A comparative evaluation of semi-active control algorithms for real-time seismic protection of buildings via magnetorheological fluid dampers

Semi-active vibration control is considered a powerful method in reducing the dynamic responses of buildings by using additional smart damping devices. In this study, magnetorheological (MR) dampers have been proposed as one of the semi-active control devices to mitigate the structural vibrations and improve the seismic performance of the structures. The performance of the MR dampers strongly depends on implemented controllers. Hence, the main purpose of this paper is to evaluate the efficiency of several semi-active control algorithms related to MR dampers for seismic control of civil building structures. A 5-story test structure is manufactured, and an MR damper is installed between the ground and the first floor. The performance of the semi-active control approach is experimentally evaluated on a shaking table under historical earthquake records. A neural network-based modeling approach is adopted in the inverse MR damper model for the current control. Three different control algorithms, namely Proportional-Integral-Derivative (PID), Sliding Mode (SMC) and Energy-based controller (EBC), are applied to the system in real-time. The shaking tests are also carried out on the structures with different natural frequencies by increasing the number of stories without changing the geometry and material properties of the 5-story building model. The results indicate that the SMC controller is the most effective control algorithm among all controllers in reducing the base shear force by 51%.