论文部分内容阅读
针对基本粒子群优化算法容易陷入局部极值的缺陷,提出了一种免疫逃避型粒子群优化算法.其基本思想是将初始粒子群划分为寄生与宿主两个种群以模拟生物寄生行为,对寄生种群的粒子采用精英学习策略,对宿主群的粒子采用探索策略,再引入免疫系统的高频变异对寄生群采用相应的免疫逃避机制,以增强群体逃离局部极值、提高算法的全局寻优能力.采用标准测试函数的实验结果表明,该算法在收敛速度和求解精度方面均有显著改进.
In order to overcome the defect that basic particle swarm optimization algorithm is easy to fall into local extremum, an immune evasive particle swarm optimization algorithm is proposed, whose basic idea is to divide the initial particle swarm into two groups, parasitic and host, to simulate the biological parasitism. The population particles adopt the elite learning strategy, explore strategies for the host population particles, and then introduce the high-frequency mutation of the immune system to the parasitic group to use the corresponding immune evasion mechanism to enhance the population to flee the local extremum and improve the global optimization ability of the algorithm Experimental results using standard test functions show that the proposed algorithm has significantly improved convergence speed and accuracy.