基于混合遗传算法的离散事件系统仿真优化方法研究

A Hybrid Genetic Algorithm Based Simulation Optimization Method and System for the Discrete Event System

在分析仿真优化原理的基础上提出一种新的混合遗传算法 ,该算法集成变尺度遗传算法与禁忌搜索算法为一体 ,处理离散事件系统性能参数仿真优化的问题 ,大大提高了仿真优化的效率和解的质量。基于此算法开发了一个仿真优化软件系统 ,并将该仿真优化系统与自主开发的离散系统仿真软件进行了有机的集成。

In this paper, on the basis of analyzing the principle of simulation optimization,we put forward a new hybrid genetic algorithm,in which a dynamic range genetic algorithm and taboo search are integrated to solve optimal problems of discrete event system(DES) performance parameters.The efficiency of simulation optimization and the solution quality are improved greatly. Based on this algorithm,a simulation optimization software system is developed, which is an organic unity with DES developed by ourselves. Finally,a manufacturing line optimization is given as an example with this system.