Support vector machine based semi‐active control of structures: a new control strategy

The present paper presents the support vector machine (SVM)-based semi-active control algorithm used for designing general dampers for multistorey structures under earthquakes. First, the linear quadratic regulator (LQR) controller for the numerical model of a multistorey structure formulated using the dynamic dense method is obtained by using the classic LQR control theory. Then, a SVM model is designed and trained to emulate the performance of the LQR controller. Likewise, this SVM model comprises the observers and controllers of the control system. Finally, in accordance with the features of general semi-active dampers, a SVM-based semi-active control strategy is put forward. More specifically, an online auto-feedback semi-active control strategy is developed and then realized by resorting to SVM. In order to numerically verify the control effectiveness of the present control strategy, the time history analysis has been implemented to a structure with general dampers designed by the SVM-based semi-active control algorithm. In numerical simulations, four seismic waves including the El Centro, Hachinohe and Kobe waves, as well as the Shanghai artificial wave, whose peak ground accelerations are all scaled to 0.1 g, are taken into consideration. Comparative results demonstrate that general semi-active dampers designed using the SVM-based semi-active control algorithm is capable of providing higher level of response reduction. Copyright © 2009 John Wiley & Sons, Ltd.

[1]  Francis Eng Hock Tay,et al.  Modified support vector machines in financial time series forecasting , 2002, Neurocomputing.

[2]  Hoon Sohn,et al.  Damage diagnosis under environmental and operational variations using unsupervised support vector machine , 2009 .

[3]  Lai Ming,et al.  Nonlinear structural response prediction based on support vector machines , 2008 .

[4]  Junsheng Cheng,et al.  Application of support vector regression machines to the processing of end effects of Hilbert Huang transform , 2007 .

[5]  Lijuan Cao,et al.  Support vector machines experts for time series forecasting , 2003, Neurocomputing.

[6]  Rong Chen,et al.  Online weighted LS-SVM for hysteretic structural system identification , 2006 .

[7]  Yutaka Inoue,et al.  Overview of the application of active/semiactive control to building structures in Japan , 2001 .

[8]  B. F. Spencer,et al.  STATE OF THE ART OF STRUCTURAL CONTROL , 2003 .

[9]  Michael C. Constantinou,et al.  Semi-active control systems for seismic protection of structures: a state-of-the-art review , 1999 .

[10]  Jian Zhang,et al.  Novel support vector regression for structural system identification , 2007 .

[11]  Nikola Pavesic,et al.  Training RBF networks with selective backpropagation , 2004, Neurocomputing.

[12]  Celestino Ordóñez,et al.  Creating a quality map of a slate deposit using support vector machines , 2007 .

[13]  M. Rabins,et al.  Semi‐Active versus Passive or Active Tuned Mass Dampers for Structural Control , 1983 .