一种改进的遗传算法及其在布局优化中的应用

该文以人造卫星舱布局为背景,研究二维带平衡及不干涉等约束的圆集在圆容器内的布局优化问题,属于NP-困难问题.文章提出一种求解此类问题的改进的遗传算法,即十进制编码控制参数自适应遗传算法,从而缓解了“组合爆炸”和遗传算法的早熟收敛问题.文章给出两个算例(其中一个为作者构造的已知最优解的算例),计算结果表明了此算法的有效性,并且优于数学规划的乘子法的计算结果.此遗传可推广应用于其他布局优化问题的求解.

A Modified Genetic Algorithm and Its Application to Layout Optimization

Taking the layout problem of satellite cabins as background, the authors make a study of the optimal layout problem of circle group in a circular container with performance constraints of equilibrium etc. in this paper, which belongs to NP-complete problem. A modified genetic algorithm called decimal coded adaptive genetic algorithm (DAGA) is presented to solve the problem. In this way, the algorithm relaxes the combinatorial explode and the premature convergence of genetic algorithm. Two examples are provided (one of them is proposed by the authors, and its optimal solution is known). The numerical results show that the DAGA is effective and superior to multiplicator algorithm. The DAGA can be developed to solve other layout optimization problems.