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

doi:10.19799/j.cnki.2095-4239.2021.0118

储能科学与技术 ›› 2021, Vol. 10 ›› Issue (6): 2053-2059.doi: 10.19799/j.cnki.2095-4239.2021.0118

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

李志浩(), 彭浩(), 陈亚琴

上海海事大学航运仿真技术教育部工程研究中心，上海 201306

收稿日期:2021-03-22 修回日期:2021-05-10 出版日期:2021-11-05 发布日期:2021-11-03
作者简介:李志浩（1997—），男，硕士研究生，研究方向为人工神经网络与传热，E-mail：lizhihao2018@gmail.com|彭浩，副教授，研究方向为微纳尺度传热、相变传热，E-mail：hpeng@shmtu.edu.cn。
基金资助:
上海市自然科学基金项目(19ZR1422300)

Neural network prediction model for temperature distribution of proton exchange membrane fuel cell membrane electrode assembly

Zhihao LI(), Hao PENG(), Yaqin CHEN

Engineering Research Center of Shipping Simulation (Ministry of Education), Shanghai Maritime University, Shanghai 201306, China

Received:2021-03-22 Revised:2021-05-10 Online:2021-11-05 Published:2021-11-03

摘要/Abstract

摘要：

质子交换膜燃料电池膜电极组件表面的温度分布会影响质子交换膜燃料电池的性能、寿命和可靠性。为探究质子交换膜燃料电池传热规律，本文提出了一种基于神经网络的质子交换膜燃料电池膜电极组件温度分布的预测模型。本研究选取径向基函数神经网络(RBF)和广义回归神经网络(GRNN)两种神经网络，以电流密度、温度点的位置作为网络输入，不同位置的温度作为网络输出，对平行流道质子交换膜燃料电池、蛇形流道质子交换膜燃料电池分别建立了神经网络预测模型。结果显示，RBF神经网络预测的均方根误差平均为0.464、平均绝对百分误差为1.179%，GRNN神经网络预测的均方根误差平均为0.7155、平均绝对百分误差为2.27%；相较于GRNN神经网络，RBF神经网络精度更高；基于RBF神经网络的平行流道质子交换膜燃料电池膜电极组件温度分布预测模型预测值与96%的实验值的相对误差在5%以内。基于RBF神经网络的蛇形流道质子交换膜燃料电池膜电极组件温度分布预测模型预测值与95%的实验值的相对误差在5%以内。

关键词: 质子交换膜燃料电池, 膜电极温度分布, 人工神经网络

Abstract:

The temperature distribution on the surface of the membrane electrode assembly of the proton exchange membrane fuel cell will affect the performance, life and reliability of proton exchange membrane fuel cell. In order to investigate the heat transfer of proton exchange membrane fuel cell, a prediction model of the temperature distribution of the membrane electrode assembly based on neural network is proposed. This study selected the radial basis function (RBF) neural network and generalized regression neural network (GRNN), two kinds of neural network, with the location of the current density and temperature point as network input, the different position as network output, the temperature of the parallel port of proton exchange membrane fuel cell, serpentine flow proton exchange membrane fuel cell neural network prediction model is established, respectively. The results showcase the average root mean square error of RBF neural network prediction is 0.464, the average absolute percentage error is 1.179%, the average root mean square error of GRNN neural network prediction is 0.7155, the average absolute percentage error is 2.27%. Compared with GRNN neural network, RBF neural network has higher accuracy. The relative error between the predicted value and the experimental value of 96% for the temperature distribution prediction model based on RBF neural network is within 5%. The relative error between the predicted value and the 95% experimental value of the temperature distribution prediction model based on RBF neural network for the membrane electrode assembly of PEMFC is less than 5%.

Key words: proton exchange membrane fuel cell, membrane electrode temperature distribution, artificial neural network

中图分类号:

TM 911.4

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.

Zhihao LI, Hao PENG, Yaqin CHEN. Neural network prediction model for temperature distribution of proton exchange membrane fuel cell membrane electrode assembly[J]. Energy Storage Science and Technology, 2021, 10(6): 2053-2059.

