Seminal
The magazine and atlas of EU-funded researchA new publication from the web and AI team at Scott Charnley

Atlas / Austria

Research and Innovation Action · Running

MaMMoS MAgnetic Multiscale MOdelling Suite

Coordinator
Universitat Fur Weiterbildung Krems, Austria
Runs
1 January 2024 to 31 December 2027
Status
Running
EU funding
€6.8m
Total cost
€6.8m
Partners
8 organisations in 4 countries
Type
Research and Innovation Action
Grant agreement
101135546

What the project is doing

In the project's own words, from its summary on CORDIS.

Magnetic materials are essential for many applications in energy, information, and communication technologies. However, the complex phenomena at different length and time scales often limit the development of new magnetic materials and devices. The goal of this project is to develop a magnetic multiscale modeling suite that will allow the design and optimisation of magnetic materials and devices based on multiscale modelling, characterisation, and numerical optimisation.

To achieve interoperability between software and analysis tools, we will establish a domain ontology for magnetic materials. We will collaborate with EU magnet industry to create standards for linking simulation software for magnetic materials from first principles simulations and micromagnetics to device level simulators. MaMMoS will use artificial intelligence (AI) to fuse modeling and characterization data. AI methods will identify and correct systematic errors in the simulation data, enabling more accurate predictions.

Moreover, AI models can fill gaps where measurements are not available. AI models can also serve as a surrogate in multi-objective optimisation. Optimisation will guide further experiments or simulations, reducing the development time. In MaMMoS, we will apply this approach to speed up the development of permanent magnets with reduced critical elements for electric machines and to optimise the layout of magnetic field sensors for high linearity range.

The MaMMoS software will be validated against benchmarks defined according to the industrial requirements for electric machine and sensor design. The multiscale magnetic materials modeling suite will be made open source to enable easy access to high-end simulation tools. Interoperability will facilitate data sharing and reuse among researchers and industries. Interpretable machine learning will reveal insights into the physics and chemistry of magnetic materials and guide the discovery of new materials for the European green deal.

Who is involved

OrganisationCityCountryRole
Universitat Fur Weiterbildung KremsKremsAustriaCoordinator
Leibniz Institut Fur Festkorper Und Werkstofforschung Dresden eVDresdenGermanyParticipant
Max-Planck-Gesellschaft Zur Forderung Der Wissenschaften eVMunchenGermanyParticipant
Robert Bosch Car Multimedia GmbHHildesheimGermanyParticipant
Siemens AktiengesellschaftMunchenGermanyParticipant
Centre National De La Recherche Scientifique CNRSParisFranceParticipant
Universite Grenoble AlpesGrenobleFranceThird party
Uppsala UniversitetUppsalaSwedenParticipant

Source: CORDIS, European Commission. Data of 5 Nov 2026, reused under the Commission's reuse policy. View on CORDIS · Grant DOI