Practical Guidelines to Train Reinforcement Learning Based Control of Electrical Drives

Konferenz: PCIM Conference 2025 - International Exhibition and Conference for Power Electronics, Intelligent Motion, Renewable Energy and Energy Management
06.05.2025 - 08.05.2025 in Nürnberg, Germany

doi:10.30420/566541245

Tagungsband: PCIM Conference 2025

Seiten: Sprache: EnglischTyp: PDF

Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt

Autoren:
Szecsenyi, Nandor; Stumpf, Peter

Inhalt:
With the recent advancements made in Artificial Intelligence, it is possible that Reinforcement Learning based data driven methods can become a next generation technology to control electrical drives instead of the classical model-based techniques. However, providing the correct setup and hyperparameters for training the agent is usually not evident, so the paper aims to lay some practical guidelines on how to properly set the environmental parameters during training. These recommendations are backed up by a detailed comparison where the effect on the performance of each environmental aspect is evaluated.