基于双层规划的飞行区地势设计优化模型

Bi-Level Programming Optimization Model of Airfield Surface Elevation Design

摘要:

为设计出一个符合标准且成本最低的机场飞行区地势表面，在飞行区内划分出骨架和扭曲面，上层规划从全局角度确定骨架具体位置，下层规划根据上层规划的结果，以局部表面跟天然地面的拟合程度为目标，合理衔接骨架并使得扭曲面设计最优，形成基于双层规划的飞行区地势设计优化模型，最后运用遗传算法求解。采用该模型计算得到的三明机场地势设计实例表明，优化模型的排水、坡度、坡段设计均符合既有要求，且与实际工程相比，更贴合天然地形，填方量下降7.50%，土方总计下降8.06%，工程成本降低10.73%。

Abstract:

In order to design an elevation surface of the airfield area that meets the standards and has the lowest cost， this paper divides the skeleton and the twisted surface in the airfield area. The upper-level planning determines the specific position of the skeleton from a global perspective. Based on the results of the upper-level planning， the lower-level planning aims at the degree of fitting between the local surface and the natural ground， connecting the skeleton and optimizing the design of the twisted surface. These form an optimization model of the airfield area elevation design based on bi-level programming. Finally the genetic algorithm is used to solve the problem. The elevation design example of Sanming Airport calculated by using this model shows that the drainage， slope， and slope section design of the optimized model meet the existing requirements， and compared with the actual project， it is more in line with the natural terrain. The filling volume is reduced by 7.50%， the total earthwork is decreased by 8.06%， and engineering cost is reduced by 10.73%.