电力通信网络海量告警信息降噪收敛方法

Noise Reduction Convergence Method of Massive Alarm Information in Power Communication Network

针对海量告警信息电力通信网络的检测率及准确率低,文章提出一种电力通信网络海量告警信息降噪收敛方法。通过电力系统通信网络的设备状态以及报文接收状态告警之间的关系,建立报文接收状态告警模型。利用覆盖规则,以少量信息代表海量信息,得到依赖关系关联两个测量模型中的海量告警信息,并构造依赖搜索树在过滤节点的同时智能归并告警信息,有效确定网络异常点。通过电力通信网络设备中的自检监测流量异常告警等信息为输入,实现电力系统通信网络各通信环节状态还原。实验结果表明应用该方法,电力系统通信网络异常检测率及准确率均在90%以上,误报率及漏报率稳定在5%以下,可以有效实现电力系统通信网络的海量告警降噪收敛。

Aiming at the low detection rate and accuracy of mass alarm information in power communication network,a convergence method for noise reduction of mass alarm information in power communication network is proposed.Through the relationship between the equipment status of the communication network of vector power system and the message receiving status alarm,a message receiving status alarm measurement model is established.Using coverage rules to represent massive information with a small amount of information,the massive alarm information in two measurement models related by dependency relationship is obtained,and a dependency search tree is constructed to intelligently merge the alarm information while filtering nodes,so as to effectively determine network anomaly points.Through the self-inspection and monitoring of abnormal flow alarm and other information in each equipment of the power communication network,the state restoration of the power system communication network can be realized.The experimental results show that the anomaly detection rate and accuracy rate of power system communication network reach 90%,and the false alarm rate and false alarm rate are below 5%,which can effectively realize the convergence of mass alarm and noise reduction in power system communication network.