PhD Diagnostics for interconnected complex dynamical systems

Are you passionate about mathematical systems and control theory? Are you interested in working with next generation high-tech systems? The main challenge lies in the reliable operation of the entire system, where many interconnected modules affect each other. In this project, you will be designing algorithms to guarantee the reliable operation of semiconductor machines, together with a highly innovative industrial partner in the Brainport region. If all these sounds fascinating, then this PhD position is made for you!

We invite highly motivated students with a strong background in mathematical control theory, and a keen interest in machine learning to apply for the PhD position within the Dynamics and Control section at the Department of Mechanical Engineering, Eindhoven University of Technology. The mission of the Dynamics and Control Section is to perform research and train next-generation students on the topic of understanding and predicting the dynamics of complex engineering systems in order to develop advanced control, estimation, planning, learning and diagnostics strategies which are at the core of the intelligent autonomous systems of the future: Designing and realizing smart autonomous systems for industry and society.

Complex dynamical systems, such as semiconductor equipment, consist of many interconnected modules, which are functionally, digitally and physically interconnected. The throughput of the equipment relies on continuous runtime while meeting stringent requirements on accuracy and performance. Therefore, monitoring the health of these systems is crucial, which is now largely performed by human experts. The aim of this PhD project is to design innovative monitoring algorithms for complex dynamical systems to automate fault isolation with diagnostics performance guarantees.

Current health monitoring technology is typically designed for either (i) individual system components/modules (which would require removing the component from the system, which is not possible in practice), or (ii) for the system as a whole. The first approach fails to account for the interaction between modules and its effect on the whole system. The second approach makes it challenging to isolate which module may be failing and how to zoom in on the part of the system that is the root cause of the failure. This PhD position will address this challenge by developing hierarchical diagnostic tools for complex dynamical systems.

• Interconnected dynamical systems
• Distributed observer design
• Data-driven approaches
• Machine learning methods
• Fault detection and isolation

This PhD position is part of the project on ‘Holistic Design Automation for Semiconductor Manufacturing Equipment’, in which 6 PhD students will be employed at the Eindhoven University of Technology in the Department of Mechanical Engineering, as well as the departments of Electrical Engineering and Mathematics and Computer Science. You will have the chance to collaborate with a semiconductor company, ASML, to apply the developed results in an industrial context, thereby building both a strong academic and industrial profile.

You will have access to the graduate courses at the Dutch Institute of Systems and Control (DISC) and will have the opportunity to collaborate with industry in the Brainport region and academic researchers worldwide. By joining us, you will be part of a vibrant community of more than 60 researchers including faculty members, postdocs and PhDs working on diverse topics in the field of dynamical systems and control and its applications.

This PhD position is jointly supervised by Nathan van de Wouw, Tom Oomen and Michelle Chong.