from the conferences organized by TANGER Ltd.
Heat treatment is usually preferred to achieve the pre-determined material properties of steel components thereby suited for several engineering applications. Gas quenching after austenitisation in vacuum is an established process for this purpose, as it is clean and environment friendly. The selection of quenching parameters depends on many factors such as sample geometry as well as material and the batches. The process parameters are adopted by many years of expert knowledge, complex calculations or from trial and error methods. This problem is addressed in this scientific study by developing a prognosis tool, which can predict the heat treatment results based on an artificial neural network (ANN). This is attained by training the ANN on the basis of experimental and numerical investigations. Therefore, the heat treatment experiments were carried out on specific components made from 42CrMo4 and 100Cr6 in a two chambered quenching setup, where N2 gas act as quenching fluid. The cooling behaviour will be investigated under the variation of process parameters such as gas pressure, geometry and batches. The development of the microstructure and hardness as a function of the process parameters are analysed metallographically. For the detailed investigation as well as to improve the training quality of ANN, FEM simulations are developed and validated, which serves afterwards to research the influence of parameter variation numerically. Thereby, sufficient data are generated numerically and experimentally for the successful training of the ANN of the prognosis tool, which can finally predict the heat treatment results.
Keywords: Steel, material properties, heat treatment, FEM, ANN© This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.