基于改进CNN和信息融合的动力电池组故障诊断方法*

doi:10.19562/j.chinasae.qcgc.2020.11.011

摘要/Abstract

摘要： 本文中提出一种基于改进CNN和信息融合的锂离子电池组故障诊断方法。首先,在全连接层前加入Concat层,以融合不同层次的特征,建立改进的CNN模型。然后,采用MCE准则来优化交叉熵损失函数,解决其在非标签维梯度不做处理的问题。针对人工确定CNN卷积核结构存在的不够实用的问题,利用BIC准则确定模型最优卷积核结构。采用以上改进CNN进行诊断后,引入确诊条件进行判断。对不符合确诊条件的诊断结果,进一步采用一般CNN网络进行辅助诊断。通过将初步诊断结果和辅助诊断结果采用D-S证据理论进行融合的方法,得到最终的诊断结果。测试结果表明,本文中提出的方法可有效提高动力电池组的故障诊断准确率。

关键词: 锂离子电池组, 故障诊断, 卷积神经网络, BIC准则, 信息融合

Abstract: A fault diagnosis method for lithium-ion battery based on improved CNN and information fusion is proposed in this paper. Firstly, the Concat layer is added before the fully connected layer to fuse the features of different levels, so an improved CNN model is established. Then, the MCE criterion is used to optimize the cross entropy loss function, solving the problem that the non-label dimension gradient is not processed. In view of the insufficient practicality of the convolution kernel structure manually determined, the BIC criterion is used to determine the optimal structure of convolution kernel of the model. After diagnosis by using the improved CNN mentioned above, the conditions for confirmed diagnosis are introduced for judgment. For the diagnosis results not meeting the conditions, the general CNN network is further used for auxiliary diagnosis. By fusing preliminary diagnosis result and auxiliary diagnosis result using D-S evidence theory, the results of final diagnosis are obtained. The test results show that the method proposed can effectively enhance the accuracy of fault diagnosis on traction battery packs.

Key words: lithium-ion battery pack, fault diagnosis, convolutional neural network, BIC criterion, information fusion

彭运赛, 夏飞, 袁博, 王志成, 罗志疆. 基于改进CNN和信息融合的动力电池组故障诊断方法^*[J]. 汽车工程, 2020, 42(11): 1529-1536.

Peng Yunsai, Xia Fei, Yuan Bo, Wang Zhicheng, Luo Zhijiang. Fault Diagnosis of Traction Battery Pack Based on Improved Convolution Neural Network and Information Fusion[J]. Automotive Engineering, 2020, 42(11): 1529-1536.

