Online Algorithms for Optimal Control of Hybrid Propulsion Systems (OASys)

OASys is an interdisciplinary project between Computer and Engineering Sciences to develop online algorithms for the optimal control of hybrid propulsion systems. Hybrid propulsion systems consist of at least two kind of propulsion systems. The control of hybrid propulsion systems should combine the engines in such a way that the required driving force is achieved while the overall energy consumption is minimized. We consider parallel hybrid cars that are equipped with a combustion and an electrical engine.

Our approach combines optimization methods from the areas of control theory and artificial intelligence with online learning. Offline generated control laws for the optimal control are selected online such that a nearly optimal control strategy is built dynamically. Mathematical models and realistic simulations are used for the evaluation of our control strategies.

Funding

This project is funded 2012-2016 by the German Research Foundation (DFG) under the project id AB 461/2-1.

People

Institut für Regelungstechnik, RWTH Aachen University

Dirk Abel
Frank-Josef Heßeler
Jan Maschuw
Matthias Hoppe
Martina Josevski
Rainer Gasper

Informatik 1, RWTH Aachen University

Berthold Vöcking
Walter Unger
Sascha Geulen
Melanie Winkler

Lehr- und Forschungsgebiet Theory Hybrider Systeme, RWTH Aachen University

Erika Ábrahám
Johanna Nellen
Ulrich Loup

Publications

2016
DownloadJohanna Nellen. Analysis and Synthesis of Hybrid Systems in Control Engineering. Phd Thesis at RWTH Aachen University, 2016.
Martina Josevski, Dirk Abel. Flatness-based Trajectory Planning for the Battery State of Charge in Hybrid Electric Vehicles. Proceedings of the 8th IFAC Symposium on Advances in Automotive Control (AAC'16), pages 49(11):134–140, IFAC-PapersOnLine, 2016.
2015
DOISascha Geulen, Martina Josevski, Johanna Nellen, Janosch Fuchs, Lukas Netz, Benedikt Wolters, Dirk Abel, Erika Abraham, Walter Unger. Learning-based Control Strategies for Hybrid Electric Vehicles. Proc. of the 2015 IEEE Conf. on Control Applications (CCA'15), pages 1722–1728, IEEE, 2015.
Sascha Geulen, Martina Josevski, Johanna Nellen, Janosch Fuchs, Lukas Netz, Benedikt Wolters, Erika Abraham, Walter Unger, Dirk Abel. Online Lernen als Kontrollstrategie in Hybridfahrzeugen. Proc. of the 7th VDI/VDE Fachtagung AUTOREG: Auf dem Weg zum automatisierten Fahren, Volume 2233 of VDI-Berichte, pages 101–112, VDI Verlag, 2015.
LinkJohanna Nellen, Benedikt Wolters, Lukas Netz, Sascha Geulen, Erika Abraham. A Genetic Algorithm based Control Strategy for the Energy Management Problem in PHEVs. Proc. of the 1st Global Conference on Artificial Intelligence (GCAI'15), Volume 36 of EPiC Series in Computer Science, pages 196–214, EasyChair, 2015.
2014
Melanie Winkler, Sascha Geulen, Martina Josevski, Michael Tegethoff, Dirk Abel, Berthold Vöcking. Online Parameter Tuning Methods for Adaptive ECMS Control Strategies in Hybrid Electric Vehicles. Proc. of the FISITA 2014 World Automotive Congress (FISITA'14), , 2014.
Martina Josevski, Dirk Abel. Energy Management of Parallel Hybrid Electric Vehicles Based on Stochastic Model Predictive Control. of the 19th World Congress The International Federation of Automatic Control (IFAC'14), Volume 19, pages 2132-2137, IFAC Papers Online, 2014.
Martina Josevski, Dirk Abel. Multi-time scale Model Predictive Control Framework for Energy Management of Hybrid Electric Vehicles. Proc. of the IEEE 53rd Annual Conference on Decision and Control (CDC'14), pages 2523-2528, , 2014.
2013
Melanie Winkler. Algorithms for Online Buffering Problems and Applications to the Power Control of a Hybrid Electric Vehicle. Phd Thesis at RWTH Aachen University, Computer science Department, 2013.