Experimental study of identification and control of structures using neural network part 2: Control

Experimental verifications of a recently developed active structural contro l method using neural networks are presented in this paper. The experiments were performed on the earthquake simulator at the University of Illinois a t Urbana-Champaign. The test specimen was a 1/4 scale model of a three-stor ey building. The control system consisted of a tendon/pulley system control led by a single hydraulic actuator at the base. The control mechanism was i mplemented through four active pre-tensioned tendons connected to the hydra ulic actuator at the first floor. The structure modelling and system identi fication has been presented in a companion paper. (Earthquake Engng. Struct . Dyn. 28, 995-1018 (1999)). This paper presents the controller design and implementation. Three controllers were developed and designed: two neurocon trollers, one with a single sensor feedback and the other with three sensor feedback, and one optimal controller with acceleration feedback. The exper imental design of the neurocontrollers is accomplished in three steps: syst em identification, multiple emulator neural networks training and finally t he neurocontrollers training with the aid of multiple emulator neural netwo rks. The effectiveness of both neurocontrollers are demonstrated from exper imental results. The robustness and the relative stability are presented an d discussed. The experimental results of the optimal controller performance is presented and assessed. Comparison between the optimal controller and n eurocontrollers is presented and discussed. Copyright (C) 1999 John Wiley & Sons, Ltd.