基于最优保留策略的改进遗传算法

Improved genetic algorithm based on elitist reserved strategy

为有效解决遗传算法收敛速度慢和早熟收敛的问题,提出一种基于最优保留策略的改进方法。对遗传算法的选择算子和变异算子同时加以改进优化,将群体优胜劣汰的思想有效融入遗传算法框架,保障最优个体的基因能迅速向后代传播,加快收敛速度。提出最优个体优化变异的思想,避免算法落入局部最优。给出算法实施的具体步骤,在8个基准测试函数上进行仿真实验。数据比较和分析结果表明,该算法在收敛速度与全局收敛能力上都有较大的改善。

Aiming at increasing the convergence speed and avoiding the premature convergence, an improved genetic algorithm based on the elitist reserved strategy was presented. Both the selection operator and the mutation operator in genetic algorithm were optimized. The principles of the superiors surviving were integrated into the frame of the genetic algorithm, thus the fast transmission of the elitist genes into the later generation and the increase of the convergence speed were guaranteed. Meanwhile, the mutation of the elitist was put forward so as to achieve the global convergence of the algorithm. Some specific implementing steps of the algorithm were presented and simulation experiments on eight benchmarking functions were conducted as well. Re- sults show that this genetic algorithm is greatly enhanced in terms of the convergence speed and the global optimality after com- paring and analyzing the data generated from the experiments.