Coprime Factorization Approach to Robust Stabilization of the Control Structures Interaction Evolutionary Model

摘要

Stabilization in the presence of uncertainties is a fundamental requirement in the design of feedback compensators for flexible structures. The search for the largest neighborhood around a given design plant for which a single feedback controller produces dosed-loop stability can be formulated as an H(sub infinity) control problem. It has been shown that the use of normalized coprime factor plant descriptions, where the plant perturbations are defined as additive modifications to the coprime factors, leads to a closed-form expression for the maximal-neighborhood boundary allowing optimal and suboptimal H(sub infinity) compensators to be computed directly without the usual gamma-iteration. This paper describes an application of normalized coprime factor stabilization theory to the computation of robustly stable compensators for the NASA Control Structures Interaction Evolutionary Model. Results indicate that the suboptimal version of the theory has the potential of providing low authority compensators that are robustly stable for significant regions of variations in design model parameters and additive unmodeled dynamics.