图/表 10

参考文献 15

1	FAGHRI A, GUO Z. Challenges and opportunities of thermal management issues related to fuel cell technology and modeling[J]. International Journal of Heat and Mass Transfer, 2005, 48(19/20): 3891-3920.
2	ABDULLAH A M, OKAJIMA T, MOHAMMAD A M, et al. Temperature gradients measurements within a segmented H2/air PEM fuel cell[J]. Journal of Power Sources, 2007, 172(1): 209-214.
3	WANG M H, GUO H, MA C F, et al. Temperature measurement technologies and their application in the research of fuel cells[C]//Proceedings of ASME 2003 1st International Conference on Fuel Cell Science, Engineering and Technology, April 21-23, 2003, Rochester, New York, USA. 2009: 95-100.
4	YU S, JUNG D. Thermal management strategy for a proton exchange membrane fuel cell system with a large active cell area[J]. Renewable Energy, 2008, 33(12): 2540-2548.
5	PARK J, LI X G, TRAN D, et al. Neutron imaging investigation of liquid water distribution in and the performance of a PEM fuel cell[J]. International Journal of Hydrogen Energy, 2008, 33(13): 3373-3384.
6	葛睿彤, 郑艺华. 燃料电池传热传质分析进展综述[J]. 储能科学与技术, 2020, 9(1): 40-56.
	GE R T, ZHENG Y H. Summary of progress in heat and mass transfer analysis of fuel cells[J]. Energy Storage Science and Technology, 2020, 9(1): 40-56.
7	高一方, 陈唐龙, 杨梓杰, 等. 质子交换膜燃料电池温度经验模型[J]. 电源技术, 2018, 42(1): 55-57,64.
	GAO Y F, CHEN T L, YANG Z J, et al. Temperature empirical models of proton exchange membrane fuel cell[J]. Chinese Journal of Power Sources, 2018, 42(1): 55-57,64.
8	马天才, 孙泽昌, 许思传. 质子交换膜燃料电池温度控制仿真模型[J]. 系统仿真学报, 2005, 17(3): 548-551.
	MA T C, SUN Z C, XU S C. Simulation model of PEMFC temperature control[J]. Acta Simulata Systematica Sinica, 2005, 17(3): 548-551.
9	CHOI E J, PARK J Y, KIM M S. A comparison of temperature distribution in PEMFC with single-phase water cooling and two-phase HFE-7100 cooling methods by numerical study[J]. International Journal of Hydrogen Energy, 2018, 43(29): 13406-13419.
10	GUO H, WANG M H, YE F, et al. Experimental study of temperature distribution on anodic surface of MEA inside a PEMFC with parallel channels flow bed[J]. International Journal of Hydrogen Energy, 2012, 37(17): 13155-13160.
11	WANG M H, GUO H, MA C F. Temperature distribution on the MEA surface of a PEMFC with serpentine channel flow bed[J]. Journal of Power Sources, 2006, 157(1): 181-187.
12	李远凯. 基于RBF神经网络的挖掘机工装轨迹控制研究[D]. 西安: 西安科技大学, 2020.
	LI Y K. Research on trajectory control of excavator working device based on RBF neural network[D]. Xi'an: Xi'an University of Science and Technology, 2020.
13	杨娟, 杨占刚. 基于GRNN算法的飞机用电设备非侵入式负荷监测方法[J]. 航空学报, 2021, 42(3): 403-413.
	YANG J, YANG Z G. Non-intrusive load monitoring method for aircraft electrical equipment based on GRNN algorithm[J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(3): 403-413.
14	李萌. 基于GRNN神经网络的科技型企业价值评估方法研究[D]. 哈尔滨: 哈尔滨理工大学, 2020.
	LI M. Research on value evaluation method of science and technology enterprise based on GRNN neural network[D]. Harbin: Harbin University of Science and Technology, 2020.
15	文旭林, 王辉涛, 王华, 等. 基于RBF神经网络的水平光管内R407C流动沸腾换热预测[J]. 昆明理工大学学报(自然科学版), 2012, 37(4): 57-62.
	WEN X L, WANG H T, WANG H, et al. Prediction of flow boiling heat transfer of R407C inside horizontal smooth tubes based on RBF neural network[J]. Journal of Kunming University of Science and Technology (Natural Science Edition), 2012, 37(4): 57-62.

[1]	胡冶州, 王双, 申涛, 朱叶, 王得丽. 限域型贵金属氧还原反应电催化剂研究进展[J]. 储能科学与技术, 2022, 11(4): 1264-1277.
[2]	陈曦, 贺凌轩, 刘芹孝, 方叶, 龙施淳, 万忠民. 动态工况下车用燃料电池系统热力学分析[J]. 储能科学与技术, 2021, 10(4): 1416-1422.
[3]	张敬, 卢雁, 李圣, 谢光彩, 万忠民. 基于模糊PID控制的家用燃料电池热电联供系统建模与仿真[J]. 储能科学与技术, 2021, 10(3): 1117-1126.
[4]	翟俊香, 何广利, 许壮, 刘聪敏. 空冷型质子交换膜燃料电池系统效率的实验研究[J]. 储能科学与技术, 2020, 9(6): 1885-1889.
[5]	裴冯来. 典型30 kW级质子交换膜燃料电池发动机性能测评对比研究[J]. 储能科学与技术, 2018, 7(3): 519-523.
[6]	朱晓舟，陈民武，刘湘东，赵航飞，韩明. 空冷型PEMFC阳极排气方法实验研究[J]. 储能科学与技术, 2018, 7(1): 118-.
[7]	朱晓舟，陈民武，刘湘东，赵航飞，韩明. 基于LabVIEW的PEMFC单电池电压巡检系统设计[J]. 储能科学与技术, 2018, 7(1): 123-.

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

Neural network prediction model for temperature distribution of proton exchange membrane fuel cell membrane electrode assembly

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 15

相关文章 7

编辑推荐

Metrics

本文评价