基于免疫量子遗传算法的多峰函数寻优

Multi-modal function optimization based on immune quantum genetic algorithm

针对多峰函数优化中的全局及局部寻优问题,提出了一种结合免疫克隆算子的量子遗传算法,给出了实现流程。该方法针对量子遗传算法在复杂连续函数优化中收敛速度慢、易陷入局部极值等缺点,采用免疫克隆操作及交叉策略提高抗体成熟力及亲和性,增强抗体群分布的多样性及稳定性,有效克服了量子遗传算法容易陷于局部最优及计算缓慢的不足。通过对多峰函数的全局寻优仿真实验,并与基本遗传算法、量子遗传算法的计算结果进行比较,结果表明在相同条件下,所提算法所需循环代数少,并且其鲁棒性高于普通量子遗传算法和遗传算法。

In order to balance the global optimization and local optimization in multi-modal function,an improved quantum genetic algorithm with immune operator was introduced.This algorithm included the idea of immune clonal,operation and cross strategy.Through this operator,the diversity of antibody and affinity maturation rate got enhanced.It not only overcame the flaw of the common quantum genetic algorithm which relapsed into local optimum result but also avoided the flaw of the common immune clone algorithm which calculated slowly.Having done the global optimization experiment on the multimodal function in the same condition,the result indicates that this algorithm can settle the problem of searching the global optimization result with less iteration,and is of more robust stability compared to common genetic algorithm and common quantum genetic algorithm.