参考文献

[1] 靳立强,孙志祥,刘志茹,等.不同温度下锂电池剩余电量估算的仿真研究[J]. 汽车工程,2019,41(5):590-598,606.
JIN L Q, SUN Z X, LIU Z R, et al. Simulation study on state of charge estimation of lithium-ion battery at different temperatures[J]. Automotive Engineering,2019, 41(5):590-598,606.
[2] 潘海鸿,吕治强,付兵,等.采用极限学习机实现锂离子电池健康状态在线估算[J].汽车工程,2017,39(12):1375-1381,1396.
PAN H H, LV Z Q, FU B, et al. Online estimation of lithium-ion battery’s state of health using extreme learning machine[J]. Automotive Engineering,2017,39(12):1375- 1381,1396.
[3] 檀斐.车用动力锂离子电池系统故障诊断研究与实现[D].北京:北京理工大学,2015.
TAN F. Fault diagnosis and implementation of electric vehicle lithium-ion battery system[D]. Beijing:Beijing Institute of Technology,2015.
[4] 古昂,张向文.基于RBF神经网络的动力电池故障诊断系统研究[J].电源技术,2016,40(10):1943-1945.
GU A, ZHANG X W. Fault diagnosis system for power battery based on RBF neural network[J]. Chinese Journal of Power Sources,2016,40(10):1943-1945.
[5] 夏飞,马茜,张浩,等.改进D-S证据理论在电动汽车锂电池故障诊断中的应用[J].智能系统学报,2017,12(4):526-537.
XIA F, MA X, ZHANG H, et al. Application of improved D-S evidence theory in fault diagnosis of lithium batteries in electric vehicles[J]. CAAI Transactions on Intelligent Systems,2017,12(4):526-537.
[6] 张倩,王建平,李帷韬.基于反馈机制的卷积神经网络绝缘子状态检测方法[J].电工技术学报,2019(16):3311-3321.
ZHANG Q, WANG J P, LI W T. Insulator state detection of convolutional neural networks based on feedback mechanism[J]. Transactions of China Electrotechnical Society,2019(16):3311-3321.
[7] 李恒,张氢,秦仙蓉,等.基于短时傅里叶变换和卷积神经网络的轴承故障诊断方法[J].振动与冲击,2018,37(19):124-131.
LI H, ZHANG Q, QIN X R, et al. Fault diagnosis method for rolling bearings based on short-time Fourier transform and convolution neural network[J]. Journal of Vibration and Shock,2018,37(19):124-131.
[8] 曲建岭,余路,袁涛,等.基于一维卷积神经网络的滚动轴承自适应故障诊断算法[J].仪器仪表学报,2018,39(7):134- 143.
QU J L, YU L, YUAN T, et al. Adaptive fault diagnosis algorithm for rolling bearings based on one-dimensional convolutional neural network[J]. Chinese Journal of Scientific Instrument,2018,39(7):134-143.
[9] 王远.基于区域卷积神经网络的物体检测方法的研究与改进[D].哈尔滨:.哈尔滨工业大学,2017.
WANG Y. Research and improvement of object detection based on region based convolutional neural networks[D]. Harbin:Harbin Institute of Technology,2017.
[10] 陈文强,熊辉,李克强,等.基于深度神经网络的行人及骑车人联合检测[J].汽车工程,2018,40(6):726-732,725.
CHEN W Q, XIONG H, LI K Q, et al. Concurrent pedestrian and cyclist detection based on deep neural networks [J]. Automotive Engineering,2018,40(6):726 -732,725.
[11] 余烨,金强,傅云翔,等.基于Fg-CarNet的车辆型号精细分类研究[J].自动化学报,2018,44(10):1864-1875.
YU Y, JIN Q, FU Y X, et al. Fine-grained classification of car models using Fg-CarNet convolutional neural network[J]. Acta Automatica Sinica,2018,44(10):1864-1875.
[12] 郑宗生,侯倩,邹国良,等.基于改进最小分类误差准则算法的深度学习研究——以台风卫星云图为例[J].计算机应用研究,2019,36(10):3160-3163.
ZHENG Z S, HOU Q, ZOU G L, et al. Research on deep learning based on improved minimal classification error criterion algorithm: take typhoon satellite image as example[J]. Application Research of Computers,2019,36(10):3160-3163.
[13] SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions[C]. Proc. of IEEE Computer Vision and Pattern Recognition,2015:1-9.
[14] JUANG B H, KATAGIRI S. Discriminative learning for minimum error classification [pattern recognition][J]. IEEE Trans. on Signal Processing, 1992, 40 (12): 3043-3054.
[15] FENG Z Y, SUN Z H, JIN L W. Learning deep neural network using max-margin minimum classification error [C]. Proc. of IEEE International Conference on Acoustics, Speech and Signal Processing,2016.
[16] 任超,吴伟,黄征凯,等.基于AIC准则的RBF神经网络在GPS高程拟合中的应用[J].测绘科学,2013,38(2):77-79.
REN C, WU W, HUANG Z K, et al. RBF neural network application in GPS height fitting based on AIC criterion[J]. Science of Surveying and Mapping,2013,38(2):77-79.
[17] 秦宣云,卜英勇,夏毅敏.基于AIC准则优化的径向神经网络微地形曲面重构[J].中南大学学报(自然科学版),2004(5):815-819.
QIN X Y, BU Y Y, XIA Y M. Surface reconstruction for micro-landform based on RBF neural network optimized by AIC criterion[J]. Journal of Central South University (Science and Technology),2004(5):815-819.
[18] 商云龙,张奇,崔纳新,等.基于AIC准则的锂离子电池变阶RC等效电路模型研究[J].电工技术学报,2015,30(17):55-62.
SHANG Y L, ZHANG Q, CUI N X, et al. Research on variable-order RC equivalent circuit model for lithium-ion battery based on the AIC criterion[J]. Transactions of China Electrotechnical Society,2015,30(17):55-62.
[19] 夏飞,袁博,彭道刚,等.基于信息量准则的锂离子电池变阶RC等效电路模型建模及优化方法[J].中国电机工程学报,2018,38(21):6441-6451,6506.
XIA F, YUAN B, PENG D G, et al. Modeling and optimization of variable-order RC equivalent circuit model for lithium-ion batteries based on information criterion[J]. Proceedings of the CSEE,2018,38(21):6441-6451,6506.
[20] 阮海斌.串联锂电池组均衡系统研究与设计[D]. 成都:西南交通大学,2017.
RUAN H B. Research and design of lithium-ion battery pack equalization system[D]. Chengdu:Southwest Jiaotong University,2017.

