Andreas Himmel successfully defended his PhD thesis
Research focused on a novel operating strategy for dynamical process operations and the sizing of storage devices.
Andreas Himmel, one of our IMPRS students and doctoral researchers, successfully defended his PhD thesis on the 23rd of June 2021. He joined the IMPRS in 2015 and worked in the Group “Process Systems Engineering” under the supervision of Prof. Sundmacher. The title of his thesis is "Optimization-based Operation Strategy and Storage Design for Coupled Processes".
Nowadays, process operation has to deal with changing feed qualities, electrical energy availability, or prices to ensure an economically optimal operation of the plant. The research of Andreas Himmel focused on the design of a novel operating strategy for dynamical process operations and the sizing of storage devices based on this strategy. He considers generic coupled processes that exchange side products via storages.
The key idea is to operate the system alternately dynamically and stationarily, while the storage systems are operated purely dynamically. In this way, the set of feasible operating points is increased, and economic efficiency is improved. The challenge here is to predict the evolution of the storage level to guarantee a safe operation. For this purpose, the dynamic properties of the process systems have to be integrated while optimizing for new operating points. The result is a bilevel optimization problem, which is hard to solve. The work of Andreas Himmel presents an approach to decouple this optimization problem by techniques of nonlinear control and machine learning. The second part of his work deals with the use of operating strategy to estimate the required and optimal size of the storage. For this purpose, the effect of stochastic scenarios of the changing feed qualities and prices were analyzed. In this way, the minimum and the scenario-based storage size can be determined. The result is storage devices that are large enough to achieve the best possible economic operating points while avoiding oversizing, resulting in reduced investment costs.
