摘要
该文对调度算法做了简单的介绍。在结合已有的模拟退火算法和遗传算法的基础上,改进了现有的遗传调度算法,自适应地保存最优个体,并对其进行模拟退火。与简单最优保存遗传调度算法进行了比较,结果表明新的算法比原有算法搜索能力更强,在跳出局部最优方面也有改进,有效地解决了原有遗传调度算法的早熟现象。
This thesis introduces the scheduling algorithm briefly.Based on combination of simulated annealing algorithm(SAA)and genetic algorithm,we have improved the existing genetic scheduling1algorithm and propose a new genetic scheduling algorithm maintaining optima adaptively with simulated annealing(SAMOAGSA).This algorithm can maintain some optimal offspring adaptively and make them simulated annealing.To compare with the simple genetic scheduling algorithm(MOSGSA)on the effectiveness,this algorithm has more strong searching ability that can abandon the local optimal solution and find the global one.So the premature of MOSGSA can be solved efficiently.
出处
《计算机工程与应用》
CSCD
北大核心
2004年第17期64-66,92,共4页
Computer Engineering and Applications
基金
上海市科委自然科学基金项目(编号:00JC14052)
上海市教委项目:网格技术-E研究院资助
关键词
自适应
遗传调度算法
最优保存
模拟退火
DAG图
adaptive properties,genetic scheduling algorithm,maintaining optimum,simulated annealing,Directed Acyclic Graph(DAG)