this work, a procedure for identifying the parameters of the vector D’Aloia–Di Francesco–De Santis (D-D-D) hysteresis model is presented. The identification task is formulated as an optimization problem based on the discrepancy between measured and simulated hysteresis loops. The comparison is performed through a dimensionless arc-lengthbased parametrization, which makes the geometrical coordinate independent of the physical units used for H and B, while a circular-shift-invariant comparison removes the dependence on the arbitrary starting point of the measured loop. In the proposed reduced formulation, only the shape parameters of the D-D-D model are optimized, whereas the amplitude coefficients are determined directly from the maximum magnetization extracted from the measured loop. A genetic algorithm is employed to minimize a composite objective function including magnetic-flux-density error, geometrical loop error, and hysteresis loop area error. The proposed approach is validated on TEAM Benchmark 32 data and on additional hysteresis loops from the literature.
A Parameter Identification Procedure for the D-D-D Vector Hysteresis Model
De Santis, Valerio
2026-01-01
Abstract
this work, a procedure for identifying the parameters of the vector D’Aloia–Di Francesco–De Santis (D-D-D) hysteresis model is presented. The identification task is formulated as an optimization problem based on the discrepancy between measured and simulated hysteresis loops. The comparison is performed through a dimensionless arc-lengthbased parametrization, which makes the geometrical coordinate independent of the physical units used for H and B, while a circular-shift-invariant comparison removes the dependence on the arbitrary starting point of the measured loop. In the proposed reduced formulation, only the shape parameters of the D-D-D model are optimized, whereas the amplitude coefficients are determined directly from the maximum magnetization extracted from the measured loop. A genetic algorithm is employed to minimize a composite objective function including magnetic-flux-density error, geometrical loop error, and hysteresis loop area error. The proposed approach is validated on TEAM Benchmark 32 data and on additional hysteresis loops from the literature.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


