The Use of Machine-Learning Techniques in Material Constitutive Modelling for Metal Forming Processes

Abstract

Accurate numerical simulations require constitutive models capable of providing precise material data. Several calibration methodologies have been developed to improve the accuracy of constitutive models. Nevertheless, a model’s performance is always constrained by its mathematical formulation. Machine learning (ML) techniques, such as artificial neural networks (ANNs), have the potential to overcome these limitations. Nevertheless, the use of ML for material constitutive modelling is very recent and not fully explored. Difficulties related to data requirements and training are still open problems. This work explores and discusses the use of ML techniques regarding the accuracy of material constitutive models in metal plasticity, particularly contributing (i) a parameter identification inverse methodology, (ii) a constitutive model corrector, (iii) a data-driven constitutive model using empirical known concepts and (iv) a general implicit constitutive model using a data-driven learning approach. These approaches are discussed, and examples are given in the framework of non-linear elastoplasticity. To conveniently train these ML approaches, a large amount of data concerning material behaviour must be used. Therefore, non-homogeneous strain field and complex strain path tests measured with digital image correlation (DIC) techniques must be used for that purpose.

Recent Advances and Applications of Machine Learning in Metal Forming Processes

“…The Special Issue is comprised of a total of ten research articles related to ML applications for metal forming processes, including: prediction of forming results [1] and their energy consumption [2]; constitutive modelling [3] and parameters identification [4]; process parameters optimization [4,5]; prediction, detection and classification of defects [6][7][8]; prediction of mechanical properties [9,10]. The following paragraphs summarize the contributions of these works.…”

“…The accuracy of the numerical simulations requires constitutive models capable of describing the mechanical behavior of the material. On this topic, two papers of the special issue [3,4] investigated the application of ML to constitutive modeling and material parameters identification. Lourenço et al [3], explored and discussed the potential contributions of ML in terms of elastoplasticity constitutive modelling.…”

“…On this topic, two papers of the special issue [3,4] investigated the application of ML to constitutive modeling and material parameters identification. Lourenço et al [3], explored and discussed the potential contributions of ML in terms of elastoplasticity constitutive modelling. This work discusses and analyses four recent advances and applications of ML: parameters identification; the enhancement of traditional constitutive models; the development of data-driven constitutive models with embedding physical and empirical knowledge; the development of constitutive models fully based on data-driven approaches.…”

Predicting the efficiency of luminescent solar concentrators for solar energy harvesting using machine learning

“…Over the past decade, the progress of ML has greatly impacted the entire spectrum of physical sciences, including materials science 6 and a reflection of that is the creation of the material genome initiative 7 as a path for accelerating materials development and rationally designing materials through the use of data-driven methods. Other works report the use of ML linear regression methods to build force constant models for thermodynamic properties of materials used in physics and chemistry 8 and the data-centric approach was used for polymer crystals 9 , pharmaceutical science 10 and material constitutive modeling for metal forming processes 11 , with likely many more important developments to come shortly. Also, the importance of the increasing quantity of data recently collected (from experiments and simulations), for the advancement of the ML as a tool for phosphor design has already been discussed 12 .…”

Anomaly Detection in the Automotive Stamping Process: An Unsupervised Machine Learning Approach

“…As a result, there is a rising demand for real-time, data-driven approaches to quality control that can swiftly identify and rectify issues. The incorporation of machine learning models, central to the Industry 4.0 era, offers a powerful approach to achieving the demand [4][5][6][7][8]. These models can analyse real-time data to predict potential defects and enable immediate corrective actions, thereby elevating the efficiency and accuracy of quality control processes [9].…”