协同进化免疫记忆克隆算法

为提高免疫克隆算法的收敛性能,基于协调勘探和开采的思想,提出一种新型协同进化免疫记忆克隆算法.将整个进化种群分为免疫克隆种群和免疫记忆种群,将免疫克隆种群又分为2个子种群,在每个免疫克隆子种群内部采用协同竞争模式,动态更新种群规模,并采用适应度共享、动态变异机制和浓度调节机制的克隆扩增以增加种群多样性；在免疫记忆种群中采用混沌精细搜索以及精英交叉机制,提高免疫记忆种群的开采能力.利用基准测试函数进行仿真并与相关方法对比分析,所提出的算法收敛效果优越,能有效地提高全局收敛性能.

Coevoluationary Immune Memory Clonal Algorithm

In order to promote the convergence of immune clonal algorithm, a new coevoluationary immune memory clonal algorithm based on the insight of coordinating the exploration and the exploitation was proposed. The whole population can be divided into two population named as immune clonal population and immune memory population, the immune clonal population has two sub-population. The competitive coevoluationary operation can be incorporated in the sub-population of immune clonal population to dynamic update the scale of population, fitness sharing, dynamic mutation mechanism and clonal reproduction based on concentration regulation mechanism can be also incorporated in the sub-population of immune clonal population to increase the diversity of the population. The immune memory population was optimized using the chaotic accurate search and elitist crossover in order to enhance the exploitation ability. Simulation results of benchmark functions show that the proposed algorithm performs well and can get better results.