DEVELOPMENT OF ADVANCED METHODOLOGY FOR GENERATING REDUCED-ORDER MODELS OF LINEAR SUB CIRCUITS

The computational modeling and analysis of dense electronic circuits involves the numerical solution of large-scale or very large scale or sparse, in general leads to a nonlinear systems of deferential-algebraic equations. Frequently, the size of these systems can be reduced model, considerably by replacing the equations corresponding to linear sub circuits by approximate models of much smaller state-space dimension. In this paper, we describe the use of Krylov-subspace methods for generating such reduced-order models of linear sub circuits.