基于OCSVM的配电通信网流量异常检测研究

Research on network traffic anomaly detection in distribution communication network based on OCSVM

随着智能配电网的发展,其业务类型不断增加,网络接入方式更加复杂,智能终端广泛接入,其所面临的网络威胁也不断增加。当智能配电网通信网络遭受攻击时,会严重影响到电力系统的安全稳定。如何及时准确识别配电网通信系统中的异常流量,对配电网的安全有着重要的意义。根据配电网网络流量的特点,本文研究基于单类支持向量机(OCSVM)的配电网通信网流量异常检测方法。该方法可以有效检测异常的网络流量,提高配电通信网的安全防护水平。

With the development of intelligent distribution network, its service types are increasing, network access mode is more complex, and intelligent terminals are widely accessed, as a result, the network threats are increasing. When the com- munication network of smart distribution network is attacked, it will seriously affect the security and stability of the power system. How to identify the abnormal flow in distribution network communication system timely and accurately is of great significance to the safety of distribution network. According to the characteristics of distribution network traffic, this paper studies the traffic anomaly detection method of distribution network communication network based on single class support vector machine （OCSVM）. This method can detect abnormal network traffic effectively and improve the security protection level of distribution network.