There are increasing needs to obtain physically realistic material constants for finite element analysis (FEA) of industrial equipment such as large motors and transformers, where vibration reduction is crucial. However, laminated iron cores commonly used in these systems consist of thin stacked steel sheets, exhibiting structural anisotropy that complicates experimental identification. In practice, only a few low-frequency vibration modes can be measured, making the inverse problem underdetermined and resulting in non-unique solutions. This study addresses the challenge by explicitly incorporating in-plane isotropy—an inherent structural property of laminated cores—into the Newton-based optimization process. Two methods are proposed: one applies an averaging correction to isotropic variable pairs, and the other adaptively tunes scaling matrices using particle swarm optimization. Numerical results show that both methods achieve high accuracy while maintaining convergence efficiency, enabling the derivation of material constants that are not only optimal but also physically plausible under limited measurement conditions.
FUKUHARA et al. (Wed,) studied this question.