Improved equivalent linearized method for double-curve adaptive variable friction pendulum system considering complex friction model based on particle swarm optimization

Various adaptive friction pendulum isolators with complex configurations have been developed to meet the increasing engineering demands. However, studies on the influence of temperature and velocity effects on the mechanical properties of new friction pendulum isolators are limited. In particular, equivalent linearized models that consider such friction-dependent effects are scarce, making traditional equivalent linearization methods unable to meet these requirements. In this study, a double-curved adaptive variable friction pendulum isolator (DCAVFPI) is developed that doubly enhances displacement capacity and maintains a small size while considering temperature-induced coefficient of friction deterioration. Meanwhile, its refined mechanical model expression is derived based on the friction dependency theory, which can simulate the effects of frictional heat and velocity on the friction coefficient, unlike the mechanical model directly based on the Coulomb friction model. The proposed mechanical model was validated through experiments on its mechanical properties. Nonlinear response history analyses determined the potential range of the nonlinear parameters and optimal equivalent damping ratio using particle swarm optimization. Nonlinear regression analysis proposes an equivalent linearized model expression for DCAVFPI based on ideal equivalent damping ratio. In conclusion, when evaluating the peak displacement of DCAVFPI, the accuracy of the proposed method can be stably controlled at -10% to 20% better than the existing equivalent linearization with an accuracy of -20% to -50%. Additionally, the proposed mechanical model accurately simulated the effects of temperature and velocity variations.