Multi-objective optimization of dual-purpose outriggers in tall buildings to reduce lateral displacement and differential axial shortening
Multi-objective optimization of dual-purpose outriggers in tall buildings to reduce lateral displacement and differential axial shortening
复制标题
DOI：
10.1016/j.engstruct.2019.03.098
复制
发表时间：
2019-06
影响因子：
5.5
通讯作者：
Han‐Soo Kim;Hye-Lym Lee;You-Jin Lim
中科院分区：
文献类型：
--
作者：
Han‐Soo Kim;Hye-Lym Lee;You-Jin Lim
最大侧向位移（LAT）和差轴缩短（DAS）是高层建筑设计中必须考虑的重要标准。支腿（OR）系统已被证明可以有效地减少高层建筑的LAT和DAS。提出了一种混合多目标优化（MOO）方法来确定ORs的最优位置，以最小化LAT和DAS。由于最小化LAT和DAS是相互冲突的目标，该方法提供了多个Pareto-front解。采用有限元方法对任意形状建筑的LAT和DAS进行了计算。采用最陡下降法进行基于梯度的直线搜索。为了克服整数设计变量带来的完整性要求，采用分段二次插值对整数变量进行松弛。采用加权和法对两个目标函数进行标化。将优化过程中识别的精英群体进行存储和比较，将进化优化的优势应用于MOO问题。结果表明，该方法得到的Pareto前沿与穷举搜索的结果相同。
Maximum lateral displacement (LAT) and differential axial shortening (DAS) are important criteria to be considered in the design of tall buildings. Outrigger (OR) systems have proven to be efficient for reducing LAT and DAS in tall buildings. We developed a hybrid multi-objective optimization (MOO) method to determine the optimal locations of ORs to minimize LAT and DAS. Because minimizing LAT and DAS are conflicting objectives, multiple Pareto-front solutions are provided by the proposed method. Finite element analysis was used to evaluate the LAT and DAS of arbitrary shaped buildings. The steepest descent method was used to perform a gradient-based line search. To overcome the integrality requirements introduced by the integer design variables, piecewise quadratic interpolation was used to relax the integer variables. The scalarization of two objective functions was performed by using the weighted-sum method. Elite populations identified during scalarized optimization are stored and compared to apply the advantages of evolutional optimization to the MOO problem. It is demonstrated that the Pareto front obtained by the proposed method provides the same results as an exhaustive search.