电信科学 ›› 2020, Vol. 36 ›› Issue (10): 153-158.doi: 10.11959/j.issn.1000-0801.2020155

• 运营技术广角 • 上一篇    下一篇

基于DBN-Softmax的电力通信网络带宽预测

李佳,丛犁,姜华,胡杨,徐梦   

  1. 国网吉林省电力有限公司信息通信公司,吉林 长春 130021
  • 修回日期:2020-05-15 出版日期:2020-10-20 发布日期:2020-11-07
  • 作者简介:李佳(1991- ),女,国网吉林省电力有限公司信息通信公司工程师,主要从事通信网络运行工作|丛犁(1984- ),女,博士,国网吉林省电力有限公司信息通信公司高级工程师,主要从事通信网运行管理工作|姜华(1985- ),男,国网吉林省电力有限公司信息通信公司工程师,主要从事通信网调度管理工作|胡杨(1988- ),男,国网吉林省电力有限公司信息通信公司工程师,主要从事通信网调度工作|徐梦(1993- ),女,国网吉林省电力有限公司信息通信公司助理工程师,主要从事通信网络检修工作

Bandwidth prediction of power communication network based on DBN-Softmax

Jia LI,Li CONG,Hua JIANG,Yang HU,Meng XU   

  1. Information and Communication Company,State Grid Jilin Electric Power Co.,Ltd.,Changchun 130021,China
  • Revised:2020-05-15 Online:2020-10-20 Published:2020-11-07

摘要:

随着电力通信网的变化,电力通信网承载业务数据呈指数级增长,对电力通信网的处理能力提出了更高要求。为保障通信网的服务质量,针对目前网络带宽分配不合理现象,提出基于深度置信的电力通信网带宽预测算法,该算法通过由限制玻尔兹曼机构成的深度置信网络获取能够完美表达网络带宽的特征,实现对电力通信网规划阶段带宽的合理预测。实验结果表明,与传统神经网络算法相比,所提算法在预测精度和稳健性方面更具有优势,可以提高电力通信网的承载能力,为电力系统的安全稳定运行提供有力的保障。

关键词: 电力通信网, 深度置信, 机器学习, 带宽预测

Abstract:

With the change of the power communication network,the data of the bearer service of the power communication network has increased exponentially,which puts higher requirements on the processing capability of the power communication network.In order to guarantee the service quality of communication network,aiming at the current unreasonable distribution of network bandwidth,a bandwidth prediction algorithm based on deep confidence for power communication network was proposed.The deep confidence network formed by the Boltzmann machine was used to obtain the characteristics that could perfectly express the network bandwidth,and the reasonable prediction of the bandwidth of the power communication network planning stage was realized.The implementation results show that the proposed algorithm is more accurate and robust than neural network.It has the advantage of improving the carrying capacity of the power communication network and providing a powerful guarantee for the safe and stable operation of the power system.

Key words: power communication network, deep confidence, machine learning, bandwidth prediction

中图分类号: 

No Suggested Reading articles found!