论文部分内容阅读
设计面向海工项目多主体协商生产调度问题的多智能体求解系统架构,明确海工生产调度多智能体系统的协商机制和冲突消解策略,为问题的求解提供基础保障。此外,为增强智能体的推理决策能力,提高问题求解效率和求解精度,基于分层强化学习理论提出面向海工生产调度问题求解的学习模型与多智能体系统学习机制。在此基础上,开发面向海工生产调度问题的多智能体求解原型系统,进行实例验证,为在海工企业中进行实际应用提供可能。
This paper designs a multi-agent solution system architecture for multi-agent negotiated production scheduling problem of marine engineering project, clarifies the negotiation mechanism and conflict resolution strategy of multi-agent system in marine production scheduling, and provides the basic guarantee for solving the problem. In addition, in order to enhance the reasoning and decision-making ability of the agent and improve the efficiency and accuracy of the problem solving, a learning model and multi-agent system learning mechanism for solving the problem of marine production scheduling are proposed based on the hierarchical reinforcement learning theory. On this basis, the development of multi-agent prototype system for marine production scheduling problem is verified by examples, which is possible for practical application in offshore enterprises.