We are pleased to announce the start of the MSCA Postdoctoral Fellowships Project. 

The project was awarded to Dr. Alvaro Moreno Soto, a member of the aerospace engineering research group, and is supervised by Prof. Manuel Soler and Prof. Stefano Discetti. 

Project title: Physics-informed nEuRal networks for SEVERe wEather event prediction 

Acronym: PERSEVERE 

Call for proposals: HORIZON-MSCA-2021-PF-01 

Project ID: 101059984 

Project description:  

The new development of Physics-Informed Neural Networks (PINNs), which incorporate constraints given by physical laws in the training process, as an excellent means to compute fluidic fields and their characteristics, such as velocity and pressure, has opened the door to numerous applications. One of these is the improvement of experiment data, as PINNs can reconstruct, by applying the Navier-Stokes equations as a loss function, the entire fluid domain in areas where experiments are limited by technology. On the other hand, the development of Generative Adversarial Networks (GANs) as robust networks with excellent accuracy but excessive computational costs leaves the door open to investigate new applications where PINNs and GANs can be combined to amplify their strengths and reduce their weaknesses. One such application concerns the prediction of severe weather, where PINNs are useful for calculating the fluidic behaviour of storms approaching a certain location, while GANs can incorporate many additional parameters, such as wind speed, humidity, temperature and electrical content, which can be essential for determining whether a certain location will be affected by severe weather in the next 48 hours. Storm estimation and forecasting is essential for the airline industry, as losses incurred due to air traffic delays and diversions caused by storms have been reported to be over $38.5 billion in the US. The development of a computational architecture that can determine whether severe weather events will occur within the next 48 hours is therefore of crucial importance. There are no models or applications in which the combined strengths of PINN and GAN have been realised, the former to rapidly estimate the fluidic behaviour of a moving storm, the latter to calculate field properties with high accuracy. 

 

We wish Alvaro all the best for this great achievement! 

Funded by the European Union under action HORIZON TMA MSCA Postdoctoral Fellowships – European Fellowships, call HORIZON-MSCA-2021-PF-01 (project number 101059984 with acronym PERSEVERE). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them.