摘要
在实际生产中,车间的扰动会造成原有调度计划受到影响。为了减少扰动带来的影响并尽快恢复车间正常运行,需要根据车间当前的状态对受到影响的任务进行动态调度。针对加工车间在受到扰动影响后的任务调度问题,提出了一种基于多智能体共享认知的车间动态调度方法。将车间中每一个制造单元映射为一个智能体,采用共享心智模型实现智能体之间的共识协商和信息传递,从而形成多个智能体之间的调度方案共享心智模型实现生产的动态调度。最后,在车间智能传感网络系统的基础上,通过实例验证说明了该方法的有效性。
In the actual production,the processing workshop will be subjected to a variety of disturbance events,such as machine failure,task delay completion,etc.The occurrence of disturbance causes that the original scheduling scheme is infeasible,in order to reduce the impact of the disturbance and restore the normal operation of the workshop as soon as possible,it is necessary to dynamically dispatch the affected tasks according to the current state of the workshop.Aiming at the problem of task scheduling in the workshop disturbed by the disturbance,this paper proposes a dynamic workshop scheduling method based on multi-agent shared cognition according to the dynamic and discret characteristics of the manufacturing environment.Each manufacturing unit in the workshop is mapped into an agent.The principle of shared mentl model is used to realize the consensus negotiation among the agents,and the dynamic scheduling is realized by establishing the scheduling scheme shared mental model among the agents.Finally,based on the intelligent sensor network system of the workshop,the effectiveness of the method is illustrated by an example.
作者
秦军
睢鹏
李昌玲
翁佩佳
QIN Jun;SUI Peng;LI Changling;WENG Peijia(Shaanxi Diesel Engine Heavy Industry Co.,Ltd.,Xingping 713105,CHN)
出处
《制造技术与机床》
北大核心
2020年第1期161-168,共8页
Manufacturing Technology & Machine Tool
基金
国家工信部智能制造标准化项目:船舶与海洋工程智能制造车间互联互通标准及试验验证
关键词
动态调度
扰动
共享心智模型
多智能体系统
强化学习
dynamic scheduling
disturbance
shared mental model
multi-agent system
reinforcement learning
