基于蚁群算法的模具制造动态调度研究  

Dynamic scheduling of mold manufacturing based on ant colony optimization

作　　者：王延斌[1] 王刚 赵立忠[1] 高国安[1] 

机构地区：[1]哈尔滨工业大学现代生产技术中心,黑龙江哈尔滨150001

出　　处：《计算机集成制造系统》2006年第7期1028-1036,共9页Computer Integrated Manufacturing Systems

基　　金：国防科学技术工业委员会"十五"重点资助项目。~~

摘　　要：为解决模具制造动态调度问题,建立了动态调度系统。该系统利用蚁群算法和优先分配启发式算法相结合的调度算法,解决具有工件约束的模具零件的调度问题。该算法首先由蚁群算法确定模具零件各工序所用加工机床,然后利用优先分配启发式算法确定在同一台机床上加工的各零件的先后顺序。考虑动态调度的实时性,提出了局部更新和全局更新相结合的、基于滑动窗口机制的动态调度方法。对于发生频率高但对调度计划执行影响不大的扰动事件采用局部更新策略,反之则采用全局更新策略,在保证获得近优解的同时提高了动态调度的效率。A scheduling system was established to solve the dynamic scheduling problem in mold manufacturing. The system combined the Ant Colony Optimization （ACO） algorithm and the dispatching priority heuristic algorithm to solve the scheduling problems existing in mold parts with work piece restrictions. Firstly, ACO was used to determine the machines needed in each working procedure, then priority heuristic algorithm was used to set the sequence of the parts to be machined in the same machine. A dynamic scheduling algorithm based on moving window and consideration of local and global update was put forward for real--time mold scheduling. Local update strategy was adopted for those turbulences, which happened in high frequency but with little influence on the scheduling; otherwise, global update strategy was adopted. The scheduling efficiency was improved as well as the near--optimum solution was guaranteed.

关 键 词：工件约束 蚁群优化算法 优先分配启发算法 模具制造 动态调度 

分 类 号：TH166[机械工程—机械制造及自动化] TG76[金属学及工艺—刀具与模具]