A dynamic flexible workshop scheduling method and system based on heterogeneous graph reinforcement learning

By using heterogeneous graph reinforcement learning, a workshop state representation is constructed and hierarchical feature extraction is performed, which solves the complex problem of dynamic flexible workshop scheduling, realizes efficient and adaptive scheduling decisions, and improves production efficiency and resource utilization.

CN122414599APending Publication Date: 2026-07-17LANZHOU UNIVERSITY OF TECHNOLOGY
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
LANZHOU UNIVERSITY OF TECHNOLOGY
Filing Date
2026-03-11
Publication Date
2026-07-17

Smart Images

  • Figure CN122414599A_ABST
    Figure CN122414599A_ABST
Patent Text Reader

Abstract

本发明属于制造业智能调度技术领域，涉及一种基于异构图强化学习的动态柔性车间调度方法及系统，步骤1：异构图构建与状态表示，将车间中的作业、机器与操作建模为异构图的节点，通过边表示节点间的加工顺序、机器选择、资源竞争与动态事件关系，构建动态可更新的图结构以表示调度状态；步骤2：分层注意力特征提取，采用节点级注意力机制捕获同类型节点间的局部依赖关系，并通过语义级注意力机制融合不同类型节点的全局语义信息，提取用于调度的低维嵌入表示；步骤3：深度强化学习决策与优化，基于提取的图嵌入表示，设计包含调度规则的策略集，通过深度强化学习代理与调度环境交互，学习最小化总延迟的调度策略，并实时响应动态事件。
Need to check novelty before this filing date? Find Prior Art