Shake table tests of a three-storey structure with an inerter-based-hysteretic-damper

Inerters have received substantial interest from the earthquake engineering community, because of their potential to improve the seismic isolation of structures. An inerter generates a force proportional to the relative acceleration across the two terminals of the device; this is in contrast to a pure mass which has one terminal and generates a force proportional to absolute acceleration. In practice the realization of an ideal inerter is very difficult, because of the inevitable influence of parasitic characteristics such as mass, damping, and (depending on the design) stiffness that are not considered in a lumped-parameter representation. The present study explores the potential of an alternative inerter design that minimizes the friction damping in the device, thereby seeking to reduce the parasitic damping within the vibration isolation system. The inerter is comprised of a flywheel and linear guideway mechanism. The center of the flywheel acts as one of the inerter terminals and at the same time is also the center of its rotation. The other terminal of the inerter is on the linear slider. Thus, the inertance generated is proportional to the relative acceleration between these two points. Unlike a traditional ball-screw inerter, this new design allows the inertance to be easily adjusted by changing the distance between the two terminals. The inerter is connected in series to a hysteretic material damper to form an inerter-based-hysteretic damper (IBhD). Two linear mathematical models are proposed for the IBhD device, namely a tuned-inerter-hysteretic damper (TIhD) model and a tuned-mass-hysteretic-damper-inerter (TMhDI). Finally, shake table experiments of a 3 storey structure with a grounded IBhD were undertaken. Particular emphasis is given to the comparison between the two models with the experimental results. It was found that the TMhDI model gives a more accurate result for this IBhD device. Furthermore, the IBhD was proven to be able to reduce the structural response due to earthquakes. Specifically, the peak response acceleration of the top storey is reduced by 38.4% for Northridge and 20.3% for El Centro earthquake. Moreover, the root-mean-square (RMS) of the acceleration response is also reduced by 51.3% and 46.1% for Northridge and El Centro earthquakes respectively.