Model order reducibility of nonlinear electro-quasistatic problems

Abstract

PurposeThis paper aims to develop an automated domain decomposition strategy that is based on the presence of nonlinear field grading material, in the context of model order reduction for transient strongly nonlinear electro-quasistatic (EQS) field problems.Design/methodology/approachThe paper provides convincing empirical insights to support the proposed domain decomposition algorithm, a numerical investigation of the performance of the algorithm for different snapshots and model order reduction experiments.FindingsThe proposed method successfully decomposes the computational domain, while the resulting reduced models are highly accurate. Further, the algorithm is computationally efficient and robust, while it can be embedded in black-box model reduction implementations.Originality/valueThis paper fulfills the demand to effectively perform model order reduction for transient strongly nonlinear EQS field problems.