Title

AI-SUPPORTED MATERIAL SIMULATION OF FORMING PROCESSES

AI-SUPPORTED MATERIAL SIMULATION OF FORMING PROCESSES

Authorship

Kruse, Michael; Wehage, Doris

Kruse, Michael

I am this author

Wehage, Doris

I am this author

DOI

10.5151/2594-5297-39816

Downloads

Abstract

Realistic material simulation is becoming increasingly important in the mapping of forming processes. The more accurate the material data and models are, the better the match between the simulation and the results of the production process will be. Thus, AI methods are also increasingly used in material simulation, which leads to further improvement of the simulation results. The paper demonstrates the use of AI in the form of neural networks for the simulation of phase transformation processes in steels during cooling from the forming heat. In this context, new possibilities arise compared to the usual use of CCT diagrams: On the one hand, the influence of changes in chemical analysis and austenitization condition can be gleaned. On the other hand, temperature-time histories with variable cooling rates can be better understood. The results significantly improve the representation of the forming process in simulation programs. Two application examples show the influence on the development of microstructures in long products.

Realistic material simulation is becoming increasingly important in the mapping of forming processes. The more accurate the material data and models are, the better the match between the simulation and the results of the production process will be. Thus, AI methods are also increasingly used in material simulation, which leads to further improvement of the simulation results. The paper demonstrates the use of AI in the form of neural networks for the simulation of phase transformation processes in steels during cooling from the forming heat. In this context, new possibilities arise compared to the usual use of CCT diagrams: On the one hand, the influence of changes in chemical analysis and austenitization condition can be gleaned. On the other hand, temperature-time histories with variable cooling rates can be better understood. The results significantly improve the representation of the forming process in simulation programs. Two application examples show the influence on the development of microstructures in long products.

Keywords

AI methods; Neural networks, Material simulation, Phase transformation, CCT diagram, Forming process

AI methods; Neural networks, Material simulation, Phase transformation, CCT diagram, Forming process