The aim of this project is to develop a process for recycling glycerin waste in the sense of the circular economy. In order to enable the optimal operation of the glycerol waste process in view of the highly fluctuating feedstocks, a digital twin for this process will be created, which will be linked to machine learning methods. The latter are expected to significantly improve the robustness and efficiency of the glycerin waste recycling process by recognizing the feedstock at hand and responding to it in the glycerin waste plant's operating mode, with the goal of minimizing waste to landfill and producing as many usable products as possible. The project will demonstrate the savings by developing specific, commercially relevant use cases.
