Arrow型蜂窝结构抛物柱面成型蚁狮算法优化

Parabolic Cylinder Formation and Ant Lion Optimization for Arrow Honeycomb

本研究将Arrow型零泊松比蜂窝结构用于空间抛物柱面天线。首先分析零泊松比蜂窝结构柱面成型所需加载方式,通过几何计算初步得到柱面成型所需载荷;再将ABAQUS分析、曲面拟合和型面精度计算整合为优化目标函数;然后采用蚁狮优化(ALO)算法求解使得型面精度最高的优化载荷,使用有效集法(Active Set)作为对照组,验证蚁狮优化算法的性能。优化结果得到了RMS为0.071 mm的抛物柱面,证明了将零泊松比蜂窝结构用于柱面天线反射面的可行性。与几何计算结果相比,通过ALO算法使得其型面精度提高了45.5%,收敛所需迭代步数为有效集法的1/7,证明了ALO算法对于解决初值未知优化问题的有效性和高效性。研究结果对于零泊松比蜂窝结构的优化设计和ALO算法的应用具有一定意义。

Arrow honeycomb was proposed for space cylindrical antenna to obtain the reflector with certain precision according to its zero Poisson′s ratio(ZPR)and single curve bending characteristic.Loading method and initial loads for cylinder formation were obtained by geometric analysis at first.Then,ABAQUS analysis,surface fitting and precision calculation were integrated into objective function and Ant Lion optimization(ALO)was used for searching optimal loads.Active Set algorithm was the compared group to verify the performance of ALO.Because of the deformed cylinder with 0.071 mm RMS error showed in the results,the feasibility of ZPR honeycomb used for cylindrical antenna reflector is proved.The precision of the deformed surface is improved by 45.5%through ALO algorithm compared with geometric analysis.The total iterations to converge are 7 times less than Active Set algorithm,which proves the high effectiveness and efficiency of ALO for solving optimization problems with unknown initial values.This research is of certain significance for the optimal design of ZPR honeycomb and application of ALO algorithm.