN, as a profound deep learning model, offers the unique capability to learn representations of high-dimensional data that traditional mathematical models often struggle to depict accurately. In order to harness this advantageous trait of CNN, the current study transforms the task of non-destructive evaluation pertaining to notched crack damage within metal plates into an image classification challenge within the domain of machine learning. Subsequently, a deep CNN model, trained on images derived from Lamb wave data transformation, is employed to tackle this issue effectively. Additionally, by utilizing the analytical formula of Lamb signals, numerical simulation techniques are utilized to generate Lamb signals, which are then coupled with the algorithm devised within this study to identify damage present within metal plate structures. Finally, the verification of aluminum plate damage identification was carried out. The research results show that the method proposed in this paper can efficiently and accurately identify the damage of metal plate structures. '''Keyword: '''Lamb waves; Deep learning; Damage localization.