一种基于少量温度传感器的超级电容模组温度监测方法

doi:10.19799/j.cnki.2095-4239.2022.0234

摘要/Abstract

摘要：

超级电容器服役性能及老化过程与温度密切相关，温度过高会引起热失控从而影响超级电容运行安全，因此监测超级电容系统中每只单体温度尤为重要，而传统的传感器监测方案存在成本高、不易安装等问题。本工作以商用超级电容模组为研究对象，提出了一种基于少量单体温度预估模组内剩余单体温度的方法，可以减少传感器的使用。通过研究不同冷却风速下、多段恒流充放电时模组内各单体的温度数据，发现单体温度间具有强相关性。建立基于BP神经网络的模组温度预估模型，通过对比不同单体组合作为输入，并分别移除电流、电压、风速等因素后的预训练效果，确定了最佳的模型架构。通过对比不同算法在同一数据集上的预测效果，选定了Levenberg-Marquardt作为模型的训练算法。模型可实现通过3只单体温度预估剩余9只单体温度，测试数据集上的总体平均绝对误差为0.06 ℃，最大绝对误差在0.30 ℃以内，满足储能系统对温度监测精度的要求。该方法所需测试条件简单，同时能够降低温度传感器购置成本，为超级电容热管理系统的温度监测提供了一种新的方法。

关键词: 超级电容模组, 温度预估, 神经网络, 温度监测

Abstract:

Supercapacitors' service performance and the aging process are closely related to temperature. Excessive temperature will cause thermal runaway and affect the operation safety of supercapacitors. Therefore, it is imperative to monitor the temperature of each unit in the supercapacitor system. However, the conventional sensor monitoring scheme has high costs and complex installation issues. With a commercial supercapacitor module as the research object, this paper suggests a method to estimate the temperature of residual cells in the module according to the temperature of a small number of cells, which can reduce the use of sensors. It is found that there is a strong correlation between cell temperatures by analyzing the temperature data of each cell in the module under various cooling wind speeds with multi-stage constant current charging and discharging. A module temperature estimation model according to BP neural network is established. The best model architecture is determined by comparing the pre-training effects of different cell combinations as input and removing current, voltage, wind speed, and other factors. By comparing the estimation impacts of various algorithms on the same data set, Levenberg Marquardt is chosen as the training algorithm of the model. The model can estimate the temperature of the remaining 9 cells through the temperature of 3 cells. The test data set's average absolute error is 0.06 ℃, and the maximum absolute error is within 0.3 ℃, which meets the energy storage system's requirements for temperature monitoring precision. The suggested method requires simple test conditions and can reduce the purchase cost of temperature sensors, which provides a new method for temperature monitoring of supercapacitor thermal management systems.

Key words: supercapacitor module, temperature estimation, neural network, temperature monitoring

中图分类号:

TM 911

韦莉, 黄雪林, 张婉婷, 白欣桐. 一种基于少量温度传感器的超级电容模组温度监测方法[J]. 储能科学与技术, 2022, 11(11): 3631-3640.

Li WEI, Xuelin HUANG, Wanting ZHANG, Xintong BAI. A temperature monitoring method of supercapacitor module based on a small number of temperature sensors[J]. Energy Storage Science and Technology, 2022, 11(11): 3631-3640.

