M. Sc. Armin Gießler
Karlsruher Institut für Technologie (KIT) Campus Süd
Institut für Regelungs- und Steuerungssysteme
Geb. 11.20 (Engler-Villa)
Studies of electrical engineering and information technology at Karlsruhe Institute of Technology (KIT) with a semester abroad at Linköping University in Sweden. Internship at Pepperl+Fuchs GmbH in Mannheim in the department of identification systems for factory automation (2017). Bachelor thesis at Vector Informatik GmbH in Stuttgart on the subject of "Optimization of parallelized flash processes within a vehicle" (2019).
Subsequent master studies at KIT with a specialization in control engineering and a semester abroad at the Instituto Superior Técnico in Portugal. Master thesis at the Institute of Control Systems (IRS) on the subject of "Distributed Optimization for Distributed Model Predictive Control" (2021).
Since January 2022 research and teaching assistant at IRS.
Control and coordination of DC microgrids
As part of my research, I deal with the control and coordination of generators, loads and storage devices in DC microgrids. The aim is the efficient and secure control of the dynamic components in the microgrid.
In the coordination task, static operating points are usually calculated, which ensure optimal economy or minimum line losses. These operating points are tracked by real-time capable primary controllers. At the same time, these primary controllers regulate the voltages in the grid such that the dynamic microgrid is stabilized in desired equilibrium points (operating points).
In my research, I try to break up this hierarchical control architecture and rethink conventional controllers, e.g. grid-forming and grid-following controllers. The aim is to achieve optimal transient economy while simultaneously stabilizing the power grid. To achieve this, innovative data-driven controllers, e.g. reinforcement learning controllers, are used, which also take into account input and state constraints. The stability of such controllers can be proven analytically (e.g., by using dissipativity theory) or verified numerically.
|Reinforcement Learning für die zentrale Regelung von DC Microgrids
|Reinforcement Learning für die primäre Regelung von DC Microgrids
|Person in Charge
|Reinforcement Learning für die Primärregelung von DC Microgrids
|Passivity-based Reinforcement Learning
Gießler, A.; Jané-Soneira, P.; Malan, A. J.; Hohmann, S.
2023. 2023 62nd IEEE Conference on Decision and Control (CDC), 1555 – 1562, Institute of Electrical and Electronics Engineers (IEEE). doi:10.1109/CDC49753.2023.10383211