|本期目录/Table of Contents|

[1]李德鑫,吕项羽,王佳蕊,等.基于熵权法和Elman神经网络相结合的储能系统SOC估计[J].电子设计工程,2020,28(01):70-74.[doi:10.14022/j.issn1674-6236.2020.01.016]
 LI Dexin,LV Xiangyu,WANG Jiarui,et al.SOC estimation of energy storage systems based on neural network integrated with entropy weight method[J].Electronic Design Engineering,2020,28(01):70-74.[doi:10.14022/j.issn1674-6236.2020.01.016]
点击复制

基于熵权法和Elman神经网络相结合的储能系统SOC估计(PDF)
分享到:

《电子设计工程》[ISSN:1674-6236/CN:61-1477/TN]

卷:
28
期数:
2020年01期
页码:
70-74
栏目:
测量与控制
出版日期:
2020-01-05

文章信息/Info

Title:
SOC estimation of energy storage systems based on neural network integrated with entropy weight method
文章编号:
1674-6236(2020)01-0070-05
作者:
李德鑫1吕项羽1王佳蕊1刘星宇2韩晓娟2
(1. 国网吉林省电力有限公司电力科学研究院 吉林 长春 130021;2.华北电力大学 控制与计算机工程学院,北京 102206)
Author(s):
LI De?xin1LV Xiang?yu1 WANG Jia?rui1 LIU Xing?yu2 HAN Xiao?juan2
(1. Electric Power Research Institute, State Grid Jilin Electric Power Co., Ltd., Changchun 130021,China;2. School of Control and Computer Engineering, North China Electric University, Beijing 102206, China)
关键词:
储能系统 SOC估计 熵权法 MEA-Elman神经网络
Keywords:
energy storage systems SOC estimation entropy weight method MEA-Elman neural network
分类号:
TM912
DOI:
10.14022/j.issn1674-6236.2020.01.016
文献标志码:
A
摘要:
电池荷电状态(State of Charge, SOC),也称电池剩余电量,是表征电池特性的关键参数。该文提出基于熵权法(Entropy Weight Method, EWM)和Elman神经网络相结合的储能系统SOC状态估计方法。针对神经网络输入参数选择通常依靠经验的问题,使用熵权法计算储能系统运转参数的权重并排序,形成待评价特征集。通过Elman神经网络对不同输入特征参数进行仿真测试,挑选功效最优的一组特征集作为Elman神经网络的输入,使用思维进化算法(Mind Evolutionary Algorithm, MEA)优化Elman神经网络初始权值和阈值,建立了基于EWM-MEA-Elman神经网络的储能系统SOC状态估计模型。经全钒液流电池充放电实测数据仿真数据表明,正确选择输入特征集可以有效提高预测精度和训练速度。该仿真结果验证了本文神经网络模型的效果和准确性,为储能系统的应用提供了优秀的理论支持。
Abstract:
State of charge (SOC) of the battery is a key parameter to characterize the performance of the battery, a SOC state estimation method of energy storage systems based on the entropy weight method (EWM) and Elman neural network is proposed in this paper. Because the selection of the input parameters for neural network usually depends on the experience, the entropy weight method is used to calculate the weights of the operating parameters for the energy storage systems and sort them to form the feature set to be evaluated. The Elman neural network is used to simulate different input characteristic parameters, and a set of optimal feature sets is selected as the input of Elman neural network. The initial weight and threshold of the Elman neural network are optimized by Mind Evolutionary Algorithm (MEA). The SOC state estimation model of energy storage systems based on the EWM-MEA-Elman neural network is established. The simulation analysis of the charge and discharge data of the all-vanadium flow battery shows that the correct selection of the input feature set can effectively improve the prediction accuracy and training speed. The simulation example verifies the validity and correctness of the proposed method and provides a theoretical basis for large-scale energy storage applications.

