Phase-space-inspired numerical methods for high-frequency wave scattering: from semiclassical analysis through numerical analysis to open-source software

The project aims to develop and implement advanced algorithms for simulating high-frequency acoustic and electromagnetic waves, enhancing speed and reliability through semiclassical analysis techniques in FreeFEM.

Subsidie
€ 8.725.537
2025

Projectdetails

Introduction

Designing fast and reliable algorithms to numerically simulate the behaviour of high-frequency acoustic and electromagnetic waves is a longstanding open problem in computational mathematics. Furthermore, these waves underpin a plethora of communication and imaging technologies; therefore any progress towards solving this problem will have wide impact.

Project Objectives

The project aims to design, analyse, and implement in open-source software new methods for the numerical solution of high-frequency acoustic and electromagnetic wave scattering problems.

Key Features

These methods will be:

  • Faster
  • More reliable
  • More widely applicable than current state-of-the-art methods

This improvement will be achieved by exploiting techniques from pure mathematics specifically designed to study high-frequency problems. These techniques are from the area of semiclassical analysis and are based on studying functions in phase space (the combination of physical and frequency space).

Current State of Research

Since both numerical analysis and semiclassical analysis of high-frequency wave scattering problems are mature areas of mathematics, it is surprising that these two fields have operated largely in isolation from each other.

Lack of Integration

Essentially none of the deep insights about high-frequency waves from semiclassical analysis have been used in the design of numerical methods for these problems.

Implementation Plan

This project will exploit this untapped potential and implement the resulting methods in the open-source multiphysics finite-element software FreeFEM (https://freefem.org/) in easy-to-use packages.

Target Audience

These packages will be designed for researchers in the applied sciences and industry seeking to simulate the behaviour of high-frequency acoustic and electromagnetic waves.

Collaboration

These goals will be achieved by combining the expertise of the three PIs:

  • JG, an expert in semiclassical analysis
  • ES, an expert in the numerical analysis of high-frequency scattering
  • PT, an expert in scientific computing and the lead developer of FreeFEM.

Financiële details & Tijdlijn

Financiële details

Subsidiebedrag€ 8.725.537
Totale projectbegroting€ 8.725.537

Tijdlijn

Startdatum1-5-2025
Einddatum30-4-2031
Subsidiejaar2025

Partners & Locaties

Projectpartners

  • UNIVERSITY OF BATHpenvoerder
  • UNIVERSITY COLLEGE LONDON
  • CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE CNRS
  • UNIVERSITE PARIS 13
  • INSTITUT NATIONAL DE RECHERCHE EN INFORMATIQUE ET AUTOMATIQUE

Land(en)

United KingdomFrance

Vergelijkbare projecten binnen European Research Council

ERC Synergy ...

Mathematics of Scattering Amplitudes

MaScAmp aims to unify scattering amplitude calculations through innovative algorithms in mathematics and physics, enhancing predictions for particle interactions and advancing theoretical research.

€ 9.986.500
ERC Synergy ...

NEw generation MEthods for numerical SImulationS

The NEMESIS project aims to develop innovative numerical simulators for complex PDE problems in magnetohydrodynamics and geological flows by creating new mathematical tools and an open-source library.

€ 7.818.782
ERC Proof of...

WAVESIM – The next generation electromagnetic wave solver

DEEP VISION aims to develop a commercial software solution that solves Maxwell's equations 2000 times faster and 10 times more accurately, based on innovative research findings.

€ 150.000
ERC Consolid...

Generating Unstable Dynamics in dispersive Hamiltonian fluids

This project seeks to rigorously prove the generation of unstable dynamics in water waves and geophysical fluid equations, focusing on energy cascades, orbital instabilities, and rogue wave formation.

€ 1.444.033
ERC Advanced...

Predictive algorithms for simulating quantum materials

This project aims to develop advanced predictive algorithms for quantum many-body systems by integrating field-theory methods with tensor techniques and machine learning to enhance understanding of quantum materials.

€ 3.499.299