About the employer
KU Leuven is an autonomous university. It was founded in 1425. It was born of and has grown within the Catholic tradition.
Visit the employer pageKU Leuven is an autonomous university. It was founded in 1425. It was born of and has grown within the Catholic tradition.
Visit the employer pageThe M-Group (Mechatronics Group) of KU Leuven and the L2EP of the University of Lille are looking for a PhD student in the framework of a Global Joint PhD partnership. The PhD topic focusses on creating reliable and safe electrical drivetrains, which are essential for electrifying the industrial landscape towards Industry 4.0 and 5.0. The project will leverage the numerical tools (code_Carmel) of L2EP and reliable and fault-tolerant control algorithms of the M-Group to thoroughly investigate different aspects of this complex behavior. The goal of the PhD project is to establish a proof-of-concept for one or multiple methodologies/algorithms that can actively dampen the rotor under various scenarios involving a combination of load, vibrations, environmental conditions.
The M-Group research group is a multidisciplinary team comprising hardware and software engineers, doctoral researchers, and professors. They specialize in ensuring the reliable and sustainable performance of systems involving electrical energy, automation, electronics, signal processing, and ICT. The group, consisting of 8 professors and 45 research assistants or PhD students, is based at the new Bruges Campus and has state-of-the-art research equipment and infrastructure for conducting tests on embedded software or communication devices. The available infrastructure also allows PhD students and research assistants to design and develop robust and dependable hardware or software design techniques. The research group has been successful in acquiring prestigious funding such as H2020, Marie Skłodowska-Curie Actions, and national funding (FWO, imec.ICON, etc.). They have a long-standing collaboration with other research groups from KU Leuven (imec.Distrinet, LMSD, etc.). The key researchers involved in this project from MGroup are prof. Hans Hallez expertise on sensors and sensor networks with edge computing, prof. Davy Pissoort, whit his expertise in dependable mechatronic systems and dr. Dries Vanoost with his expertise in condition monitoring and electric drive systems. Additionally, our well-equipped laboratories for validating the dependability of programmable electronics, including a fully equipped EMC testing laboratory, a Highly Accelerated Lifetime Testing Chamber, and a mechatronic test-bench, will provide an excellent platform for the project.
The Laboratory of Electrical Engineering and Power Electronics of Lille (https://l2ep.univ-lille.fr/) brings together the research work of 4 research teams, including the team ‘Numerical Tools and Methods’. The latter is a research group focused on the development of numerical models and effective methodologies aimed to the optimal design and study of electromagnetic devices within their environment. The developed models are used to design and study the performance of electrical machines in different operating cases but also in faulty cases for diagnosis purposes, quantify losses, and explore demagnetization effects in permanent magnets. The key researchers involved in this project from L2EP are Prof. Abdelmounaim Tounzi, who specializes in numerical modeling of electrical machines, and Yvonnick Le Menach, with expertise in the development of numerical modeling and its use to study electromagnetic devices. The modelling work will use code-_Carmel, a 3D finite element electromagnetic calculation software co-developed with EDF R&D
Mechatronic systems face a significant hurdle in the form of mechanical vibrations, leading to performance degradation, higher costs, and decreased reliability[1]. As reported by JEMA[2], bearing failures account for 30% to 40% of overall motor faults. This problem is intensified by the high stiffness of modern mechatronic systems, allowing vibrations to reach the electric motor rotor with little damping. Furthermore, these vibrations are being amplified due to the unbalanced magnetic pull [3](UMP).
To address these challenges, we aim to improve condition monitoring and motor control algorithms to dampen vibrations without needing more hardware. Our approach involves multiphysics simulations to study the complex interplay between magnetic fields, motor input, and vibration dynamics. This will enable the development of cost-effective surrogate models using data-driven techniques like Deep Koopman models, which will be utilized for creating resilient control algorithms.
Furthermore, by using Deep Koopman linearization models, we aim to integrate rotor eccentricity estimation from stator voltages and currents to enhance fault detection without invasive measurements. Here, our goal is to optimize rotor eccentricity and prevent accelerated damage under varying operating conditions by integrating linear control methods with Koopman-based models.
We are also looking to shift from predictive maintenance to active solution strategies, driven by machine learning, to revolutionize electric motor maintenance. However, accurately predicting and implementing mitigation strategies presents its own set of challenges. These strategies necessitate multiphysics simulation encompassing electrical, mechanical, control, and magnetic aspects. These interconnected simulations, utilizing finite element method (FEM)-based analysis, will provide a comprehensive resource for optimizing electric drive control strategy.
Within the project, the candidate will be responsible for developing a new control algorithm which, in real-time, can dampen vibrations caused by eccentricity errors using advanced Deep Koopman models. These models enable the linearization of the inherent non-linear dynamics of the electrical machine using data-driven approaches, allowing for the application of control algorithms to correct the eccentric behavior during rotation. The candidate will also document and present these demonstrations to an industrial audience.
1 Desenfans, P., Gong, Z., Vanoost, D., Gryllias, K., Boydens, J., Pissoort, D. (2023). The influence of the unbalanced magnetic pull on fault-induced rotor eccentricity in induction motors. Journal Of Vibration And Control. doi: 10.1177/10775463231162908
2 On recommended interval of updating induction motors, Tech. rep., The Japan Electrical Manufacturers’ Association (JEMA) (2000).
3 Z. Gong, P. Desenfans, D. Pissoort, H. Hallez and D. Vanoost, "Multiphysics Coupling Model to Characterise the Behaviour of Induction Motors With Eccentricity and Bearing Faults," in IEEE Transactions on Energy Conversion, vol. 39, no. 1, pp. 146-159, March 2024, doi: 10.1109/TEC.2023.3305717.
KU Leuven strives for an inclusive, respectful and socially safe environment. We embrace diversity among individuals and groups as an asset. Open dialogue and differences in perspective are essential for an ambitious research and educational environment. In our commitment to equal opportunity, we recognize the consequences of historical inequalities. We do not accept any form of discrimination based on, but not limited to, gender identity and expression, sexual orientation, age, ethnic or national background, skin colour, religious and philosophical diversity, neurodivergence, employment disability, health, or socioeconomic status. For questions about accessibility or support offered, we are happy to assist you at this email address.
KU Leuven is an autonomous university. It was founded in 1425. It was born of and has grown within the Catholic tradition.
Visit the employer page