No abstract

Nowadays, virtual predictions are essential in the design and development of new engineering parts. A critical aspect for virtual predictions is the accuracy of the constitutive model to simulate the material behavior. A state-of-the-art constitutive model generally involves a large number of parameters, and according to classical procedures, this requires many mechanical experiments for its accurate identification. Fortunately, this large number of mechanical experiments can be reduced using heterogeneous mechanical tests, which provide richer mechanical information than classical homogeneous tests. However, the test’s richness is much dependent on the specimen's geometry and can be improved with the development of new specimens. Therefore, this work aims to design a uniaxial tensile load test that presents heterogeneous strain fields using a shape optimization methodology, by controlling the specimen's interior notch shape. The optimization problem is driven by a cost function composed by several indicators of the heterogeneity present in the specimen. Results show that the specimen's heterogeneity is increased with a non-circular interior notch. The achieved virtual mechanical test originates both uniaxial tension and shear strain states in the plastic region, being the uniaxial tension strain state predominant.

No abstract

The simulation of deep drawing processes and its quality is intrinsically dependent on the accuracy of the constitutive model in reproducing the mechanical behaviour of the sheet metal material. Today, the calibration of elastoplastic models – correspondent to the inverse identification of the material parameters – often uses full-field measurements, through Digital Image Correlation (DIC) techniques, to capture non-homogeneous strain fields and states, coupled with non-straightforward numerical inverse methodologies. In the last decade, new parameter identification methodologies, such as the Finite Element Model Updating (FEMU), the Constitutive Equation Gap (CEG) method, the Equilibrium Gap Method (EGM) and the Virtual Fields Method (VFM) have been developed and have proven to be effective for non-linear plasticity models. Nonetheless, the FEMU and the VFM have distinguished themselves from the others. More recently, supervised Machine Learning (ML) techniques have been also used as an inverse identification method. These artificial intelligence-based methods use large datasets of numerical tests to train an inverse model in which the input is the history of the strain field and loads during the test, and the output are directly the material parameters. The goal of this paper is to analyse, compare and discuss these inverse identification methods, with particular focus on the FEMU, VFM, and ML methodologies. A heterogeneous tensile-load test is considered to compare in detail the FEMU, VFM, and ML strategies.

Abstract. The efficient development of metal products with high quality usually requires realistic numerical simulations before the manufacturing procedure. The choice of the constitutive model has a considerable influence on the predicted material behavior’s description. Several material constitutive models have been proposed to describe different mechanical phenomena. However, its selection is a labored task that requires expertise. This lack of knowledge can lead to errors in the numerical predictions and, consequently, large costs and delays in the manufacturing procedure. To overcome this problem, an automatic material model selection tool is necessary. This work aims to compare the impact of different constitutive models and their features on the simulation of a forming process and develop a rational and systematic strategy for model selection. The approach focuses on the study of a hole expansion test using Abaqus and a statistical analysis of variance (ANOVA). It was possible to establish a ranking for the importance of the types of models that can help with model selection decision-making and efficient parameter calibration for accurate predictions.