基于LVQ神经网络的电力网络智能风险预警方法  

Intelligent risk early warning method for power network based on LVQ neural network

作　　者：袁杰生 刘志民 吴天磊 YUAN Jiesheng;LIU Zhimin;WU Tianlei(Zhongshan Power Supply Bureau of Guangdong Power Grid Co.,LTD,guangdong,zhongshan 528400,China;Baili Data Co.,Ltd,guangdong,guangzhou 510000,China)

机构地区：[1]广东电网有限责任公司中山供电局,广东中山528400 [2]佰聆数据股份有限公司,广东广州510000

出　　处：《长江信息通信》2024年第1期17-19,共3页Changjiang Information & Communications

摘　　要：针对现有电力网络风险预警方法在实际应用中,存在无法找出全部异常数据,无法实现对电力网络安全的准确判别问题,引入LVQ神经网络,开展对电力网络智能风险预警方法的设计研究。从多个来源采集电力网络运行数据,并对其处理。利用LVQ神经网络,实现对异常数据的排查和定级。最后,应用全业务合规管理监控,实现对电力网络风险的智能化预警和对预警结果的可视化展现。通过对比实验证明,新的预警方法可以实现对电力网络中异常数据的全部查出,促进电力网络运行安全性提升。In view of the practical application of the existing power network risk warning method,it is impossible to find out all abnormal data and realize the accurate discrimination of power network security,LVQ neural network is introduced to carry out the design and research of the intelligent risk warning method of power network.Collect the operation data of the power network from multiple sources and process it.The LVQ neural network is used to check and level the abnormal data.Finally,the whole-service compliance management and monitoring is applied to realize the intelligent warning of powernetwork risks and the visual display of early warning results.Through the comparative experiment,it is proved that the new early warning method can realize all the abnormal data in the power network and promote the security of the power network operation.

关 键 词：LVQ神经网络 网络 预警 风险 智能 电力 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]