![\"\"](\"https:\/\/fusion.bsc.es\/wp-content\/uploads\/1774980330751-e1789732567885-400x226.jpg\")
<\/p>\n<\/p>\n
Hello everyone! My name is Adam Canarutto<\/strong> and over the past six months, I had the opportunity to carry out my Master’s thesis within the Fusion Group at the Barcelona Supercomputing Center (BSC)<\/a><\/strong>. My work focused on a problem at the intersection of plasma physics, numerical modelling and high-performance computing: the simulation of electromagnetic wave propagation for Ion Cyclotron Resonance Heating (ICRH).<\/p>\n Throughout my project, I had the privilege of being guided by Alejandro Soba<\/strong>, Ezequiel Goldberg<\/strong>, and Hern\u00e1n Domingo<\/strong>, and I am incredibly grateful for their support. From the very beginning, they were always there to guide me, answer my questions, and help me overcome any difficulties along the way, while creating a friendly and enjoyable working environment.<\/p>\n <\/p>\n Understanding wave propagation in a magnetised plasma is a challenging numerical problem. The plasma response is anisotropic and depends on parameters such as the magnetic field, density, particle species and wave frequency. At the same time, realistic fusion simulations require large computational resources and therefore efficient parallel numerical methods.<\/p>\n My work was carried out using Alya<\/strong><\/a>, BSC’s multiphysics simulation code designed for large-scale parallel computing, and specifically its EMWAVE<\/strong> module for electromagnetic-wave simulations based on the finite element method. The plasma description used in the project relies on the cold-plasma approximation, which provides a useful foundation for studying wave propagation before moving towards more complete plasma models.<\/p>\n A significant part of the thesis involved extending and testing EMWAVE’s numerical capabilities. I implemented and verified quadratic triangular and quadrilateral finite elements<\/strong>, introduced electrical conductivity<\/strong> into the electromagnetic formulation, and extended the available outputs to include the magnetic flux density B<\/strong>, magnetic field H<\/strong>, and current density J<\/strong>, in addition to the electric field. This required working directly with Alya’s Fortran code, understanding its finite-element formulation, debugging the implementation and designing numerical tests to verify the new features.<\/p>\n Verification was a central part of the project.<\/strong> Numerical results were first tested using controlled electromagnetic problems and then compared with results from ERMES<\/strong>, an independent electromagnetic solver. The agreement between the two codes for the electromagnetic fields and derived quantities provided confidence in the implementation and highlighted the importance of independent verification when developing scientific software.<\/p>\n I also moved from local tests to large-scale simulations on MareNostrum 5<\/strong><\/a>, BSC’s supercomputer. Parallel scalability studies allowed me to investigate how the solver behaves as the computational workload is distributed across increasing numbers of processors. This provided practical experience with parallel efficiency, workload distribution, memory requirements and the challenges of running large HPC simulations. It also demonstrated that, for realistic fusion applications, reproducing the correct physics is only part of the problem: numerical methods must also exploit modern supercomputing architectures efficiently.<\/p>\n In the final stage of the thesis, I explored coupling Alya with DAKOTA<\/strong><\/a>, a toolkit for parameter studies, optimisation and uncertainty quantification. I used this coupling to automate frequency-sensitivity studies and examine how the electromagnetic response changes with excitation frequency. These studies also showed the importance of ensuring sufficient numerical resolution before drawing physical conclusions from parameter scans. A more detailed explanations of the results can be found in my Master’s Thesis<\/a>.<\/p>\n The thesis concluded with its presentation in September 2026<\/strong>, but the work forms part of a longer-term development of EMWAVE. Future directions include more realistic geometries and, eventually, moving beyond the cold-plasma approximation towards hot-plasma models<\/strong> capable of describing the wave-particle interactions responsible for plasma heating.<\/p>\n Beyond the technical developments, the internship gave me the opportunity to experience scientific research within a large computational organisation. Coming from a background in plasma and fusion physics, I had to connect physical concepts with finite-element methods, scientific software development and HPC. Working on a large research code also taught me the importance of careful testing, integration and reproducibility in a parallel computing environment.<\/p>\n For me, the project brought together three areas I would like to continue exploring: fusion plasma physics, numerical modelling and high-performance scientific computing<\/strong>. It was also a memorable opportunity to spend six months in Barcelona, surrounded by an international research community and supported by colleagues who made the experience valuable both scientifically and personally. I am very grateful to my supervisors and to the BSC Fusion Group for the opportunity and for everything I learned during these six months. Finally, I would also like to acknowledge the financial support provided by FuseNet and the Erasmus+ programme, which contributed to making this research stay at BSC possible.<\/p>\n","protected":false},"excerpt":{"rendered":" Hello everyone! My name is Adam Canarutto and over the past six months, I had the opportunity to carry out my Master’s thesis within the Fusion Group at the Barcelona Supercomputing Center (BSC). My work focused on a problem at the intersection of plasma physics, numerical modelling and high-performance computing: the simulation of electromagnetic wave … Read more<\/a><\/p>\n","protected":false},"author":47,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":"","_members_access_role":[],"_members_access_error":""},"categories":[9,1],"tags":[],"class_list":["post-93219","post","type-post","status-publish","format-standard","hentry","category-group","category-news"],"_links":{"self":[{"href":"https:\/\/fusion.bsc.es\/index.php\/wp-json\/wp\/v2\/posts\/93219","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/fusion.bsc.es\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/fusion.bsc.es\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/fusion.bsc.es\/index.php\/wp-json\/wp\/v2\/users\/47"}],"replies":[{"embeddable":true,"href":"https:\/\/fusion.bsc.es\/index.php\/wp-json\/wp\/v2\/comments?post=93219"}],"version-history":[{"count":14,"href":"https:\/\/fusion.bsc.es\/index.php\/wp-json\/wp\/v2\/posts\/93219\/revisions"}],"predecessor-version":[{"id":93274,"href":"https:\/\/fusion.bsc.es\/index.php\/wp-json\/wp\/v2\/posts\/93219\/revisions\/93274"}],"wp:attachment":[{"href":"https:\/\/fusion.bsc.es\/index.php\/wp-json\/wp\/v2\/media?parent=93219"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/fusion.bsc.es\/index.php\/wp-json\/wp\/v2\/categories?post=93219"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/fusion.bsc.es\/index.php\/wp-json\/wp\/v2\/tags?post=93219"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"\"](\"https:\/\/fusion.bsc.es\/wp-content\/uploads\/scaling_test_tol1e6-1024x634.png\")