Title :
Computation of structural system properties, integer matrices, and Bipartite Graphs
Author :
Sagianos, E. ; Karcanias, N.
Author_Institution :
Control Eng. Res. Centre, City Univ., London, UK
Abstract :
The computation of the McMillan degree and structure at infinity for large scale uncertain models of the type known as Structural Transfer Functions (STF) have been shown to be equivalent to determining the maximal weight of associated integer matrices. These problems are equivalent to the class of “Optimal Assignment Problems” developed for the family of resource allocation studies. A new algorithm that exploits notions of “reduceness” of integer matrices and the notions of Bipartite Graphs associated with them is presented here. The complexity of the algorithm is examined and its performance with respect to known optimal assignment methods is considered in terms of features and demonstrated by examples.
Keywords :
computational complexity; graph theory; matrix algebra; transfer functions; McMillan degree; McMillan structure; STF; algorithm complexity; bipartite graphs; integer matrices; large scale uncertain models; optimal assignment methods; resource allocation; structural transfer functions; Bipartite graph; Complexity theory; Cost accounting; Matrix decomposition; Poles and zeros; Polynomials;
Conference_Titel :
Control Conference (ECC), 2007 European
Conference_Location :
Kos
Print_ISBN :
978-3-9524173-8-6