The use of neural networks in describing magnetisation phenomena
暂无分享,去创建一个
Abstract The application of multilayer feed-forward neural network architecture is presented for mapping of the B — H plane with a major loop and first order transition curves, and for approximation of dynamic hysteresis of soft magnetic specimens. The neural network correctly models hysteresis characteristics and constitutes a viable alternative to analytical approaches.
[1] Mayergoyz,et al. Mathematical models of hysteresis. , 1986, Physical review letters.
[2] Numerical implementation and testing of new vector isotropic Preisach-type models , 1994 .