Model order reduction of large-scale state-space models in fusion machines via Krylov methods

This work presents a robust technique, based on the Krylov subspace method, for the reduction of large-scale state-space models arising in many electromagnetic problems in fusion machines. The proposed approach aims at reducing the number of states of the system and lowering the computational effort, with a negligible loss of accuracy in the numerical solution. It is built on the Arnoldi algorithm, which allows to avoid numerical instabilities when computing the reduced model, and exploits both input/output Krylov methods. In the full paper a detail performance study will be presented on an ITER-like machine.