|
|
|
| Seq2Seq prediction method of state of charge of electric bus battery |
Hong-zhao DONG( ),Zhen WANG,Nan ZHANG,Yi-ni SHE,Ying-ying LIN |
| Joint Institute of Intelligent Transportation System, College of Mechanical Engineering, Zhejiang University of Technology, Hangzhou 310014, China |
|
|
|
Abstract A Seq2Seq prediction method for the state of charge of electric bus batteries was proposed to address the difficulty of predicting the state of charge under high-dimensional input features and long-term prediction needs. The battery state prediction was enhanced by incorporating vehicle driving status and operating conditions. The impact of various factors on the state of charge was analyzed using feature selection algorithms. The WN-Seq2Seq algorithm integrated Seq2Seq and WaveNet cyclic structures, which could enhance the memory and representation ability of high-dimensional input features and predicted state of charge sequence information. The accuracy of the model's prediction was improved. Validation using actual driving data from four electric buses in Hangzhou from 2021 to 2022 demonstrated the effectiveness of the WN-Seq2Seq model. The mean squared error (MSE) reached 0.505%, the mean absolute error (MAE) reached 0.479%, the mean absolute percentage error (MAPE) reached 0.656%, and the model computation time (MCT) reached 0.017 s. The prediction accuracy and stability were significantly promoted compared with traditional models. The WN-Seq2Seq model exhibited good prediction performance across different temperature ranges. The findings provide reliable decision-making support for energy consumption control strategies and safety management of electric buses.
|
|
Received: 06 November 2022
Published: 18 October 2023
|
|
|
| Fund: 国家自然科学基金资助项目（61773347）；浙江省公益技术研究项目（LGF20F030001） |
电动公交车电池荷电状态的Seq2Seq预测方法
针对高维输入特征和长时预测需求下荷电状态预测困难的问题，提出一种电动公交车电池荷电状态的Seq2Seq预测方法. 在电池状态的基础上引入车辆行驶状态和行驶工况，利用特征选择算法分析各因素对荷电状态的影响. 以LSTM为基本单元结构，WN-Seq2Seq算法融合Seq2Seq与WaveNet循环结构，可以强化高维输入特征与预测荷电状态的序列信息记忆与表征能力，从而提高模型的预测精度. 通过2021—2022年杭州市4辆电动公交车实际行驶数据验证表明，在引入车辆行驶状态和行驶工况后，WN-Seq2Seq模型的评价指标均方误差（MSE）、平均绝对误差(MAE)、平均绝对百分比误差(MAPE)和模型计算时间（MCT）分别为0.505%、0.479%、0.656%和0.017 s. 研究结果表明相比传统模型，预测精度及稳定性都有所提升，在不同温度区间下都具有良好的预测效果，为电动公交车能耗控制策略、安全管理提供合理且可靠的参数决策支持.
关键词：
电动公交车,
荷电状态预测,
深度学习,
序列到序列,
循环神经网络
|
|
| [1] |
MANZOLLI J A, TROVÃO J P, ANTUNES C H A review of electric bus vehicles research topics–methods and trends[J]. Renewable and Sustainable Energy Reviews, 2022, 159: 112211
doi: 10.1016/j.rser.2022.112211
|
|
|
| [2] |
朱丽群, 张建秋 一种联合锂电池健康和荷电状态的新模型[J]. 中国电机工程学报, 2018, 38 (12): 3613- 3620
ZHU Li-qun, ZHANG Jian-qiu A new model of jointed states of charge and health for lithium batteries[J]. Proceedings of the CSEE, 2018, 38 (12): 3613- 3620
|
|
|
| [3] |
常九健, 张煜帆 基于EMB的纯电动汽车制动能量回收优化控制策略研究[J]. 汽车工程, 2022, 44 (1): 64- 72
CHANG Jiu-jian, ZHANG Yu-fan Research on optimization control strategy for braking energy recovery of a battery electric vehicle based on EMB system[J]. Automotive Engineering, 2022, 44 (1): 64- 72
|
|
|
| [4] |
SOLDO J, ŠKUGOR B, DEUR J Online synthesis of an optimal battery state-of-charge reference trajectory for a plug-in hybrid electric city bus[J]. Energies, 2021, 14 (11): 3168
doi: 10.3390/en14113168
|
|
|
| [5] |
胡韵华, 冯瑾涛, 邓清闯, 等 电动汽车直流充电桩自动化测试平台的设计与应用[J]. 电力系统保护与控制, 2021, 49 (7): 150- 159
HU Yun-hua, FENG Jin-tao, DENG Qing-chuang, et al Development and application of automated test platform for DC charging piles of electric vehicles[J]. Power System Protection and Control, 2021, 49 (7): 150- 159
|
|
|
| [6] |
王义军, 左雪 锂离子电池荷电状态估算方法及其应用场景综述[J]. 电力系统自动化, 2022, 46 (14): 193- 207
WANG Yi-jun, ZUO Xue Review on estimation methods for state of charge of lithiumion battery and their application[J]. Automation of Electric Power Systems, 2022, 46 (14): 193- 207
|
|
|
| [7] |
来鑫, 李云飞, 郑岳久, 等 基于SOC-OCV优化曲线与EKF的锂离子电池荷电状态全局估计[J]. 汽车工程, 2021, 43 (1): 19- 26
