Metaheuristic Optimization of Controller Design for Active Mass Dampers

Metaheuristic Optimization of Controller Design for Active Mass Dampers

The linear-quadratic regulator (LQR) solution has been applied to active structural control theory and has been extensively investigated over past decades. State feedback gains can be obtained by minimizing the cost function that contains weighted states and control inputs; however, the appropriate selection of weights mostly depends on trial and error combined with engineering experiences. In this study, a novel metaheuristic optimization algorithm named symbiotic organisms search (SOS) was applied to optimize the LQR weighting matrices with respect to different objective functions. In addition, three objective functions were proposed, which contain the root-mean-square of modal acceleration, the peak absolute modal acceleration, and the square root of the sum of squares of the modal acceleration. A 10-story shear building with an active mass damper installed on the top floor was adopted to validate the effectiveness of the proposed method.