图/表 13

表1

图1

图2

图3

图4

图5

图6

表2

图7

表3

表4

图8

表5

参考文献 28

1	曹广华, 高佶, 高洁, 等. 超级电容的原理及应用[J]. 自动化技术与应用, 2016, 35(5): 131-135.
	CAO G H, GAO J, GAO J, et al. Principle and application of super capacitor[J]. Techniques of Automation and Applications, 2016, 35(5): 131-135.
2	韩冬林, 徐琤颖, 陈愚. 基于超级电容的后备式UPS系统设计[J]. 电源技术, 2016, 40(10): 2036-2039.
	HAN D L, XU Z Y, CHEN Y. Design of backup UPS system based on super capacitors[J]. Chinese Journal of Power Sources, 2016, 40(10): 2036-2039.
3	王民华, 李凤霞. 混合储能平抑微电网功率波动控制策略研究[J]. 电力电容器与无功补偿, 2020, 41(4): 215-220.
	WANG M H, LI F X. Study on control strategy of hybrid energy storage used in stabilization power fluctuation of microgrid[J]. Power Capacitor ＆ Reactive Power Compensation, 2020, 41(4): 215-220.
4	华黎, 杨恩东, 梁全顺, 等. 车用超级电容器动力系统在世博公交的应用及推广[C]. 中国智能交通协会, 2011: 9.
	HUA L, YANG E D, LIANG Q S, et al. Application and promotion of automotive supercapacitor power system in expo bus[C]. China Intelligent Transportation Association, 2011: 9.
5	芮学宝, 朱刚阳. 电容储能型再生电能吸收装置在地铁系统中的应用[J]. 电气化铁道, 2020, 31(S1): 195-198.
	RUI X B, ZHU G Y. Application of capacitive energy storage regenerative energy absorption device in subway system[J]. Electric Railway, 2020, 31(S1): 195-198.
6	陈朝晖, 张伟先, 黄钰强. 48V超级电容模组的研究[J]. 技术与市场, 2015, 22(5): 68-69.
	CHEN C H, ZHANG W X, HUANG Y Q. Research on 48V supercapacitor module[J]. Technology and Market, 2015, 22(5): 68-69.
7	陈书礼, 韩金磊, 荣常如, 等. 温度和电压对超级电容器单体内阻影响的研究[J]. 汽车技术, 2016(3): 40-44.
	CHEN S L, HAN J L, RONG C R, et al. Influence of temperature and voltage on internal resistance of supercapacitor[J]. Automobile Technology, 2016(3): 40-44.
8	KÖTZ R, HAHN M, GALLAY R. Temperature behavior and impedance fundamentals of supercapacitors[J]. Journal of Power Sources, 2006, 154(2): 550-555.
9	ZHOU Y T, WANG Y N, WANG K, et al. Hybrid genetic algorithm method for efficient and robust evaluation of remaining useful life of supercapacitors[J]. Applied Energy, 2020, 260: doi: 10.1016/j.apenergy.2019.114169.
10	FLETCHER S I, SILLARS F B, CARTER R C, et al. The effects of temperature on the performance of electrochemical double layer capacitors[J]. Journal of Power Sources, 2010, 195(21): 7484-7488.
11	顾帅, 韦莉, 张逸成, 等. 超级电容器不一致性研究现状及展望[J]. 中国电机工程学报, 2015, 35(11): 2862-2869.
	GU S, WEI L, ZHANG Y C, et al. Review of nonuniformity research and analysis on supercapacitor[J]. Proceedings of the CSEE, 2015, 35(11): 2862-2869.
12	Q31/0115000918C005—2018, 船用超级电容器系统[S]. 上海: 上海奥威科技开发有限公司, 2018.
	Q31/0115000918C005—2018, Marine supercapacitor system[S]. Shanghai: Shanghai Aowei Technology Development Co., LTD, 2018.
13	VOICU I, LOUAHLIA H, GUALOUS H, et al. Thermal management and forced air-cooling of supercapacitors stack[J]. Applied Thermal Engineering, 2015, 85: 89-99.
14	FRIVALDSKY M, CUNTALA J, SPANIK P. Simple and accurate thermal simulation model of supercapacitor suitable for development of module solutions[J]. International Journal of Thermal Sciences, 2014, 84: 34-47.
