Harald Özelt

Harald Özelt is a Researcher at the Center for Modelling and Simulation within the University for Continuing Education Krems. His work focuses on computational magnet design, micromagnetics, and machine learning applications in materials science. He leads projects such as 'Data Driven Magnet Design through Combinatorial Synthesis and Micromagnetic Graph Networks' (FWF-funded) and 'Design of Nanocomposite Magnets by Machine Learning' (FWF-funded). His research emphasizes the development of advanced simulation techniques and optimization strategies for magnetic materials, particularly permanent magnets and composite systems.

• Key Projects:
  • Magnetic Multiscale Modelling Suite (EU-funded, PI: Thomas Schrefl)
  • Data Driven Magnet Design (FWF-funded, PI: Özelt)
  • Design of Nanocomposite Magnets (FWF-funded, PI: Özelt)

Özelt's research interests include the simulation of magnetic materials' microstructure-property relationships, defect manipulation for enhanced coercivity, and the integration of physics-informed machine learning into computational frameworks. He has authored numerous peer-reviewed articles in journals like Journal of Magnetism and Magnetic Materials and Frontiers in Materials, focusing on topics such as micromagnetic modeling, hysteresis optimization, and magnet design.

His lectures and conference presentations highlight contributions to inverse design methodologies, multiscaling strategies in magnet design, and the use of neural networks for predicting magnetic properties. He collaborates internationally on projects addressing sustainable materials and rare-earth reduction in permanent magnets.