A general synthesis of identification and vibration control of building structures under unknown excitations

• Published in: Mechanical systems and signal processing Vol. 143; p. 106803
• Main Authors: Lei, Ying, Lu, Jubin, Huang, Jinshan, Chen, Siyi
• Format: Journal Article
• Language: English
• Published: Berlin Elsevier Ltd 01.09.2020; Elsevier BV
• Subjects: Absolute response; Acceleration measurement; Accelerometers; Active control; Active damping; Algorithms; Data integration; Earthquake dampers; Excitation; Extended Kalman filter with unknown input; Ground motion; Identification; Kalman filters; Linear quadratic Gaussian control; LQG control; Mathematical analysis; Optimal control; Rheological properties; Semiactive damping; Smart structures; Structural health monitoring; Structural identification; Synthesis; Synthesis algorithm; System identification; Vibration control; Vibration monitoring; Absolute response; Extended Kalman filter with unknown input; Vibration control; Synthesis algorithm; Structural identification; LQG control
• ISSN: 0888-3270; 1096-1216
• DOI: 10.1016/j.ymssp.2020.106803

•It provides a more general synthesis algorithm of identification and control.•The absolute responses collected by SHM are adopted under unknown earthquake input.•A GEFK-UI is proposed when measurement equations do not contain unknown excitation.•The drifts in identified results are prevented in real-time by the data fusion method.•The identified state is synthesized in real time with the optimal control strategy. The synthesis of structural health monitoring with vibration control is cost-effective and beneficial for developing smart structures. In the last decade, some synthetic algorithms have been proposed for combining system identification and vibration control. Among them, an approach has been developed recently for the synthesis of identification and vibration control under unknown excitation. However, these approaches are only suitable when structural measurement/observation equations contain the unknown excitations, which limits the application of this approach. In this paper, a general approach is proposed to tackle this limitation problem. A generalized extended Kalman filtering with unknown input (GEKF-UI) is proposed to circumvent the limitations of previous EKF-UI approaches. The proposed GEKF-UI can simultaneously identify structural system and unknown excitation when structural measurement/observation equations contain or do not contain the unknown excitations. Moreover, data fusion of measured acceleration and displacement responses is adopted to prohibit the drifts in the identified structural state and unknown excitations. Then, the identified structural state is synthesized in real time with the linear-quadratic-Gaussian (LQG) control strategy for optimal semi-active optimal control provided by magneto-rheological (MR) dampers. Some numerical examples for the synthesis of identification and vibration control of buildings subject to unknown external forces or unknown earthquake ground motion are adopted to verify the feasibilities of the proposed approach.