基于改进CNN和信息融合的动力电池组故障诊断方法^*

Fault Diagnosis of Traction Battery Pack Based on Improved Convolution Neural Network and Information Fusion

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 10

[1]	张鹏博, 陈仁祥, 邵毅明, 孙世政, 闫凯波. 纯电动汽车电驱动系统故障诊断研究进展[J]. 汽车工程, 2024, 46(1): 61-74.
[2]	管欣,仲昭辉,詹军,奚腾龙,叶昊,高深圳,成健,廖世辉,蔡均. 基于卷积神经网络的汽车操纵稳定性试验类型识别方法[J]. 汽车工程, 2023, 45(9): 1765-1771.
[3]	王明,唐小林,杨凯,李国法,胡晓松. 考虑预测风险的自动驾驶车辆运动规划方法[J]. 汽车工程, 2023, 45(8): 1362-1372.
[4]	赵东宇, 赵树恩. 基于级联YOLOv7的自动驾驶三维目标检测[J]. 汽车工程, 2023, 45(7): 1112-1122.
[5]	张雷, 关可人, 丁晓林, 郭鹏宇, 王震坡, 孙逢春. 基于图像识别与动力学融合的路面附着系数估计方法[J]. 汽车工程, 2023, 45(7): 1222-1234.
[6]	赵霞,李朝,付锐,葛振振,王畅. 基于深度卷积-Tokens降维优化视觉Transformer的分心驾驶行为实时检测[J]. 汽车工程, 2023, 45(6): 974-988.
[7]	洪吉超,梁峰伟,杨海旭,李克瑞. 大数据驱动动力电池智能安全管理与控制方法研究[J]. 汽车工程, 2023, 45(10): 1845-1861.
[8]	毕贵红,谢旭,蔡子龙,骆钊,陈臣鹏,赵鑫. 动态条件下基于深度学习的锂电池容量估计[J]. 汽车工程, 2022, 44(6): 868-878.
[9]	马彦,李佳怡,马乾,陈明超. 基于迭代动态规划的动力电池组热管理优化策略[J]. 汽车工程, 2022, 44(5): 709-721.
[10]	张奇祥,靳立强,靳博豪,张伊晗,陈鹏飞,刘永腾,李建华. EMB夹紧力控制与传感器故障诊断研究进展[J]. 汽车工程, 2022, 44(5): 736-746.
[11]	冯润泽,江昆,于伟光,杨殿阁. 基于两阶段分类算法的中国交通标志牌识别[J]. 汽车工程, 2022, 44(3): 434-441.
[12]	夏秋,陈特,陈龙,徐兴,蔡英凤. 基于冗余信息融合的车辆质心侧偏角估计方法[J]. 汽车工程, 2022, 44(2): 280-289.
[13]	于海,邓钧君,王震坡,孙逢春. 基于卷积神经网络的逆变器故障诊断方法[J]. 汽车工程, 2022, 44(1): 142-152.
[14]	夏祥腾,王大方,曹江,赵刚,张京明. 基于稀疏卷积神经网络的车载激光雷达点云语义分割方法[J]. 汽车工程, 2022, 44(1): 26-35.
[15]	施冬梅,肖锋. 基于改进长短时记忆网络的驾驶行为检测方法研究[J]. 汽车工程, 2021, 43(8): 1203-1209.