Document Type : Article
Authors
Dept. of Civil and Environmental Engineering Amirkabir University of Technology
Abstract
Nowadays, basic isolation is an extensive technology that is applied in many countries, and the construction basics of their different types are well known. It seems that inventors have an inevitable interest in this concept and, so,
they propose innovative base isolation systems each year. Many of these systems are not applicable, and may be, in some cases, perilous, but their numbers are vastly growing, year by year. In recent years, the damage to well designed structures, due to ground motion, has attracted the attention of engineers to near-field sources of motion and their effects on building performance. Triple Friction Pendulum Bearings (TFPBs), as an adaptive seismic isolator, with
different stiffness and damping properties, can guarantee the seismic performance of an upper structure for long periods and amplitudes of near-field ground motion. TFPBs are made of multiple concave surfaces with different
friction coefficients. The magnitude of displacement will cause a transition of sliding on surfaces, and produces appropriate damping and stiffness. Hence, optimization of effective design parameters for an objective performance is
critical.
First, the behavior of TFPBs is investigated to identify its dominant design parameters on the response of structures, such as story drift, roof acceleration and displacement of isolated levels. Then, a specific numerical optimization method, based on Genetic Algorithms, has been applied to determine the optimum value of these parameters to achieve the minimum response of the structure. In this process, near-field ground motion with different
characteristics, such as a pulse period, at different hazard levels has been used.
As the results of GA analysis shows, it was realized that the optimum design parameters have significantly different optimum intervals for different target responses. So, different response targets were combined linearly, to make a new fitness function. The partnership coefficients of each single objective function can be chosen by the desire of the designer.
The superstructure was assumed to have rigid behavior, so, the vibration period of the structure adheres to the TFPB period. Thus, optimum design parameters can be used for different types of superstructure with the same behavior.
Keywords
)