TY - CONF AB - While trade-offs between modeling effort and model accuracy remain a major concern with system identification, resorting to data-driven methods often leads to a complete disregard for physical plausibility. To address this issue, we propose a physics-guided hybrid approach for modeling non-autonomous systems under control. Starting from a traditional physics-based model, this is extended by a recurrent neural network and trained using a sophisticated multi-objective strategy yielding physically plausible models. While purely data-driven methods fail to produce satisfying results, experiments conducted on real data reveal substantial accuracy improvements by our approach compared to a physics-based model. AU - Schön, Oliver AU - Götte, Ricarda-Samantha AU - Timmermann, Julia ID - 31066 IS - 12 KW - neural networks KW - physics-guided KW - data-driven KW - multi-objective optimization KW - system identification KW - machine learning KW - dynamical systems T2 - 14th IFAC Workshop on Adaptive and Learning Control Systems (ALCOS 2022) TI - Multi-Objective Physics-Guided Recurrent Neural Networks for Identifying Non-Autonomous Dynamical Systems VL - 55 ER -