多种群变换遗传算法及其在优化调度中的应用(英文)  

The Comparison of Multi-reproduction Groups of Genetic Algorithms and Its Application in the Optimization Schedule

作　　者：姚志红[1] 赵国文[2] 韩兵[3] 

机构地区：[1]上海交通大学信息与控制系,上海200030 [2]上海交通大学土木建筑系,上海200030 [3]上海交通大学自动化系,上海200030

出　　处：《控制理论与应用》2001年第6期882-886,共5页Control Theory & Applications

摘　　要：提出了一种基于多种群思想的遗传算法 ,研究了种群数目对收敛速度的影响及确定原则 ,设计了适合连续离散规划问题求解的交叉与变异算子以及评价函数参数 .通过与一般遗传算法比较、统计计算及对某载重轮胎厂生产调度系统仿真 ,表明该算法的收敛速度有很大提高 。The comparison approach of genetic algorithms based on the multi_reproduction thinking is introduced, the judgement principle about the number of reproduction groups and its influence on the algorithm convergence speed are compared, and the genetic operator and the parameter of evaluation function that was built to slove the programming problem of hybrid system are designed. Compared with ordinary genetic algorithms, statistical calculation, and the simulation to the Shanghai Heavy Duty Tyre production process, the results of calculation indicate that this approach had a fast speed of convergence, and can optimize the production schedule of hybrid system in which there are continuous and discrete processes and it is broad in scale.

关 键 词：多种群变换 遗传算子 遗传算法 优化调度 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]