A Kalman Filter-based model predictive control scheme with actuator saturation consideration for active control of a nine-story benchmark SAC building

Model predictive control is one of the optimal control methods for systems based on their behavior on future horizons. One of the salient features of this control method is the optimal consideration of control constraints in the system control process. Accurate states are required to perform an optimal control performance with this technique. On the other hand, state sensors are unable to provide accurate states due to the uncertainty in their structures. This shortcoming causes problems in the optimal control process. In this study, a discrete-time Kalman filter-based model predictive control scheme with actuator saturation consideration is presented. As a state estimator, the Kalman filter is able to provide more accurate states. On the other hand, the application of performance constraints in the control process causes the saturation of the actuators to be optimally regarded. In the present study, to investigate the effectiveness of the proposed control method in reducing seismic responses, a nine-story benchmark steel structure (SAC) under seismic excitation is utilized. Then, the results obtained from the proposed method considering three different control force constraint scenarios are compared with the results of the uncontrolled case. The results of numerical studies demonstrate the appropriate performance of the proposed control process in reducing seismic responses. Also, the replacement of low-capacity actuators with high-capacity ones, while making the control process more economical, do not significantly change the other responses quantity. For example, the highest change in the Drift Ratio Index (J1) for the controlled case with control force constraints to the controlled case without control force constraints for the Elcentro earthquake is by up to 4%, while the same conditions for the maximum control force index (J12) is 78%.