电网异常检测模型方法设计

Research on Modeling of Power Network Anomaly Detection

采用序列模式挖掘算法构建电网异常检测模型,能够更好地表现电网异常行为。本文主要是进行了电网异常检测建模方法的研究,并提出了一个基于模糊序列模式的电网异常检测模型。本文采用改进后的SPADE算法从电网连接记录中挖掘序列模式,据此构建电网异常检测模型,采用规则裁减策略来剔除冗余规则,从而缩减了模式库,加快了检测过程,同时也降低了误报率。通过理论分析和仿真实验证明,本文提出的模型不仅具有检测异常行为的实际能力,而且与传统模型相比,由于引入了模糊逻辑理论和规则裁减技术,模型具有更简洁的行为模式库和更低的误报率,检测效率和检测性能都得到了较大的改善。

It can better show the abnormal behavior of the power line when using the Sequence pattern mining algorithm to build power network anomaly detection model.This article is mainly a research of the method of power network anomaly detection model,and proposes a model based on fuzzy sequence pattern.In this thesis,the improved SPADE algorithm is applied to mining sequential pattern from network connection records,which results in a more efficient network anomaly detection model.The rewards are the reduced pattern,accelerated detection process and less false positive rates.Theoretical analysis and simulation experiments show that the model presented in this paper not only has the actual capability,but also has a more concise behavior pattern model and lower false alarm rate compared with the traditional model,because it made the introduction of fuzzy logic theory and the rules reduction techniques,the detection efficiency and detection performance have been improved considerably.