ABSTRACT

Electromagnetic solvers are of paramount importance to simulate the behavior of complex microwave and photonic devices and systems, which have many applications in domains as communication, sensing, imaging and computing. However, EM solvers can result very computationally expensive. In a design flow it is needed to carry out design tasks such as optimization and uncertainty quantification, and therefore a high number of EM simulations can be required. The direct use of EM solvers into a design flow then can easily become unfeasible. Advanced data-driven and model-driven modeling techniques (e.g., machine learning, surrogate modeling, parameterized model order reduction) have been proposed for simulation and design of microwave and photonic devices and systems. These modeling techniques can be used to tremendously speed-up the design flow while retaining high accuracy in the obtained results. Multiple applications will be presented where different modeling techniques are used for passive and active electromagnetic devices (e.g., microwave filters, antennas, metasurfaces for (bio)imaging and communication, electronic circuits, etc.) and can be tailored towards optimization and uncertainty quantification. The importance of data collection (sampling) will be highlighted, since it is a very important aspect related to the model accuracy and computational cost of the model training and validation.

SPEAKER CV

Prof. Francesco Ferranti received the B.S. degree in electronic engineering from the Università degli Studi di Palermo, Palermo, Italy, in 2005, the M.S. degree in electronic engineering from the Università degli Studi dell'Aquila, L'Aquila, Italy, in 2007, and the Ph.D. degree in electrical engineering from Ghent University, Ghent, Belgium, in 2011. He is currently a Professor at the Vrije Universiteit Brussel (VUB), Belgium. Francesco is also an Adjunct Professor at Carleton University, Ottawa, Canada and at the Indian Institute of Technology (IIT) Madras, Chennai, India.

Francesco's research activities have mainly focused on advanced data-driven and model-driven modeling (e.g., machine learning, surrogate modeling, parameterized model order reduction) for the analysis, simulation and design of complex dynamical systems, such as microwave filters, antennas and power amplifiers, high-speed interconnects and circuits, metasurfaces/metamaterials, nanophotonics, and smart contact lenses. His research is largely interdisciplinary and it is driven by his broad interest for fundamental scientific problems and their application in challenging domains with a significant societal and economic impact. Adaptive sampling and model generation methods have enriched these research results and can effectively reduce the user interaction and expertise level needed for a real-life use by designers and practitioners.