参考文献/References:

[1] 林娅,陈则王.锂离子电池剩余寿命预测研究综述[J].电子测量技术,2018,41(4):29-35.[2] 范兴明,曾求勇,张鑫.基于改进安时积分法的电动汽车电池SOC估计与仿真研究[J].电气应用,2015,34(8):111-115.[3] 付浪,杜明星,刘斌,等.基于开路电压法与卡尔曼滤波法相结合的锂离子电池SOC估算[J].天津理工大学学报,2015,31(6):9-13. [4] 樊波,栾新宇,张瑞,等.基于改进PNGV模型的电池SOC估计算法研究[J].电测与仪表,2018,55(20):46-51.[5] 方磊,陈勇,赵理,等. 基于模糊控制的扩展卡尔曼滤波SOC估计研究[J]. 系统仿真学报, 2018, 30(1):325-331.[6] 杨春生,牛红涛,隋良红,等. 基于贝叶斯正则化算法BP神经网络钒电池SOC预测[J]. 现代电子技术, 2016,39(8):158-161.[7] 周美兰,王吉昌,李艳萍. 优化的BP神经网络在预测电动汽车SOC上的应用[J]. 黑龙江大学自然科学学报, 2015,32(1):129-134.[8] Liu F, Liu T, Fu Y. An improved SoC estimation algorithm based on artificial neural network[C]//International Symposium on Computational Intelligence and Design. IEEE, 2016:152-155.[9] Sun B, Wang L. The SOC estimation of NIMH battery pack for HEV based on BP neural network[C]//International Workshop on Intelligent Systems and Applications. IEEE, 2009:1-4.[10]周美兰,赵强,周永勤. 改进的PSO-BP神经网络估算磷酸铁锂电池SOC[J]. 哈尔滨理工大学学报, 2015,20(4):88-92.[11]赵钢,朱芳欣,窦汝振. 基于PSO-BP的电动汽车锂离子电池SOC估算[J]. 电源技术,2018,42(9):62-64.[12]Cai C, Du D, Liu Z, et al. State-of-charge (SOC) estimation of high power Ni-MH rechargeable battery with artificial neural network[C]//International Conference on Neural Information Processing. IEEE, 2003(2):824-828.[13]马纪,刘希喆. 基于序关系-熵权法的低压配网台区健康状态评估[J]. 电力系统保护与控制, 2017,45(6):87-93.[14]杨旭. 基于物元分析法的堤防工程管理评价模型及应用[J]. 水土保持应用技术, 2018,183(3):21-23.[15]赵希梅,金鸿雁.基于Elman神经网络的永磁直线同步电机互补滑模控制[J].电工技术学报,2018,33(5):973-979.[16]吴佳懋,李艳,符一健. 基于粗糙集-混沌时间序列Elman神经网络的短期用电量预测[J]. 电力系统保护与控制, 2019,47(3):29-36.

相似文献/References:

[1]张仰维.基于动态变系数的混合储能系统频率控制方法[J].电子设计工程,2020,28(05):27.[doi:10.14022/j.issn1674-6236.2020.05.006]
 ZHANG Yangwei.Frequency control method of hybrid energy storage system based on dynamic variable coefficient[J].Electronic Design Engineering,2020,28(01):27.[doi:10.14022/j.issn1674-6236.2020.05.006]
[2]王 磊.超级电容储能系统的变换器拓扑结构研究[J].电子设计工程,2018,26(08):129.
 WANG Lei.Research of converter topology based on super capacitor energy storage system[J].Electronic Design Engineering,2018,26(01):129.
[3]张 骁,胡忠平,时金林,等.家用光伏储能系统的设计与实现[J].电子设计工程,2017,(19):165.
 ZHANG Xiao,HU Zhong-ping,SHI Jin-lin,et al.Design and implementation of household photovoltaic energy storage system[J].Electronic Design Engineering,2017,(01):165.
[4]宋才华,蓝源娟,王永才,等.综合价值评估法在电力企业客户细分中的应用[J].电子设计工程,2014,(12):111.
 SONG Cai-hua,LAN Yuan-juan,WANG Yong-cai,et al.Application of synthetic value valuation method in power enterprise customer segmentation[J].Electronic Design Engineering,2014,(01):111.
[5]潘燕华,刘晨熙. 基于改进AHP和熵权法的预制节段梁生产运输方案研究[J].电子设计工程,2013,(14):45.
 PAN Yan-hua,LIU Chen-xi. Based on improved AHP and entropy weight method research of prefabricated segment girder’s production and transport[J].Electronic Design Engineering,2013,(01):45.

备注/Memo

备注/Memo:
收稿日期:2019-05-23 稿件编号:201905120基金项目:国家自然科学基金项目(51577065);国家重点研发计划资助项目(2017YFB0903505)作者简介:李德鑫(1985—),男,吉林长春人,硕士,高级工程师。研究方向:新能源并网技术与网源协调技术等。
更新日期/Last Update: 2019-12-30