摘要
A gen t和多A gen t系统已经成为计算机科学和分布式人工智能研究的一个重要方向。该文重点介绍了基于多A gen t技术的操作票生成系统的模型的建立,研究了通过A gen t选择意图模型来实现操作任务的规划,以及对单个A gen t的推理机制和知识库中的符号化描述,并讨论了A gen t自学习及多A gen t之间的交互通信等问题。开发实现基于多A gen t变电站倒闸操作票生成系统,在某变电站得到良好的应用。实践表明,基于多A gen t系统能大大提高工作效率和准确度,特别是基于A gen t选择意图模型来实现操作任务的规划更能适应外界环境的变化。
The model of MAS-based switching operation sheet making system is introduced. More researches have been made in task operation layout through the agent purpose model. The theories of single agent＇s consequence and describing repository by symbol are studied. MAS＇s self-study, communication and cooperation are also discussed. It shows that the system based on MAS can enhance the work efficiency and improve the accuracy of making switch operation sheet greatly, the task operation layout using the agent purpose model is adaptive to the changing of condition.
出处
《电力系统及其自动化学报》
CSCD
北大核心
2005年第6期17-21,共5页
Proceedings of the CSU-EPSA
关键词
变电站
智能的
多智能系统
倒闸操作票
transformer substation
Agent
multi-Agent systems （MAS）
switching operation sheet
