Spatio-temporal data-driven modelling, simulation and testing of thermal-fluid systems

This research introduces an innovative model structure that helps us better predict, design and understand how heat and fluids behave in various systems. By using advanced techniques like ma- chine learning and computer simulations, we can analyse and predict the behaviour of complex systems such as wind turbines, drones, and combustion processes.
What sets our modelling framework apart is its ability to combine different types of data. We don't rely solely on real-world measurements; we also incorporate data obtained from computer
simulations. This integration of experimental and computational data allows us to quantify uncertainties and optimize system performance even under uncertain conditions.
The implications of this research extend to a wide range of fields that rely on thermal-fluid processes (and beyond). From energy systems and environmental engineering to aerospace applications and related domains, our integrated modelling framework offers valuable insights. It proves particularly beneficial for thermal-fluid systems like wind turbines, drones, plasma-assisted combustion, district heating networks, combined heat and power (CHP) systems, particle-laden flows, and emissions of particulate matter.