基于改进随机森林算法的电力业务实时流量分类方法

发布时间：2018-02-05 19:22

本文关键词： 随机森林数据剪辑分类间隔电力业务流量分类　出处：《电力系统保护与控制》2016年24期 　论文类型：期刊论文

【摘要】：为了更有效地对电力业务系统安全接入过程中日渐增多的流量进行实时分类,提高电力系统的业务处理速度,提出了一种基于改进随机森林算法的电力业务实时流量分类方法。在分析电力业务安全接入实时流量特征的基础上,改进传统随机森林算法,基于分类间隔加权对随机森林进行修剪来提高分类实时性;对新的样本数据进行数据剪辑来提高分类的准确性。在此改进算法的基础上设计了电力业务安全接入实时流量分类流程。最后以某省电力公司安全接入实时流量分类为例,验证了所提方法的准确性和实时性。
[Abstract]:In order to effectively classify the increasing traffic in the process of secure access to power system, and improve the processing speed of power system. This paper presents a real-time traffic classification method for power services based on improved stochastic forest algorithm, which improves the traditional stochastic forest algorithm on the basis of analyzing the characteristics of power traffic security access real-time traffic. In order to improve the real-time of classification, the random forest is pruned based on the classification interval weighting. The new sample data is edited to improve the accuracy of classification. On the basis of the improved algorithm, the flow of real-time traffic classification for secure access to power services is designed. Finally, a provincial power company is used to secure access to real-time traffic. Classified as an example. The accuracy and real-time of the proposed method are verified.
【作者单位】：北京中电普华信息技术有限公司;国网新疆电力公司;国网浙江电力公司;华北电力大学控制与计算机工程学院;国家开发投资公司;
【分类号】：TM73

【相似文献】