NEW ALGORITHM FOR ACTIVE STRUCTURAL CONTROL

A simple effective algorithm for active structural control is proposed . It applies the control force at every time step, minimizing the syst em energy that is carried over from one time step to the next. The con trol force needed for the next time step is computed by a simple close d form solution at the current time utilizing the available informatio n. Therefore, the proposed control algorithm is free from time delay p roblems. The control algorithm is derived from the theory of single-de gree-of-freedom (SDF) systems and is extended to multiple-degree-of-fr eedom (MDF) systems by making use of the modal synthesis. Numerical ex amples are used to demonstrate the effectiveness of the proposed contr ol law. Practical problems in active control, such as the spillover ef fect, the actuator's capacity limit, and the problems associated with the limited number of sensors are discussed. Various strategies to cop e with these problems, in particular, an artificial neural network as the state estimator, are proposed in the paper. The results of the exa mples show that the proposed control algorithm can reduce the structur al response by one order of magnitude and has the ability to reduce th e peak that occurs during the first few cycles of the time history, an ability that linear control laws lack. Unlike most control algorithms that deal with the state space of the system that usually involves co mplex-valued eigenvalues and eigenvectors, the proposed control algori thm deals with the real-valued normal modes of the system.