Robustness of semi-active stiffness damper with system uncertainty

A study is conducted to investigate the robustness of various control laws in a semi-active stiffness damper subjected to system uncertainty. Variability in stiffness, damping and sampling time is introduced to demonstrate the effect of system uncertainty on control robustness. Nine control cases of system uncertainty are considered. The control laws used are resetting control, switching control and LQR. They are applied to a five-storey model under two benchmark earthquake excitations. Control law performances and their sensitivity to system uncertainty are investigated through numerical simulation. The results show that the effect of system uncertainty depends on the particular control laws used and the type of earthquake excitation. The estimation error in stiffness affects the performance of the control system more than the estimation error in damping. Moreover, the performance depends on whether the estimation error in stiffness ends up underestimating or overestimating the stiffness. Resetting and switching control algorithms are sensitive to varying sampling time, with LQR displaying more flexibility.