Experimental and numerical studies on pseudo-negative-stiffness control of a base isolated building using magneto-rheological dampers

This article presents a series of experimental and numerical studies on the feasibility and effectiveness of pseudo-negative-stiffness (PNS) control system using magneto-rheological (MR) dampers. First, a MR damper with nominal force capacity of 10 kN was self-developed. A set of experiments was conducted to study the dynamic performance of the MR damper. Then, a PNS control system was established with NI Compact RIO as hardware platform and Labview as software platform. The displacement-based PNS (DPNS) control algorithm was employed to control the MR damper. Dynamic performance tests were performed to investigate the hysteretic behavior of DPNS control force under different excitation conditions and with different control parameters. Next, shaking table tests were conducted on a four-floor steel frame base isolated structure model employing the PNS control system. The seismic responses under uncontrolled case, passive off case, passive on case and DPNS control case were compared. For each control type, the control efficacy was verified for the El Centro, Ji Ji, Kobe_FN and Kobe_FP with increasing peak ground acceleration. The test results show that the DPNS control can further reduce the isolation displacement compared to the passive off case with no accompanying increase of superstructure responses under seismic excitation with different intensities and spectral characteristics. Finally, numerical studies were conducted to compare the effectiveness of DPNS and clipped-optimal control and investigate the effect of time delay of DPNS control. Results indicate that the DPNS control outperforms the clipped-optimal control and the time delay of the DPNS control leads to the adjustment of the DPNS control force at zero displacement, which is beneficial to control efficacy. (paper)