core-loss

CAHR-Net proposes a condition-adaptive hysteresis reconstruction network that improves magnetic core loss modeling by injecting operating conditions into intermediate representations, achieving lower errors with fewer parameters compared to existing methods.

This preprint proposes PI-HNO, a physics-informed hybrid neural operator for transient magnetization prediction in power magnetics, achieving low B-H energy consistency errors with only 4,777 trainable parameters per material model.