改进的多目标遗传算法在无人机机翼结构优化中的应用

An Updated Multi-objective Genetic Algorithm and Its Application to Unmanned Aircraft Wing Structural Optimization

现有的多目标遗传算法往往只能求得整个非劣曲线的一部分,同时局部搜索能力差,收敛速度较慢。为了解决这些问题,提出了一种改进算法,该算法将非劣分层遗传算法(NSGA)与向量评估遗传算法(VEGA)的优点结合起来,并且提供了一个利用往代信息构造搜索方向的局部搜索算子,有效扩展了非劣曲线的范围,加快了收敛速度。以某无人机机翼结构的多目标优化问题为例,证明本文改进算法可以较为快速地获得一个分布均匀的非劣解集。

Current multi -objective genetic algorithms usually can only attain part of the whole pareto front, at the same time, because of the worse local searching ability, the convergence speed is slow. In order to overcome these disadvantages, an updated multi -objective genetic algorithm is proposed in this paper. The updated algorithm not only integrates the merits of the Non - dominated Sorting Genetic Algorithm （NSGA） and the Vector Evaluated Genetic Algorithm （VEGA）, but also has a local searching operator which constructs the searching direction by using the previous population＇s information, so it can effectively expand the scope of non - inferior solutions and improve the convergence speed. Using the updated algorithm, this paper succeeds in optimizing a large unmanned aircraft wing structure. The result indicates that the new algorithm can rapidly acquire uniform non - inferior solutions and prove the superiority of the algorithm.