中文摘要：

针对树型网络的拓扑结构和数学模型,从个体编码、种群初始化、种群进化、适应度函数等方面构建基于遗传算法的网络优化方法.实验结果表明：所构建的方法进一步修正了适应度函数,增强了弱势个体被选择的概率,避免遗传算法优化过程的过早收敛问题,缩短了执行时间,取得了较佳的网络优化结果.

英文摘要：

Based on genetic algorithm, a network optimization method is proposed according to the topology and mathe- matical model of tree-shape network from the aspects of individual encoding, population initialization, population evolu- tion, fitness function and so on. Experimental results show that the proposed method can further modify the fitness func- tion, enhance the probability of the weak individuals＇ being chosen, avoid the premature convergence of genetic algo- rithm, and reduce the execution time. The results show good networking optimization.