This paper presents a new technique for optimizing the design of viscous dampers when they are used to improve the seismic performance of existing buildings. The technique uses a genetic algorithm optimization routine that is applicable to the highly complex and nonlinear relationship between damper properties, building response and cost. The paper describes the use of nonlinear static analyses in a manner that significantly reduces the amount of analysis required to determine building response. This makes the genetic algorithm highly efficient and reduces the time required to obtain an optimized solution. The method is illustrated with two buildings - a four-story reinforced concrete moment frame building and a nine-story steel moment frame building. Analyses of the buildings shows that the method produces excellent results, even after relatively few iterations. Typical values of optimization parameters that give good convergence are recommended. The paper also suggests criteria that can be used to determine when a near optimum design has been achieved.
展开▼
