Develop data-enabled system and control co-design algorithms, synergizing control theory with novel machine learning techniques, to reinforce the performance of future-engineered systems.
In the last decades, the classical control theory has proven its effectiveness in terms of analysis and control (design) methodologies. Control theory has been a key enabler in the realization of complex systems. However, increasing system complexity leads to an increasingly indirect relationship between linear to practically meaningful performance. Moreover, the intricate requirements for system performance complicate controller design through the sole use of the classical control theory. Synergizing established fundamental control with promising data-enabled machine-learning techniques could effectively solve these present-day design challenges.
You will develop algorithms to effectuate efficient data-enabled systems and control co-design approaches. Therefore, you will tightly synergize the established control theory with novel data-enabled techniques from the fields of machine learning (ML) and artificial intelligence (AI). This allows for the design and efficient calibration of the system and controller in unison.
The algorithms that you develop will be applied to wind turbines, both in simulation and experimentally on lab-scale wind tunnel set-ups. The application area is highly relevant, as wind turbines see a rapid increase in system complexity. Next-generation large-scale wind turbines are becoming ever larger with increasing performance demands to satisfy the net-zero emission targets. This size increase leads to greater complexity through dynamic interactions. The exponential growth of data from wind turbines motivates the development and application of novel co-design techniques to achieve next-level performance, ultimately lowering the costs of renewable energy.
The PhD position is in collaboration with Jens Kober, http://www.jenskober.de/
Want to know more? See link below. Also feel free to contact me, Sebastiaan Mulders (S.P.Mulders@tudelft.nl).
https://www.tudelft.nl/over-tu-delft/werken-bij-tu-delft/vacatures/details/?nPostingId=3881&nPostingTargetId=10902
#MACHINELEARNING #CONTROL #LEARNING #RENEWABLEENERGY #WINDTURBINES #TUDELFT
