This paper presents a theoretical and experimental study on active control structure excited by seismic loads using artiï¬cial neural network Artiï¬cial neural network is used to calculate the control force based on acceleration of the structure which is obtained by accelerometer measurement. The control mechanism is implemented on the structure using active stiflness and mass based on continuous vibration measurement. The reability of the active control system with artiï¬cial neural network has been tested experimentally using a reduced model of two=storey steel flame excited by base acceleration through a small shaking table. The experimental study shows that the artiï¬cial neural network control method gives satisfactory results for many types of base excitation such as random and El-Centro N-S earthquake accelerations. The neuro control algorithm is simple and reduces consideranbly computational time.
Copyrights © 2003