A TARGET APPROXIMATION METHOD FOR SPILLOVER EFFECT MINIMIZATION

Control and observation spillover minimization is a major concern in the control design of distributed-parameter systems when controller dimensions are small due to limitations on on-board computers and limited availability of actuators and sensors. The spillovers couple the controlled and uncontrolled modes and shift the system's closed-loop poles away from their desired locations. Thus, a desirable controller is the one which simultaneously provides enough control force and minimizes spillover effect. In this paper, we propose a target approximation method which can achieve this goal. The method not only optimizes the locations of actuators and sensors, but also utilizes feedback and observer gain structures to minimize the spillover effect. The design method has two stages. The first stage generates an idealized, totally distributed closed-loop system as a target which meets the posed performance specifications and has no spillovers. The second stage designs an actual controller including the locations of actuators and sensors. The design is carried out in such a way that the closed-loop system approximates the performance of the target. At the end of the paper, this simple novel design procedure is shown superior to the other approaches such as H and other classical methods in the stabilizing controller design of a widely studied simple flexible beam.

A TARGET APPROXIMATION METHOD FOR SPILLOVER EFFECT MINIMIZATION Academic Article