基于有导向变异算子的GM-EA算法  

GM-EA:guided mutation evolutionary algorithm

作　　者：闭应洲[1,2] 陆建波[1] 丁立新[2] 元昌安[1] 

机构地区：[1]广西师范学院计算机与信息工程学院,南宁530004 [2]武汉大学软件工程国家重点实验室,武汉430072

出　　处：《计算机应用研究》2010年第4期1249-1251,共3页Application Research of Computers

基　　金：国家自然科学基金资助项目(60763012;40761027);广西自然科学基金资助(0991104)

摘　　要：为了进一步提高演化算法的效率,提出基于有导向变异算子的GM-EA算法(guided mutation evolutio-nary algorithm)。通过结合粒子群优化的方法改进郭涛算法,更好地利用当前最优解指导变异,并将算法分为探索与开采两个阶段;在开采阶段基于模拟退火方法决定是否用新个体取代旧个体,在巩固所获取的建筑块成分的同时,尽可能克服早熟收敛问题。实验结果证明了新算法的有效性。To design a more effective evolutionary algorithm,this paper introduced a new guided mutation evolutionary algorithm by combining Guotao algorithm with the idea from particle swarm optimization,which focused on exploiting the global best solution in population to direct the mutation.In order to preserve the components of building-blocks and avoid the premature problem,separated the search process as the exploration phase and exploitation phase,and in exploitation phase simulated annealing was applied as the replace policy.The experimental results show that the proposed algorithm is significantly superior to Guotao algorithm.

关 键 词：有导向的变异 郭涛算法 粒子群优化 模拟退火 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]