Machine Learning the Central Magnetic Flux Density of Superconducting Solenoids
Yun Zhang, Xiaojie Xu
Abstract
The central magnetic flux density is usually simulated via finite element methods that require a significant amount of inputs and computation resources. We develop the Gaussian process regression (GPR) model to shed light on the statistical relationship between structural descriptors of the iron yoke and the central magnetic flux density for superconducting solenoids. The model is highly stable and accurate, contributing to fast and robust estimations of the central magnetic flux density.
Topics & Concepts
Yoke (aeronautics)Magnetic fluxFlux (metallurgy)SuperconductivityFinite element methodComputationPhysicsMagnetCondensed matter physicsStatistical physicsMagnetic fieldMaterials scienceMechanicsComputer scienceQuantum mechanicsAlgorithmMetallurgyAerodynamicsThermodynamicsFlight control surfacesSuperconducting Materials and ApplicationsMagnetic confinement fusion researchNuclear Physics and Applications