基于高风险模式树挖掘方法的电力系统风险设备集分析
发布时间：2018-12-24 16:36
【摘要】：迅速积累的调度控制系统大数据为电网设备风险分析提供了充足的条件,在分析调度控制系统大数据特性的基础上,给出了具有普遍适用性的调度控制系统大数据分析的总体架构,并针对在电网风险管控中的应用,提出一种基于高风险模式树(HRT)的高风险设备集挖掘方法。通过分析电力系统中设备的风险诱发因素,定义了设备风险影响度,用于量化设备发生告警或故障后对电网运行的影响程度,并提出设备风险影响度计算指标体系,通过融合设备故障发生频次,计算设备风险值。以设备风险值为目标进行高风险设备集挖掘,通过构建HRT,保留原始事务数据库中各设备风险值及设备风险先验知识信息,根据HRT的路径信息输出满足一定风险阈值的高风险设备集。以调度控制系统的海量历史告警数据为基础进行了仿真,结果表明,HRT可以在告警数据中迅速挖掘出满足条件的高风险设备集,并且能够反映出高风险设备组合之间存在的潜在关联性。
[Abstract]:Big data, a rapidly accumulating dispatching control system, provides sufficient conditions for the risk analysis of power network equipment. On the basis of analyzing the characteristics of the dispatching control system big data, In this paper, the general framework of big data analysis of dispatching control system with universal applicability is presented. Aiming at the application in power network risk control, a mining method of high-risk equipment set based on high risk pattern tree (HRT) is proposed. By analyzing the risk inducing factors of the equipment in the power system, the paper defines the risk influence degree of the equipment, which is used to quantify the influence degree of the equipment alarm or fault on the power network operation, and puts forward the calculating index system of the equipment risk influence degree. The risk value of the equipment is calculated by combining the frequency of fault occurrence of the equipment. Taking the equipment risk value as the target, the high risk equipment set mining is carried out, and the priori knowledge information of each device risk value and equipment risk in the original transaction database is retained by constructing HRT,. According to the path information of HRT, a set of high risk devices satisfying certain risk threshold is outputted. Based on the massive historical alarm data of the dispatching control system, the simulation results show that HRT can quickly mine the high risk equipment set satisfying the condition in the alarm data. And can reflect the potential correlation between the portfolio of high-risk equipment.
【作者单位】： 华北电力大学电气与电子工程学院;北京国电通网络技术有限公司;
【基金】：国家自然科学基金资助项目(51507063) 国家电网公司科技项目(B34681150152)~~
【分类号】：TM73
，
本文编号：2390836
[Abstract]:Big data, a rapidly accumulating dispatching control system, provides sufficient conditions for the risk analysis of power network equipment. On the basis of analyzing the characteristics of the dispatching control system big data, In this paper, the general framework of big data analysis of dispatching control system with universal applicability is presented. Aiming at the application in power network risk control, a mining method of high-risk equipment set based on high risk pattern tree (HRT) is proposed. By analyzing the risk inducing factors of the equipment in the power system, the paper defines the risk influence degree of the equipment, which is used to quantify the influence degree of the equipment alarm or fault on the power network operation, and puts forward the calculating index system of the equipment risk influence degree. The risk value of the equipment is calculated by combining the frequency of fault occurrence of the equipment. Taking the equipment risk value as the target, the high risk equipment set mining is carried out, and the priori knowledge information of each device risk value and equipment risk in the original transaction database is retained by constructing HRT,. According to the path information of HRT, a set of high risk devices satisfying certain risk threshold is outputted. Based on the massive historical alarm data of the dispatching control system, the simulation results show that HRT can quickly mine the high risk equipment set satisfying the condition in the alarm data. And can reflect the potential correlation between the portfolio of high-risk equipment.
【作者单位】： 华北电力大学电气与电子工程学院;北京国电通网络技术有限公司;
【基金】：国家自然科学基金资助项目(51507063) 国家电网公司科技项目(B34681150152)~~
【分类号】：TM73
，
本文编号：2390836
本文链接：https://www.wllwen.com/kejilunwen/dianlidianqilunwen/2390836.html
