Design of observers for large scale singular systems

This paper is concerned with the problem of designing minimal-order observers for reconstructing the state of large-scale singular systems. Based on the generalization of block Hessenberg form, a block decomposition technique is presented to construct a minimal order observer for such systems. It is shown that the system in this design coordinate along with certain assumed block forms for the observer matrices enable the generalized observer constraint equation to be decomposed and solved effectively. A distinct point of the proposed algorithm is that the solution procedure can be tailored to parallel implementation and is therefore fast.