About the project
This PhD project explores the use of Physics-Informed Neural Networks (PINNs) to solve environmental flow problems, including the 2D Shallow Water Equations. Combining advanced artificial intelligence (AI) with fluid mechanics, the research aims to develop fast, accurate, and robust simulations for applications like flood modelling and water management.
PINNs combine data-driven machine-learning techniques with the governing physics of fluid systems to create fast, accurate, and computationally efficient models. This innovative approach has the potential to revolutionize how environmental flows are simulated, with applications in flood prediction, engineering design and water resource management.
Building on recent advancements in PINNs and their integration into fluid mechanics, this project will develop state-of-the-art methodologies for modelling complex environmental systems. You will explore and refine these techniques, pushing the boundaries of their application to real-world challenges.
This project is hosted by the Water and Environmental Engineering Research Group, a leader in environmental and computational modelling. You will join a dynamic cohort of researchers working on cutting-edge machine-learning tools for fluid dynamics, fostering collaboration and innovation in this rapidly growing field.
To support this work, You will have access to one of the UK’s fastest supercomputing facilities, enabling high-performance simulations and advanced computational analysis. This project offers an excellent opportunity to contribute to impactful research at the intersection of artificial intelligence and environmental science, shaping solutions for critical global challenges.
