基于随机森林算法的电力骨干通信网故障诊断研究

Failure Diagnosis Method Based on Random Forest in Electric Backbone Communication Networks

在如今现代化智能电网中,电力骨干通信网已经成为其非常重要的组成部分。如何对电力骨干通信网故障的类型做出快速、准确的判断,对安排人员进行故障维修具有重要的指导意义。现有的电力通信网故障诊断要么是人工经验法,要么所用诊断算法效果不佳。文章将随机森林算法用于电力骨干通信网故障诊断中,快速、准确地学习出故障诊断模型。从实验结果可看出,基于随机森林算法对通信网故障诊断可达85.02%的准确率。

In today’s modern smart grid, the electric backbone communication networks has become a very important part of it.How to quickly and accurately classify the types of electric backbone communication network faults has important guiding significance for arranging personnel for fault repair. The existing fault diagnosis methods of electric communication network are either artificial experience method or the fault diagnosis algorithm used is not working well. In this paper, we use the random forest algorithm in the fault diagnosis of electric backbone communication network to quickly and accurately learn the fault diagnosis model. It can be seen from the experimental results that the diagnosis based on the random forest algorithm, the communication network fault diagnosis can reach 85.02% accuracy.