15	戴朝华, 傅雪婷, 杜云, 等. 不同空间结构的有轨电车超级电容热行为[J]. 西南交通大学学报, 2020, 55(5): 920-927, 900.
	DAI C H, FU X T, DU Y, et al. Supercapacitor thermal behavior of trams with different spatial structures[J]. Journal of Southwest Jiaotong University, 2020, 55(5): 920-927, 900.
16	郭文强, 董瑶, 李清华, 等. PSO-BP神经网络在开关柜设备温度预测中的应用[J]. 陕西科技大学学报, 2020, 38(1): 149-153.
	GUO W Q, DONG Y, LI Q H, et al. Application of PSO-BP neural network in temperature prediction for switchgear equipment[J]. Journal of Shaanxi University of Science ＆ Technology, 2020, 38(1): 149-153.
17	赵明珠, 王丹, 方杰, 等. 基于LSTM神经网络的地铁车站温度预测[J]. 北京交通大学学报, 2020, 44(4): 94-101.
	ZHAO M Z, WANG D, FANG J, et al. Prediction of subway station temperature based on LSTM neural network[J]. Journal of Beijing Jiaotong University, 2020, 44(4): 94-101.
18	贾永会, 杜建桥, 汪潮洋, 等. 基于BP神经网络的燃煤锅炉温度分布预测[J]. 热能动力工程, 2020, 35(7): 130-138.
	JIA Y H, DU J Q, WANG C Y, et al. Prediction model of temperature distribution in combustion zone of coal-fired boiler based on BP neural network[J]. Journal of Engineering for Thermal Energy and Power, 2020, 35(7): 130-138.
19	FANG K Z, MU D B, CHEN S, et al. A prediction model based on artificial neural network for surface temperature simulation of nickel-metal hydride battery during charging[J]. Journal of Power Sources, 2012, 208: 378-382.
20	LIU K L, LI K, DENG J. A novel hybrid data-driven method for li-ion battery internal temperature estimation[C]//2016 UKACC 11th International Conference on Control (CONTROL). Belfast, UK. IEEE : 1-6.
21	HASAN M M, POURMOUSAVI S A, JAHANBANI ARDAKANI A, et al. A data-driven approach to estimate battery cell temperature using a nonlinear autoregressive exogenous neural network model[J]. Journal of Energy Storage, 2020, 32: doi:10.1016/j.est.2020.101879.
22	KIM T J, YOUN B D, KIM H J. Online-applicable temperature prediction model for EV battery pack thermal management[C]. ASME 2013 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference. 2013.
23	BO Z, LI H W, YANG H C, et al. Combinatorial atomistic-to-AI prediction and experimental validation of heating effects in 350 F supercapacitor modules[J]. International Journal of Heat and Mass Transfer, 2021, 171: doi:10.1016/j.ijheatmasstransfer.2021.121075.
24	赵崇文. 人工神经网络综述[J]. 山西电子技术, 2020(3): 94-96.
	ZHAO C W. A survey on artificial neural networks[J]. Shanxi Electronic Technology, 2020(3): 94-96.
25	RUMELHART D E, HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors[J]. Nature, 1986, 323(6088): 533-536.
26	HORNIK K, STINCHCOMBE M, WHITE H. Multilayer feedforward networks are universal approximators[J]. Neural Networks, 1989, 2(5): 359-366.
27	SUTSKEVER I, MARTENS J, DAHL G, HINTON G. On the importance of initialization and momentum in deep learning[C]. 30th International Conference on Machine Learning, ICML 2013, 1139-1147.
28	YADAV S, SHUKLA S. Analysis of k-fold cross-validation over hold-out validation on colossal datasets for quality classification[C]//2016 IEEE 6th International Conference on Advanced Computing. Bhimavaram, India. IEEE : 78-83.

