Semi-active control of three-story benchmark structure using LQG algorithm with a fuzzy-genetic system

Damages of natural disasters such as earthquakes has led engineering sciences to research and expand the use of earthquake-resistant systems. Semi-active control systems have been considered due to their ability to change the mechanical properties of their system and the use of limited energy sources. On the other hand, taking advantage of soft computing and fuzzy inference systems due to the compensation for existing uncertainties, considerably helps to reduce structure’s responses during earthquakes. In this paper, a method for controlling a three-story nonlinear benchmark structure equipped with an MR magnetic damper is presented. The required voltage of the damper is obtained from the design of a LQG linear control system with a fuzzy inference system. In this system, the linear control force of LQG and the force of the previous step of the damper enter the Takaki-Sugeno fuzzy system and the desired voltage of the damper is obtained. In order to optimize the controller performance, the rules of the fuzzy inference system are taught by a genetic algorithm. Modeling of the damper is based on the Bouc-wen model. The results of the evaluation criteria and time history charts show the high efficiency of the proposed controller in reducing seismic responses against other proposed systems