Seismic optimum design of triple friction pendulum bearing subjected to near-fault pulse-like ground motions

Abstract

Triple Friction Pendulum Bearing (TFPB) as a novel seismic isolator, provides different combinations of stiffness and damping during its course of motion. The adaptive behavior of TFPBs is one of the practical solutions for unsuitable performance of seismic isolation systems under near-fault ground motions. Selecting the TFPB's design variables (curvature radii, friction coefficients and displacement capacity of sliding surfaces) is complicated process while finding the optimized combination of these variables depends on input ground motion characteristics and seismic performance objectives of the superstructure. Here first, comprehensive nonlinear dynamic analyses are performed to identify influence of the design variables on superstructure response (roof acceleration and displacement of isolated level). Next, a specific numerical optimization method based on Genetic Algorithms (GA) is applied to determine the optimum values of the design variables that minimize superstructure demands. In this process, near-fault ground motions are employed with ranges of pulse periods and hazard levels as input excitations. According to GA results, the derived optimum design variables of TFPB have significantly distinct intervals for different target responses such as story drift and TFPB displacement. Therefore response targets (single objective functions) are combined to make a new fitness function. The proposed optimization method for determining design variables and design intervals can be used for investigating many other types of superstructures with similar behaviors.