Distributed Control of Large-Scale Structural Systems

This article develops a methodology for the distributed active control of large structural systems. The central idea is to use a set of active members in a structural system, each {locally} controlled. The method relies on an understanding of (1) the peculiar interaction of force feedback and velocity feedback in an active member and (2) the interaction between such active members, as well as that between active and passive members in a structural system. Using a commonly occurring MDOF model of a structural system, we show that the distributed-local-control design is robust. Locating the active members so that all the eigenvalues of the damping and stiffness matrices of the controlled system are assuredly increased leads to several results on where and how active members might be located in such a structure.