技术指标	数值
额定工作电压区间/V	2.5～4.0
截止电压/V	2.5
额定充放电电流/A	50
最大充放电电流/A	100
标称容值/F	34000
静电容量/Ah	15
储存能量/Wh	54
直流内阻/mΩ	0.7
工作温度/℃	-25～55

模型代号	相较于NNv1的操作	MAE/℃
NNv1	无	0.0651
NNv2	移除电流	0.1290
NNv3	移除电压	0.0601
NNv4	移除风速	0.1169
NNv5	移除Tin_k-1	0.0851
NNv6	移除全部温度	0.8366
NNv7	移除环境温度	0.1091

算法	MAE/℃
Levenberg-Marquardt	0.0601
动量法	1.7444
拟牛顿法	0.0848
学习率可变的BP算法	0.5318

单体	最大绝对误差/℃	平均绝对误差/℃
#1	0.2496	0.0639
#2	0.2396	0.0650
#3	0.1682	0.0520
#5	0.1671	0.0575
#7	0.2225	0.0529
#9	0.1703	0.0442
#10	0.2807	0.0693
#11	0.2908	0.0749
#12	0.2234	0.0604

[1]	黄鹏, 聂枝根, 陈峥, 舒星, 沈世全, 杨继鹏, 申江卫. 基于优化Elman神经网络的锂电池容量预测[J]. 储能科学与技术, 2022, 11(7): 2282-2294.
[2]	耿萌萌, 范茂松, 杨凯, 赵光金, 谭震, 高飞, 张明杰. 基于EIS和神经网络的退役电池SOH快速估计[J]. 储能科学与技术, 2022, 11(2): 673-678.
[3]	王鲁, 王峰, 徐竞, 赵延鹏, 李玮, 王艳艳, 王应彪. 基于SOM+SVM的退役锂离子电池分选[J]. 储能科学与技术, 2022, 11(11): 3623-3630.
[4]	张少凤, 张清勇, 杨叶森, 苏义鑫, 熊斌宇. 基于滑动窗口和LSTM神经网络的锂离子电池建模方法[J]. 储能科学与技术, 2022, 11(1): 228-239.
[5]	王帅, 马鸿雁, 窦嘉铭, 张英达, 李晟延, 胡璐锦. 基于UGOA-BP的锂电池SOC估算[J]. 储能科学与技术, 2022, 11(1): 258-264.
[6]	王凡, 史永胜, 刘博亲, 左玉洁, 符政, ALI Jamsher. 基于注意力改进BiGRU的锂离子电池健康状态估计[J]. 储能科学与技术, 2021, 10(6): 2326-2333.
[7]	李志浩, 彭浩, 陈亚琴. 质子交换膜燃料电池膜电极组件温度分布的神经网络预测模型[J]. 储能科学与技术, 2021, 10(6): 2053-2059.
[8]	李练兵, 李思佳, 李洁, 孙坤, 王正平, 杨海跃, 高冰, 杨少波. 基于差分电压和Elman神经网络的锂离子电池RUL预测方法[J]. 储能科学与技术, 2021, 10(6): 2373-2384.
[9]	曹新宇, 彭飞, 李立伟, 尹建光. 基于IBAS-NARX神经网络的锂电池荷电状态估计[J]. 储能科学与技术, 2021, 10(6): 2342-2351.
[10]	张远进, 吴华伟, 叶从进. 基于AUKF-BP神经网络的锂电池SOC估算[J]. 储能科学与技术, 2021, 10(1): 237-241.
[11]	陈峥, 李磊磊, 舒星, 沈世全, 刘永刚, 申江卫. 基于特征处理与径向基神经网络的锂电池剩余容量估算方法[J]. 储能科学与技术, 2021, 10(1): 261-270.
[12]	王泰华, 张书杰, 陈金干. 基于BP-PSO算法的锂电池低温充电策略优化[J]. 储能科学与技术, 2020, 9(6): 1940-1947.
[13]	张新锋, 姚蒙蒙, 王钟毅, 饶勇翔. 基于ACO-BP神经网络的锂离子电池容量衰退预测[J]. 储能科学与技术, 2020, 9(1): 138-144.
[14]	陈德海, 马原, 潘韦驰. 改进PSO-RBF模型的分阶查表法荷电状态估计[J]. 储能科学与技术, 2019, 8(6): 1190-1196.
[15]	苏振浩, 李晓杰, 秦晋, 杜文杰, 韩宁. 基于BP人工神经网络的动力电池SOC估算方法[J]. 储能科学与技术, 2019, 8(5): 868-873.