Optimization of Open-Die Forging Using Fast Models for Strain, Temperature, and Grain Size in the Context of an Assistance System

Abstract

Besides achieving the intended final shape, one main aim of open-die forging is the adjustment of the mechanical properties by transforming the cast structure into a fine-grained microstructure. To achieve this, the process needs to be designed in a way that ensures achieving all the required part properties, such as grain size, which up to now often requires a lot of operator experience. This paper presents the concept of a forging assistance system, since during forging small deviations from the previously designed pass schedule might add up to unacceptable errors. Such an assistance system requires the evolution of part geometry and surface temperature as input, which are captured with a thermographic camera. The assistance system then uses fast models for equivalent strain, temperature, and microstructure which allow calculation of these properties for the core fibre within seconds on the basis of semi-empirical and physical formulae. However, in the context of an assistance system, which gives real-time advice in case of process deviations, these calculation times are still fairly long, if the hundreds of iteration necessary for process optimization are taken into account. Therefore, three scenarios of deviations, which have to be solved within different time frames, are examined to explore the limits of the chosen classical optimization algorithm.