Model Order Reduction for High Dimensional Linear Systems based on Rank-1 Incremental Proper Orthogonal Decomposition

This work considers a modified incremental proper orthogonal decomposition (iPOD) method and applications to model order reduction (MOR) of linear evolutionary distributed parameter systems. A recursive matrix transformation approach is proposed to obtain reduced order models from high-dimensional systems with less computational cost than the non-recursive case. A detailed analysis of the computational complexity is carried out to compare different numerical procedures. Simulation results based on the heat conduction process in a two dimensional container validate the effectiveness of the proposed method.