首页 | 本学科首页   官方微博 | 高级检索  
     

基于时空分布映射的大规模电池健康状态研究
引用本文:王波,陈东东,张锦霞,张之琛,马星星,张志宏.基于时空分布映射的大规模电池健康状态研究[J].陕西电力,2022,0(6):85-91.
作者姓名:王波  陈东东  张锦霞  张之琛  马星星  张志宏
作者单位:(1.国网陕西省电力公司信息通信公司,陕西西安 710065;2.上海交通大学生物医学工程学院,上海 200030;3.厦门大学信息学院,福建厦门 361005)
摘    要:针对传统方法难以对不间断电源系统、微网储能系统的蓄电池组进行高效准确的状态估计问题,提出了一种全新的基于数据驱动的方法应用于大规模电池的健康状态预测。首先,分析电池的历史数据,从电池健康状态(空间维度)和时间演变放电过程(时间维度)2个层面对电池状态进行有效聚类。其次,将数据分布映射到高维空间,设计相应的基于时空分布映射的电池健康状态预测深度神经网络模型。最后,实验模型验证了所提方法的有效性。

关 键 词:电池健康状态预测  深度神经网络  多尺度聚类  数据驱动

Large-scale Battery Health State Prediction Based on Spatio-temporal Distribution Mapping
WANG Bo,CHEN Dongdong,ZHANG Jinxia,ZHANG Zhichen,MA Xingxing,ZHANG Zhihong.Large-scale Battery Health State Prediction Based on Spatio-temporal Distribution Mapping[J].Shanxi Electric Power,2022,0(6):85-91.
Authors:WANG Bo  CHEN Dongdong  ZHANG Jinxia  ZHANG Zhichen  MA Xingxing  ZHANG Zhihong
Affiliation:(1. State Grid Shaanxi Information and Telecommunication Company,Xian 710065,China;2. School of Biomedical Engineering, Shanghai Jiao Tong University,Shanghai 20030,China;3. School of Information,Xiamen University,Xiamen 361005,China)
Abstract:Efficient and accurate state estimation of storage battery is of great significance for the normal operation of UPS system and microgrid energy storage system. A new data-driven method is proposed to predict the health state of large-scale batteries. Firstly, by analyzing the large-scale historical data of battery,the battery state is effectively clustered from the two levels of battery health state (spatial dimension) and time evolution discharge process (time dimension). Then,by mapping data distribution to higher dimensional space, the corresponding deep neural network model for battery health state prediction based on spatial-temporal distribution mapping is designed. Experimental results demonstrate the effectiveness of proposed method.
Keywords:battery health status prediction  deep neural network  multiscale cluster  data driven
点击此处可从《陕西电力》浏览原始摘要信息
点击此处可从《陕西电力》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号