基于分布式免疫进化算法的函数优化

针对遗传算法在寻优过程和多峰值函数求解中出现的“早熟”问题以及免疫算法收敛速度较慢问题，将免疫算法和进化算法进行优势融合，并结合改进的进化算法的并行模型，提出一种新的算法--分布式免疫进化算法（distributed immune evolutionary algorithm，DIEA）。新算法主要包括记忆种群进化模块和子种群进化模块两个部分，子种群的主要功能是找出各个区间的局部最优解；主种群主要是进行全局搜索，寻找全局最优解。仿真实验表明，该算法具有很高的全局寻优能力和很快的收敛速度，适合求解复杂多峰函数优化问题。

Function optimization based on distributed immune evolutionary algorithm

In the light of the genetic algorithm’s "precocity" problem appearing in the optimization process and multi-peak function solution and the slow convergence speed in the immune algorithm,this paper proposes a new algorithm—distributed immune evolutionary algorithm(DIEA) through combining the modified distributed model of an evolutionary algorithm with the advantages of the existing evolutionary algorithm and the immune algorithm.This algorithm includes one master module which is a major global search to find the global optimal solution and several slave modules which are to find all intervals of the local optimal solution.Experimental results show that DIEA is effective in achieving multiple solutions of the multimodal optimization problems.