基于NSGA-Ⅱ&DEA混合算法的高速铁路桥梁风屏障高度多目标优化研究

MULTI-OBJECTIVE OPTIMIZATION OF WIND SCREEN HEIGHT FOR HIGH-SPEED RAILWAY BRIDGES BASED ON NSGA-Ⅱ AND DEA HYBRID ALGORITHM

为探究高速铁路桥梁风屏障高度的多目标优化问题,基于计算流体动力学理论,采用数值模拟方法计算设置有不同高度风障时,列车及桥梁各自的气动力系数。以车辆侧倾稳定性力矩系数及桥梁阻力系数为优化目标,风屏障高度为设计变量,采用多目标遗传算法(NSGA-II)求解Pareto最优解集,采用数据包络分析方法(DEA)评价Pareto解集中各个解的相对效率,得到最优风屏障高度。结果表明:采用NSGA-II&DEA混合算法对风屏障高度进行多目标优化是可行的。该优化设计方法为风屏障高度优化问题提供了一种新思路。

In order to explore the multi-objective optimization of wind screen height for high-speed railway bridges, the aerodynamic coefficients of the train and the bridge with different wind screen heights were analyzed, based on the theory of computational fluid dynamics. The roll stability moment coefficient of train and the drag coefficient of bridge were defined as optimization objectives. The height of wind screen was selected as design variable, and the multi-objective optimization genetic algorithm （NSGA-Ⅱ） was employed as the optimization method to achieve the Pareto solution set. The optimal height of wind screen was obtained by using the Data Envelopment Analysis （DEA） to evaluate the relative efficiency of the Pareto solution set. The results show that it is feasible to add the NSGA-Ⅱ and DEA hybrid algorithm to the multi-objective optimization of wind screen height for high-speed railway bridges. This optimization design method provides a new design method with the optimization problem of wind screen heights.