A Clustered Dynamic Model for a class of LInear Autonomous Systems Unsing Simple Enumerative Sorting

Simple, computationally efficient, enumerative techniques are introduced for establishing a clustered semidecoupled dynamic model for a class of autonomous linear systems exemplified by the electromechanical dynamics of the electric power system. Such a model consists of intercluster (usually slow) dynamics and intracluster dynamics (usually fast) which are respectively weakly coupled. The process starts by selecting the point in frequency where intracluster and intercluster dynamics tend to separate. The sorting technique then identifies the number of clusters, the width of separation in frequency, the specifics of the dynamic components and the coupling between them. No matrix manipulations like finding eigenvalues or inverses are required.