BA | SA - On Understanding of gym-electric-motor (GEM) Toolbox
- Institut
- Lehrstuhl für Nachhaltige Mobile Antriebssysteme
- Typ
- Bachelorarbeit Semesterarbeit
- Inhalt
- theoretisch
- Beschreibung
The simulation is an important aspect throughout the whole stages of research of electrical drive systems. Simulation softwares such as Maxwell and MATLAB/Simulink provide a bundle of functions and interfaces after years of developement in classic modelling and control of electricmagnetical as well as mechanical dynamics within electrical drives. With the emerging Machine Learning techniques such as Neural Networks and Reinforcement Learning being applied to approximate and control real systems, the Python-based simulation is showing great capability of powering Machine Learning techniques in programming-environment with their user-defined scalability and adapability to parallel computing.
The gym-electric-motor (GEM) toolbox is a Python package for the simulation and control of various electric motors. It is built upon Faram Gymnasium Environments, and, therefore, can be used for both, classical control simulation and Reinforcement Learning experiments. The goal of the thesis is to understand the mechanism of GEM package and to utilize the toolbox to build electric motor environment models for Reinforcement Learning control.
Your tasks:
- Study of instruction of GEM toolbox
- Study of examples of GEM toolbox cookbook and examples
- Set up and configuration of GEM toolbox in Python environment
- Building of electric motor environment models
- Implementing of usecases of Reinforcement Learning control
- Voraussetzungen
- Interest in modelling and control of electrical drives
- Interest in Machine Learning
- Experience in Python
- Good English knowledge
- Möglicher Beginn
- sofort
- Kontakt
-
M.Sc. Kai Cui
Raum: 2107.EG.008
Tel.: +49 8928924108
k.cuitum.de