基于免疫规划的单亲遗传算法研究及其应用  

New partheno genetic algorithm based on immune programming and its application

作　　者：姚建斌[1] 行小帅[1] 霍冰鹏[1] 

机构地区：[1]山西师范大学物理与信息工程学院,山西临汾041004

出　　处：《系统工程与电子技术》2007年第4期636-639,共4页Systems Engineering and Electronics

基　　金：山西省青年科技研究项目资助课题(20031008)

摘　　要：在分析了单亲遗传算法的优越性与存在不足的基础上,借鉴生物免疫概念与理论,提出了一种新的单亲遗传算法——基于免疫规划的单亲遗传算法。该算法的核心在于使用最优保留策略前提下,合理地构造了非均匀算子和免疫算子。理论分析和仿真结果表明,该算法不仅能够有效地保持群体多样性,而且减轻了遗传算法的后期波动现象,同时收敛速度明显提高。After analyzing the advantages and disadvantages of the partheno genetic algorithm, a novel algorithm, partheno genetic algorithm based on immune programming （IPPGA） is proposed with the concept and theory of immunity in biotic science being used for reference, which incorporates the partheno genetic algorithm into the immune programming. On the premise of the elitist reserve strategy, the core of the algorithm lies on constructing reasonably the nonuniform mutation operator and immune operator. The theory analysis and experimental results show that the algorithm not only effectively keeps population diversity, but also alleviates the undulate phenomenon in evolutionary latter stages, meanwhile increases the convergent speed greatly.

关 键 词：单亲遗传算法 免疫规划 非均匀变异算子 免疫算子 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]