基于多Agent的加工时间定额确定方法

Multiagent-based Approach for Determining Machining Time-quota

研究了CAPP和生产调度集成环境中的时间定额问题 ,提出了一种基于Agent的时间定额确定方法。采用智能Agent的方法 ,建立了集成环境中确定时间定额的结构模型。将影响时间定额的因素映射成与工序状态有关的零件特征Agent和加工方法Agent。以神经网络Agent为计算工具 ,借助于Agent的判断推理能力 ,将各因素处理成神经网络可以识别和接受的数据模式。利用黑板结构处理各Agent之间的信息交互和协同控制 ,以快速确定工序的时间定额问题。实验表明 ,只要选择适当的样本 ,并在系统中不断进行自学习和自组织 。

Time quota in CAPP and production scheduling integrated environment is studied.An agent based time quota method is put forward and the structural model is established by means of intelligent agent in integrated environment.The method can map the factors influencing time quota into part agent related to process condition and machining method agent. With the aid of the function of agent rule based reasoning,the agents can transform these factors into data modes that artificial neural network(ANN) can accept and recognize.As a tool, ANN agent can calculate time quota. A blackboard method is used as the means of communication between and collaborative control of agents. The experiments show that precise process time quota can be obtained rapidly with proper samples selected, continuous self study and self organization in system, and multi agent approach is an effective method to determine time quota.