Research On Multi-objective Optimization Of Base Isolation System

With the popularization of shock isolation technology, performance optimization becomesan important aspect in base isolation research. In this paper, the model of base isolation syste-m is analyzed in Simulink environment , a Matlab tool and multi-objective optimization ofseismic responses in rubber bearing base isolation system, sliding base isolation system andresilient-friction base isolation system is performed. The seismic responses include the maximu-m story shear force of structures and the maximum drift of isolation layer. The decision vari-ables are yield coefficient , namely period and damping ratio. Characteristics of Pareto frontof the optimization are given, and relations between seismic responses and isolation layer par--ameters are analyzed.The main contents are as follows:1) First, detail on the status of performance optimization of base isolation system andthe current key issues facing and determine the main work of the paper: research into themulti-objective optimization of the seismic response of base isolation system; then, study avariety of algorithms of multi-objective optimization and selecte NSGA-Ⅱas the approachof multi-objective optimization in this paper.2) Through the analysis of the upper structure and the base isolation layer, build themodel of the base isolation in Simulink.3) Using Taft wave and El-Centro wave, multi-objective optimization of seismic responsesin rubber bearing base isolation system, sliding base isolation system and resilient-frictionbase isolation system is performed and Pareto front is given. Research shows that: for multiobjectiveoptimization of the seismic response of base isolation system, the NSGA-Ⅱalgorith-m is feasible; Pareto front of rubber bearing base isolation system has the character of Multi-mode and the character don′t change with the form of the upper structure,the type of seis-mic waves and seismic intensity; in rubber bearing base isolation system, reasonable periodof isolation layer is about 3s; when the damping ratio is one of the decision variables, thedamping ratio of the member of Pareto front is generally larger.