Transferability of predictive black box models to different application scenarios

The aim of the research was to develop a concept for the parameter optimization of a fused deposition modeling (FDM) process based on predictive models. Within this project empirical models (black box) were created based on a specific application scenario. Therefore, their transferability to other application scenarios has to be analyzed, before they can be utilized for the parameter optimization of the considered process. In addition to the original black box models, two different approaches for adaptive black box models were designed. An experiment for the evaluation of the accuracy of the three different black box modeling approaches was conducted and the results of the analysis will be presented within this paper.