基于改进遗传算法的大跨机库柱间消能支撑的位置优化研究

Research on location optimization of energy dissipation braces between columns of long-span hangar based on improved genetic algorithm

大跨维修机库动力特性复杂,针对该类结构的减震研究则刚刚起步。结合大跨机库的地震响应特点,采用布置柱间消能支撑来进行机库结构地震响应的被动控制。结合减震结构优化模型特点与主要优化算法的数学原理,选取遗传算法作为结构减震装置位置优化分析工具,并在Matlab遗传工具箱基础上进行改进以提高寻优效率,通过编制目标函数程序实现改进遗传算法在结构被动控制系统位置优化分析中的应用。最后设置两类不同的适应度函数,分别进行机库结构减震位置优化与考虑阻尼器耗能效率的减震方案优化分析,结果表明遗传算法以及相应的优化程序可以有效地实现机库结构减震位置优化分析。

Dynamic characteristics of long-span hangar are very complex and the study on the shock absorption of this type of structure is just at the early stage. Combined with seismic response characteristics, fluid viscous dampers are set between columns and joints of the roof to reduce the seismic response. Based on the characteristics of optimization model of seismic reduction structure, the genetic optimization algorithm is employed for the optimization analysis of damper location. The algorithm is improved for the increased optimization efficiency and accuracy. By means of the Matlab toolbox, a damper program for the location optimization of seismic reduction structure is coded. Two types of fitness functions are provided and used for the damper location optimization of one hangar. Results show that the location optimization analysis may be conducted effectively by using the proposed optimization algorithm and the coded program.