SAM-GRU-based fusion neural network for SOC estimation in lithium-ion batteries under a wide range of operating conditions

doi:10.19799/j.cnki.2095-4239.2023.0292

Abstract

Abstract:

Accurate estimation of the state of charge (SOC) of lithium-ion batteries under a wide range of operating conditions is crucial for ensuring the operational safety and reliability of electric vehicles; therefore, estimating SOC is one of the most important tasks of battery management systems. In this study, a fusion neural network model combining Self-Attention Mechanism (SAM) and Gated Recurrent Unit (GRU) is proposed to capture the long-term nonlinear mapping relationship between the measurable parameters (voltage and current) and SOC of lithium-ion batteries. This SAM-GRU neural network model makes full use of the short-time processing capability of GRU and the long-time sequence feature-extraction capability of SAM. Additionally, this model simplifies the effective characterization of the long-sequence-related features of SOC. Based on the results of the Beijing Bus Dynamic Stress Test, the proposed SAM-GRU neural network model yields more accurate SOC estimates than the traditional GRU neural network under different discharge rates, environmental temperatures, and combinations of both. Specifically, the improvements in accuracy are no less than 26%, 25%, and 11%, respectively.

Key words: lithium-ion battery, state of charge, self-attentive mechanism, gated recurrent unit neural network

CLC Number:

TM 912

Hongsheng GUAN, Cheng QIAN, Binghui XU, Bo SUN, Yi REN. SAM-GRU-based fusion neural network for SOC estimation in lithium-ion batteries under a wide range of operating conditions[J]. Energy Storage Science and Technology, 2023, 12(7): 2229-2237.

Figures/Tables 14

Fig. 1

Fig. 2

Fig. 3

Table 1

Fig. 4

Table 2

Fig. 5

Fig. 6

Table 3

Fig. 7

Table 4

Fig. 8

Table 5

Fig. 9

References 18

1	YANG C F, WANG X Y, FANG Q H, et al. An online SOC and capacity estimation method for aged lithium-ion battery pack considering cell inconsistency[J]. Journal of Energy Storage, 2020, 29: doi: 10.1016/j.est.2020.101250.
2	HOSSAIN LIPU M S, HANNAN M A, KARIM T F, et al. Intelligent algorithms and control strategies for battery management system in electric vehicles: Progress, challenges and future outlook[J]. Journal of Cleaner Production, 2021, 292: doi: 10.1016/j.jclepro. 2021.126044.
3	NI Z C, YANG Y. A combined data-model method for state-of-charge estimation of lithium-ion batteries[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 1-11.
4	WU L F, ZHANG Y. Attention-based encoder-decoder networks for state of charge estimation of lithium-ion battery[J]. Energy, 2023, 268: doi: 10.1016/j.energy.2023.126665.
5	BIAN C, YANG S K, LIU J, et al. Robust state-of-charge estimation of Li-ion batteries based on multichannel convolutional and bidirectional recurrent neural networks[J]. Applied Soft Computing, 2022, 116:doi: 10.1016/j.asoc.2021.108401.
6	GUO S S, MA L. A comparative study of different deep learning algorithms for lithium-ion batteries on state-of-charge estimation[J]. Energy, 2023, 263: doi: 10.1016/j.energy.2022.125872.
7	ZHANG G Y, XIA B Z, WANG J M. Intelligent state of charge estimation of lithium-ion batteries based on L-M optimized back-propagation neural network[J]. Journal of Energy Storage, 2021, 44: doi: 10.1016/j.est.2021.103442.
8	HOW D N T, HANNAN M A, HOSSAIN LIPU M S, et al. State-of-charge estimation of Li-ion battery in electric vehicles: A deep neural network approach[C]//2019 IEEE Industry Applications Society Annual Meeting. September 29—October 3, 2019, Baltimore, MD, USA. IEEE, 2019: 1-8.
9	HANNAN M A, HOW D N T, HOSSAIN LIPU M S, et al. SOC estimation of Li-ion batteries with learning rate-optimized deep fully convolutional network[J]. IEEE Transactions on Power Electronics, 2021, 36(7): 7349-7353.
10	LIU Y F, LI J Q, ZHANG G, et al. State of charge estimation of lithium-ion batteries based on temporal convolutional network and transfer learning[J]. IEEE Access, 2021, 9: 34177-34187.
11	CHEN J X, ZHANG Y, WU J, et al. SOC estimation for lithium-ion battery using the LSTM-RNN with extended input and constrained output[J]. Energy, 2023, 262: doi: 10.1016/j.energy.2022.125375.
12	TIAN J P, XIONG R, SHEN W X, et al. State-of-charge estimation of LiFePO₄ batteries in electric vehicles: A deep-learning enabled approach[J]. Applied Energy, 2021, 291:doi: 10.1016/j.apenergy. 2021.116812.
13	YANG F F, LI W H, LI C, et al. State-of-charge estimation of lithium-ion batteries based on gated recurrent neural network[J]. Energy, 2019, 175: 66-75.
14	HANNAN M A, HOW D N T, BIN MANSOR M, et al. State-of-charge estimation of Li-ion battery using gated recurrent unit with one-cycle learning rate policy[J]. IEEE Transactions on Industry Applications, 2021, 57(3): 2964-2971.
15	朱元富, 贺文武, 李建兴, 等. 基于Bi-LSTM/Bi-GRU循环神经网络的锂电池SOC估计[J]. 储能科学与技术, 2021, 10(3): 1163-1176.
	ZHU Y F, HE W W, LI J X, et al. SOC estimation for Li-ion batteries based on Bi-LSTM and Bi-GRU[J]. Energy Storage Science and Technology, 2021, 10(3): 1163-1176.
16	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]. La Jolla: Neural Information Processing Systems (nips), 2017: doi: 10.48550/arXiv.1706.03762.
17	HU C S, CHENG F J, MA L A, et al. State of charge estimation for lithium-ion batteries based on TCN-LSTM neural networks[J]. Journal of the Electrochemical Society, 2022, 169(3): 030544.
18	CHEN Z, ZHAO H Q, SHU X, et al. Synthetic state of charge estimation for lithium-ion batteries based on long short-term memory network modeling and adaptive H-Infinity filter[J]. Energy, 2021, 228: doi: 10.1016/j.energy.2021.120630.

网络层	特征数/隐藏节点数
输入	2
FC0	64
SAM层	64
GRU层	64
FC1	8
FC2(输出)	1

组号	电池编号	温度/℃	放电倍率
G1	B211、B212、B213	25	1 C
G2	B221、B222、B223	25	2 C
G3	B421、B422、B423	45	2 C

网络模型	1 C(B213)			2 C(B223)
网络模型	RMSE	MAE	MAX	RMSE	MAE	MAX
GRU	0.762	0.700	3.936	0.852	0.723	4.566
SAM-GRU	0.557	0.493	3.271	0.486	0.391	3.164

网络模型	25 ℃(B223)			45 ℃(B423)
网络模型	RMSE	MAE	MAX	RMSE	MAE	MAX
GRU	0.774	0.652	5.072	0.540	0.422	1.907
SAM-GRU	0.511	0.385	3.015	0.407	0.321	1.748

网络模型	1 C, 25 ℃(B213)			2 C, 25 ℃(B223)			2 C, 45 ℃(B423)
网络模型	RMSE	MAE	MAX	RMSE	MAE	MAX	RMSE	MAE	MAX
GRU	0.699	0.621	4.129	0.698	0.600	3.094	0.451	0.356	1.983
SAM-GRU	0.619	0.549	3.109	0.421	0.355	2.601	0.423	0.342	1.762