摘要
研究了多目标柔性作业车间调度问题,提出了一种改进的自适应免疫遗传算法。算法根据搜索的历史信息,自适应的调整遗传过程中的遗传参数以提高算法的稳定和效率。针对遗传算法的局部搜索能力差和全局搜索效率低的问题,结合免疫算法的免疫记忆和接种疫苗,对各近似最优解进行动态邻域搜索,提高算法的局部搜索能力和解的质量;免疫反馈和免疫选择能淘汰相似个体,维持种群的多样性,避免算法陷入早熟,改善算法的性能和稳定性。最后通过仿真实例验证了算法的有效性。
An improved self-adaptive immune genetic algorithm was put forward for multi-objective flexible job shop scheduling. It self-adaptively adjusted genetic operators based on history information of algorithm so as to increase search stability and efficiency. According to solve poor ability of local search and lower efficiency of genetic algorithm, immune memory and vaccination were adopted to dynamic search near each approximate optimal solution, so it accelerated search and improved quality of solution. Meanwhile, immune feedback and selection could eliminate semblable individuals and maintain variety of population for avoiding premature and improving performance and stability. Finally the result of simulation indicates validity of algorithm.
出处
《系统仿真学报》
EI
CAS
CSCD
北大核心
2008年第22期6163-6168,共6页
Journal of System Simulation
基金
国家科技攻关计划资助项目(2001BA201A01-01)
科技部科技攻关资助项目(2001BA201A56)
关键词
柔性车间调度
多目标
自适应
免疫算法
遗传算法
flexible job shop scheduling,multi-objective,self-adaptive,immune algorithm,genetic algorithm
