Identifying linear reduced-order models for systems with arbitrary initial conditions using Prony signal analysis

A method of identifying reduced-order linear models for systems operating in the neighborhood of an equilibrium point is presented. The method is based on Prony signal analysis, which has recently received considerable attention in the study of power system electromechanical oscillations. Prior to the application of the input test signal, the system can be in a transient state. The system input test signal is piecewise continuous and allows several Prony analyses to be performed during a transient, with each analysis conducted between input discontinuities. Results of these Prony analyses can be combined in various ways to obtain system eigenvalues, transfer-function residues, and initial condition residues. Two examples are given to illustrate the use of the method.<
>