LAI Xin, LI Yun-fei, ZHENG Yue-jiu, et al An overall estimation of state-of-charge based on SOC-OCV optimization curve and EKF for lithiumion battery[J]. Automotive Engineering, 2021, 43 (1): 19- 26
|
|
|
| [8] |
罗勇, 祁朋伟, 黄欢, 等 基于容量修正的安时积分SOC估算方法研究[J]. 汽车工程, 2020, 42 (5): 681- 687
LUO Yong, QI Peng-wei, HUANG Huan, et al Study on battery SOC estimation by ampere hour intergral method with capacity correction[J]. Automotive Engineering, 2020, 42 (5): 681- 687
|
|
|
| [9] |
寇发荣, 王甜甜, 王思俊, 等 基于ABC-RFEKF算法的锂电池SOC估计[J]. 电力系统保护与控制, 2022, 50 (4): 163- 171
KOU Fa-rong, WANG Tian-tian, WANG Si-jun, et al Lithium battery SOC estimation base on an ABC-RFEKF algorithm[J]. Power System Protection and Control, 2022, 50 (4): 163- 171
|
|
|
| [10] |
孙国强, 任佳琦, 成乐祥, 等 基于分数阶阻抗模型的磷酸铁锂电池荷电状态估计[J]. 电力系统自动化, 2018, 42 (23): 57- 63
SUN Guo-qiang, REN Jia-qi, CHENG Le-xiang, et al State of charge estimation of LiFePO4 battery based on fractional-order impedance model[J]. Automation of Electric Power Systems, 2018, 42 (23): 57- 63
|
|
|
| [11] |
孙金磊, 邹鑫, 顾浩天, 等 基于FFRLS-EKF联合算法的锂离子电池荷电状态估计方法[J]. 汽车工程, 2022, 44 (4): 505- 513
SUN Jin-lei, ZOU Xin, GU Hao-tian, et al State of charge estimation for Lithiumion battery based on FFRLS-EKF joint algorithm[J]. Automotive Engineering, 2022, 44 (4): 505- 513
|
|
|
| [12] |
SCHWUNK S, ARMBRUSTER N, STRAUB S, et al Particle filter for state of charge and state of health estimation for lithium-iron phosphate batterie[J]. Journal of Power Sources, 2013, 239: 705- 710
doi: 10.1016/j.jpowsour.2012.10.058
|
|
|
| [13] |
WU J, ZHANG C H, CUI N X Fuzzy energy management strategy for a hybrid electric vehicle based on driving cycle recognition[J]. International Journal of Automotive Technology, 2012, 13 (7): 1159- 1167
doi: 10.1007/s12239-012-0119-z
|
|
|
| [14] |
宋媛媛. 基于行驶工况的纯电动汽车能耗建模及续驶里程估算研究[D]. 北京: 北京交通大学, 2014: 11-13.
SONG Yuan-yuan. Energy consumption modeling and cruising range estimation based on driving cycle for electric vehicles [D]. Beijing: Beijing Jiaotong University, 2014: 11-13.
|
|
|
| [15] |
赵轩, 马建, 刘瑞, 等 基于GGAP-RBF神经网络的多参数纯电动客车蓄电池荷电状态预测[J]. 中国公路学报, 2015, 28 (4): 116- 126
ZHAO Xuan, MA Jian, LIU Rui, et al Multiple parameters state-of-charge estimation of battery for pure electric bus based on GGAP-RBF neural network[J]. China Journal of Highway and Transport, 2015, 28 (4): 116- 126
|
|
|
| [16] |
鲍伟, 葛建军 基于稀疏采样数据的电动公交车电池SOC预测方法研究[J]. 汽车工程, 2020, 42 (3): 367- 374
BAO Wei, GE Jian-jun Study on battery SOC prediction method for electric bus based on sparsely sampled data[J]. Automotive Engineering, 2020, 42 (3): 367- 374
|
|
|
| [17] |
WANG S, CAO J, CHEN H, et al SeqST-GAN: Seq2Seq generative adversarial nets for multi-step urban crowd flow prediction[J]. ACM Transactions on Spatial Algorithms and Systems, 2020, 6 (4): 1- 24
|
|
|
| [18] |
QIN Y, SONG D, CHEN H, et al. A dual-stage attention-base recurrent neural network for time series prediction [C]// Proceedings of 26th International Joint Conference on Artificial Intelligence. Melboume: [s. n. ], 2017: 399-404.
|
|
|
| [19] |
HOCHREITER S, SCHMIDHUBER J Long short term memory[J]. Neural Computation, 1997, 9 (8): 1735- 1780
doi: 10.1162/neco.1997.9.8.1735
|
|
|
| [20] |
CHEMALI E, KOLLMEYER P J, PREINDL M, et al Long short termmemory networks for accurate state-of-charge estimation of Li-ion batteries[J]. IEEE Transactions on Industrial Electronics, 2017, 65 (8): 6730- 6739
|
|
|
| [21] |
DONG G, WEI J, CHEN Z Kalman filter for onboard state of charge estimation and peak power capability analysis of lithium-ion batteries[J]. Journal of Power Sources, 2016, 328: 615- 626
doi: 10.1016/j.jpowsour.2016.08.065
|
|
|
| [22] |
胡杰, 高志文 基于数据驱动的电动汽车动力电池SOC预测[J]. 汽车工程, 2021, 43 (1): 1- 9
HU Jie, GAO Zhi-wen A data-driven SOC prediction scheme for traction battery in electric vehicles[J]. Automotive Engineering, 2021, 43 (1): 1- 9
|
|
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
| |
Shared |
|
|
|
|
| |
Discussed |
|
|
|
|
