An image-based approach for Preisach function calculation

Abstract

Abstract-This paper proposes an image-based calculation of Preisach function. The function is calculated from a series of distinct magnetized domain images. The domain images derive the state transition matrices from the image Helmholtz equation, which is a governing partial differential equation of digital dynamic images. It is described that the state transition matrix of the equation makes it possible to approximate the Preisach function. An intensive verification is carried out using scanning electron microscope domain images of a grain-oriented electrical steel. The magnetization curve reconstructed by means of the state transition matrices is shown to closely reproduce experimental result. Moreover, the state transition matrices reveal the iron loss generating parts on the target specimen.

Computation of Local Magnetization Curve from Visualized Magnetic Domain Dynamics By Bitter Method

“…We have measured entire magnetization characteristics while the magnetic domains such as shown in Fig.1 are visualizing [1][2][3]. Since entire magnetization characteristic of a specimen is an integration of all magnetic charge movements, then pixels having large and small in values on the visualized domain image respectively associate with the positive and negative magnetic flux densities, leading to the pixel representation for magnetization characteristics,…”