Autonomous exploration and exploitation of structure-property linkages and optimization of microstructures

The development of new materials plays an important role in technological innovations. The focus of interest is the relationship between the process parameters, the local material structure and the resulting properties. Since the local structure of materials essentially dictates their properties, the provision and understanding of structure-property linkages (SPL) are of great importance.
This project aims to establish reliable SPLs quickly and cost-effectively, thereby identifying realistic, manufacturable material structures with optimal properties.
The experimental exploration of SPL causes a considerable technological effort. The achievable limited data basis is a major obstacle for the application of modern machine learning methods. Model-based numerical simulations provide a remedy in the form of synthetic data. A practically relevant algorithm also requires the automated generation of synthetic microstructures and the performance of numerical simulations on selected samples to specifically extend the experimental data base. This procedure, referred to as autonomous research, is developed, implemented and validated in a material-independent manner.
Overall, the synthesis of model-based numerical simulation and data-driven algorithms promises a deeper understanding of the mechanisms of action underlying SPLs and an acceleration of material development.