A Combined Data Science and Simulation-Based Methodology for Efficient and Economic Prediction of Thermoplastic Performance for Automotive Industry

There are significant predictive tool usages by design engineers in automotive industry to capture material composition and manufacturing process-induced variables. In specific, an accurate modeling of material behavior to predict the mechanical performance of a thermoplastic part is an evolving subject in this field as one needs to consider multiple factors and steps to achieve the right prediction accuracies. The variability in prediction comes from different factors such as polymer type (filled vs. unfilled, amorphous vs semi crystalline etc.), design and manufacturing features (weldline, gate locations, thickness, notches etc.), operating conditions (temperature, moisture etc.) and finally load states (tension, compression, flexural, impact etc.). Using traditional numerical simulation-based modelling to study and validate all these factors requires significant computational time and effort. An alternative method by using data science and AI-ML models is proposed to reduce the overall validation time needed for simulation. To validate this methodology, extensive part level experiments were done on a representative cylindrical geometry to accommodate all these factors using different ULTEM™ Resin materials (PEI). The results show that by using neural network ML model, it is possible to accurately predict the structural response like maximum displacement and force. The ML model results were compared to the CAE based approaches and results overlapped with each other well within the 95% scatter band. By combining both the CAE modelling and ML modelling it is possible to accurately predict the critical structural response of applications more efficiently and economically.