This paper presents an efficient application of Machine Learning (ML) to derive models for accurately predicting the inductance value and mechanical constraints in widely used air-cored inductors in power electronics systems for accelerators. The ML is trained on Finite Elements Analyses (FEA) obtained data. The obtained Artificial Neural Network (ANN) based models are then used in a numerical optimization environment able to efficiently provide optimal solution in terms of speed and accuracy.
Détails
Titre
Using supervised machine learning in power converters design for particle accelerators : application to magnetic components design
Auteur(s)/ trice(s)
Cajander, David (School of Engineering and Architecture (HEIA-FR), HES-SO University of Applied Sciences and Arts Western Switzerland ; LEEPCI Lab., Laval University, Quebec, Canada) Aguglia, Davide (CERN - European Organization for Nuclear Research, Geneva, Switzerland ; LEEPCI Lab., Laval University, Quebec, Canada) Viarouge, Isabelle (Electrotechnologies Selem Inc, Quebec, Canada) Viarouge, Philippe (LEEPCI Lab., Laval University, Quebec, Canada)
Date
2023-05
Publié dans
Proceedings of the IPAC'23 -14th International Particle Accelerator Conference, 7–12 May 2023, Venice, Italy
Pagination
4 p.
Présenté à
IPAC'23 -14th International Particle Accelerator Conference, Venice, Italy, 2023-05-07, 2023-05-12
Type de papier
short paper
Domaine
Ingénierie et Architecture
Ecole
HEIA-FR
Institut
Energy - Institut de recherche appliquée en systèmes énergétiques
