Prediction of strain distribution during the plane strain tensile test based on artificial neural networks

Meknassi, Raid Fekhreddine and Béres, Gábor József and Lukács, Zsolt (2023) Prediction of strain distribution during the plane strain tensile test based on artificial neural networks. DESIGN OF MACHINES AND STRUCTURES, 13 (1). pp. 74-83. ISSN 1785-6892

Preview

Text Meknassi_2.pdf Download (623kB) | Preview

Abstract

In this study, finite element method was used to study the effects of various notch geometries on the strain field distributions during the plane strain tensile test for cold-rolled steel (DC01). The artificial neural network approach (ANN) and the response surface methodology (RSM) were adopted to develop the mathematical prediction models applied in the optimization procedure. The strain state was expressed by self-defined metrics, namely, the Plane Strain State Index (PSSI) and the Homogeneity Index (HI) were predicted by changing the notch angle (X degree), notch width (d mm), and notch length (c mm). The Quadratic mathematical models obtained by the RSM, and ANN presented the evolution of PSSI, and HI based on (X, d, and c). The results show that the ANN method provides more precise results compared to those of the RSM approach.

Actions (login required)

Edit Item