一种结合混沌搜索的自适应遗传算法(英文)

AN ADAPTIVE GENETIC ALGORITHM COMBINED WITH CHAOS SEARCHING

在系统分析和研究自适应遗传算法特点的基础上,提出了一种新颖的混合软计算:结合混沌搜索的自适应遗传算法.一方面,算法将具有对初值敏感、易跳出局部极小、搜索速度快和计算精度高的混沌优化算法引入到自适应遗传算法中,以平衡其"开发"和"探测"之间的性能;另一方面,算法设定群体早熟收敛的量化计算公式和判定阈值,并引入了一组新的自适应交叉率和变异率的计算函数,从而有效防止了算法陷入局部最优的缺点.通过对4个基准测试函数的仿真计算,证明该算法能有效提高全局寻优的性能,且鲁棒性好.

A novel hybrid soft computing： an adaptive genetic algorithm combined with chaos searching is proposed in this paper. In order to enhance the performance of the genetic search process, on the one hand, two sets of crossover and mutation rates are employed to automatically maintain the balance between exploration and exploitation. On the other hand, the chaos searching is introduced so as to avoid being trapped into the local optimum. The effectiveness of the proposed approach is demonstrated by applying it to four benchmark functions obtained from the literature. Furthermore, the simulation results show that this approach can provide favorable performance over other existing algorithms.