An algorithm for model reduction of large-scale systems via equality constrained least squares

摘要

A new SVD-Krylov based method is proposed, which is equivalent to compute an equality constrained least-squares problem. The reduced model matches the first r + i Markov parameters of the full order model. Based on the rational equality constrained least-squares method, an iterative algorithm for ‛2 model reduction is prensented. Moreover, both algorithms of IRKA ( An Iterative Rational Krylov Algorithm) [10] and ISRK(An iterative SVD-rational Krylov based model reduction method) [9] turns out to be two special cases of the proposed algorithm. The algorithm is numerically effective and suited for large-scale problem, which can be verified in the numerical examples.