Master Thesis on Data-Based Modelling of Electric Drives for Reinforcement Learning-Based Controller Design

The performance and efficiency of electric drives are fundamentally determined by their control methods and modulation schemes. While conventional approaches rely on simplified models and control structures these limitations often restrict optimal performance in real-world applications. Reinforcement Learning (RL) has emerged as a promising solution offering the potential to enhance performance through more sophisticated models and control structures e.g. direct switching control which directly manipulates the switching time instants of the inverter terminals. However RL agents trained in simulation environments using simplified models frequently experience performance gaps when deployed in real-world scenarios. The main objective of this thesis is the development of an innovative electric drive model suitable for a direct switching controller design using reinforcement learning.

• During your thesis you will conduct a comprehensive review of existing literature on data-based modeling techniques and advanced control strategies applied to electric drives.
• You will develop a novel and effective concept for systematically exciting electric drive systems. The primary objective is to generate rich and informative training data that accurately captures the switching behavior.
• Based on the collected training data and insights from the literature review you will develop an advanced electric drive model that precisely captures the switching behavior combining physics-based and data-based modeling techniques.
• An optional extension involves training and evaluating a direct switching controller using reinforcement learning techniques and the developed models.
• Finally you will thoroughly document all developed concepts and results culminating in a comprehensive thesis report.

Qualifications :

• Education: Master studies in the field of Cybernetics Computer Science Engineering Mathematics or comparable
• Experience and Knowledge: profound knowledge of machine learning and control theory; experience in MATLAB/Simulink Python and ideally in DL frameworks; knowledge of electrical machines is a plus
• Personality and Working Practice: you work autonomously with a systematic practice and analytic thinking quickly grasping concepts and structuring tasks
• Languages: very good in English

Additional Information :

Start: according to prior agreement  
Duration: 6 months

Requirement for this thesis is the enrollment at university. Please attach your CV transcript of records examination regulations and if indicated a valid work and residence permit.

Diversity and inclusion are not just trends for us but are firmly anchored in our corporate culture. Therefore we welcome all applications regardless of gender age disability religion ethnic origin or sexual identity.

Need further information about the job
Felix Berkel (Functional Department)
 92301

#LI-DNI

Remote Work :

No

Employment Type :

Full-time