In this study, artificial neural networks were used to predict the plastic flow behaviour of S355 steel in the process of high-temperature deformation. The aim of the studies was to develop a model of changes in stress as a function of strain, strain rate and temperature, necessary to build an advanced numerical model of the soft-reduction process. The high-temperature characteristics of the tested steel were determined with a Gleeble 3800 thermo-mechanical simulator. Tests were carried out in the temperature range of 400-1450 °C for two strain rates, i.e. 0.05 and 1 s-1. The test results were next used to develop and verify a rheological model based on artificial neural networks (ANNs). The conducted studies show that the selected models offer high accuracy in predicting the high-temperature flow behaviour of S355 steel and can be successfully used in numerical modelling of the soft-reduction process.
The calculation results of the temperature field during multi-beads GMAW (Gas Metal Arc Welding) cladding of the S355 steel plate are presented in the paper. Numerical simulations were performed using the SysweldR program. Two of Goldak’s heat source models were chosen for calculating the temperature field for each weld bead. The original article achievement is, by selecting the right heat source model and heat loading of the finite elements, obtaining an irregular shape of the fusion zone. This irregular shape of the fusion zone is very complicated to obtain using other commercial programs for numerical welding simulation. The calculation results were verified by the dimensions (critical temperatures) of the heat affected zones (HAZ) determined in the experiment, obtaining a satisfactory agreement.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.