机器学习法在计量系统告警信息中的研究与应用  

Research and Application of Machine Learning Method in the Management of the Alarm Information in Measuring System

作　　者：韦瑞录[1] 覃能杰 

机构地区：[1]广西机电职业技术学院电气工程系,南宁530007 [2]广西博联信息通信技术有限责任公司,南宁530021

出　　处：《价值工程》2016年第11期171-174,共4页Value Engineering

摘　　要：随着计算机技术、网络技术和数据库技术的快速进步和发展,电力运行网络已经引入了先进的自动化管理技术,开发了许多的分布式管理系统,电力系统运行积累了海量的数据。尤其是计量告警系统规模巨大,告警信息种类繁多,人工识别已经无法满足现代电网实时监控需求,需要采用机器学习、数据挖掘技术,从海量告警信息中挖掘潜在的模式,对告警信息进行自动化分类管理,以提高报警信息分析的准确度,提高电网监控管理成效,为电网计量系统正常运行提供支撑。With the fast growth and the rapid development of the computer, internet and database technology, some advanced automation technologies have been imported into the electric power operation network, and many distributed management systems have been developed. The power system operation has accumulated vast amounts of data, especially the alarming system. Because of so many different kinds of alarm information, artificial recognition has been unable to meet the demand of real-time monitoring of the modern electric network. Therefore it should use the machine learning method and the data mining technology to mine potential models from these different kinds of alarm information and apply the automatic classification management to the alarm information so as to enhance the accuracy of alarm information analysis and improve the efficiency of grid monitoring management, which can support the normal operation of the power grid measuring system.

关 键 词：机器学习 计量系统 告警信息 数据挖掘 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]