基于多Agent的变电站操作票自动生成系统  

Multi-agent based operation instructing system in substation

作　　者：姜惠兰[1] 刘洁[1] 梁昭君[1] 岳亚林 

机构地区：[1]天津大学电气与自动化工程学院,天津300072 [2]天津高压公司,天津300250

出　　处：《电工电能新技术》2005年第4期18-21,67,共5页Advanced Technology of Electrical Engineering and Energy

基　　金：天津大学留学回国人员基金资助项目(200447)

摘　　要：变电站操作票制度是保证变电站安全、可靠运行的重要措施,因此开具正确无误的倒闸操作票对于变电站的运行操作和维护管理具有重要的意义。本文提出将分布式人工智能之多Agent理论应用于操作票自动生成系统中,并结合RBF神经网络技术,建造了由设备操作序列生成Agent、校核Agent、解释Agent、协调控制Agent组成的具有分布式结构的变电站操作票MAS。研究表明该系统的实现过程简单、维护管理方便、具有较强的智能性。The operation instructing system in substations of power systems is an important guarantee for secure and reliable operation in substations, so it is significant to making up correct switch operation-order-sheet for the operation and maintenance management of substations. This paper proposes to apply multi-agent theory of distributed artificial intelligent technique in operating-order-sheet automatic generating system and build a distributed MAS system to accomplish the operation instructing system in substation. This system consists of devices＇ operating sequence＇s generating Agent, checking Agent, misoperation prevention Agent and coordination control Agent. For every sub-Agents of devices＇ operating sequence＇s generating Agent which plays kernel function in whole system, RBFNN is adopted to accomplish hidden inference to gain device operating sequences by using the performance of strong associative memory and high fault-tolerance of RBFNN, so that some of the disadvantages of expert system technique can be avoided. Study shows this system has the teatures of simple processing, convenient maintenance management, and high intelligence, and plays intelligent assistant role for field operation and management.

关 键 词：变电站 多AGENT 操作票 RBFNN 

分 类 号：TM732[电气工程—电力系统及自动化]