摘要
通信机房蓄电池是通信机房系统重要的组成部分之一,其运行状态对于保障现网稳定运行具有直接影响。针对通信机房蓄电池的运行状态,文章提出了一种基于深度置信网络(deep belief network,DBN)神经网络的蓄电池荷电状态(state of charge,SOC)和基于粒子数滤波算法的蓄电池健康状态(state of health,SOH)的联合评估方案。该方案以蓄电池电压的时间序列为研究对象,利用DBN神经网络实现蓄电池SOC分类,利用粒子数滤波算法实现蓄电池SOH计算,实现通信机房蓄电池运行状态的实时评估。仿真结果表明,所提出的方法与专家经验评估方案相比,其容量估计和寿命预测的准确性保持在较高水准。
The battery in the communication room is one of the important components of the communication room system,and its operating status has a direct impact on ensuring the stable operation of the existing network.This paper proposes a joint evaluation scheme for the state of charge(SOC)of battery cells based on deep belief network(DBN)neural network and the state of health(SOH)of battery cells based on particle number filtering algorithm for the operation status of communication room batteries.This scheme takes the time series of battery voltage as the research object,uses DBN neural network to classify battery SOC,and uses particle number filtering algorithm to calculate battery SOH,achieving real-time evaluation of battery operation status in communication rooms.The simulation results show that the proposed method maintains a high level of accuracy in capacity estimation and life prediction compared to expert experience evaluation schemes.
作者
王磊
黄丰
莫辉强
王强
杨震
黄先南
WANG Lei;HUANG Feng;MO Huiqiang;WANG Qiang;YANG Zhen;HUANG Xiannan(Guangzhou Branch,China Mobile Communications Group Guangdong Co.,Ltd.,Guangzhou 510335,Guangdong Province,China;Zhejiang Rail Transit Operation Management Group Co.,Ltd.,Hangzhou 310005,Zhejiang Province,China;Haining Rail Transit Operation Management Co.,Ltd.,Jiaxing 314400,Zhejiang Province,China;China Mobile Communications Group Guangdong Co.,Ltd.,Guangzhou 510623,Guangdong Province,China)
出处
《电力信息与通信技术》
2025年第9期67-72,共6页
Electric Power Information and Communication Technology
关键词
通信机房蓄电池
深度置信网络
蓄电池荷电状态
粒子数滤波
蓄电池健康状态
communication room battery
deep belief network
state of charge
particle number filtering
the state of health of battery
作者简介
王磊(1991),男,硕士研究生,工程师,从事传输机房动力配套及传输维护及管理,13922209937@139.com;黄丰(1973),男,硕士研究生,副高级工程师,从事铁路、轨道交通车辆技术、运营管理方向的研究工作;莫辉强(1979),男,硕士研究生,副高级工程师,从事铁路、轨道交通车辆技术、运营管理方向的研究工作;王强(1982),男,硕士研究生,高级工程师,从事数据中心动力能源方面的维护及管理工作;杨震(1988),男,硕士研究生,工程师,从事核心机楼机房动力和节能维护及管理工作;黄先南(1992),男,本科,工程师,从事网络运维管理和IT研发创新工作。
