基于ELM和MCSCKF的鋰離子電池SOC估計

王橋; 葉敏; 魏孟; 廉高棨; 武晨光

doi:10.13374/j.issn2095-9389.2022.05.10.003

基于ELM和MCSCKF的鋰離子電池SOC估計

doi: 10.13374/j.issn2095-9389.2022.05.10.003

王橋^1,,
葉敏^1, ,,
魏孟^{1, 2},
廉高棨¹,
武晨光¹

1.
長安大學公路養護裝備國家工程研究中心，西安 710064
2.
新加坡國立大學機械工程系，新加坡 117576

基金項目: 陜西省科技創新團隊（2020TD0012）；中央高校基本科研業務費專項資金-長安大學優秀博士學位論文培育資助項目（300102252710）；陜西省重點研發計劃資助項目（2023-GHYB-05）

詳細信息

通訊作者:
E-mail: mingye@chd.edu.cn

中圖分類號: TM911.3
計量
- 文章訪問數: 639
- HTML全文瀏覽量: 226
- PDF下載量: 77
- 被引次數: 0
出版歷程
- 收稿日期: 2022-05-10
- 網絡出版日期: 2022-07-05
- 刊出日期: 2023-05-31

ELM- and MCSCKF-based state of charge estimation for lithium-ion batteries

WANG Qiao^1
,,
YE Min^{1
, ,},
WEI Meng^{1, 2},
LIAN Gao-qi¹,
WU Chen-guang¹

1.
National Engineering Research Center for Highway Maintenance Equipment, Chang'an University, Xi'an 710064, China
2.
Department of Mechanical Engineering, National University of Singapore, Singapore 117576, Singapore

More Information

Corresponding author: E-mail: mingye@chd.edu.cn

摘要

摘要: 為了減少噪聲對鋰離子電池荷電狀態估計的影響，本文提出一種新穎的基于極限學習機和最大相關熵平方根容積卡爾曼濾波的SOC估計方法。首先，利用泛化性好、運行速度快的極限學習機作為卡爾曼濾波的測量方程；其次，基于灰狼優化算法，極限學習機的超參數被優化以提高電池荷電狀態的估計精度；最后，基于最大相關熵平方根容積卡爾曼濾波，極限學習機的測量噪聲被進一步減弱。所提方法可以簡化極限學習機繁瑣的調參過程，且為閉環的SOC估計方法。所提方法在多工況和寬溫度范圍內被測試以驗證其泛化性能。測試結果顯示，所提方法明顯地提高了鋰離子電池的荷電狀態估計精度。同時，對比其他算法，所提方法的平均運行時間僅僅為長短時序列和循環門控單元網絡的三分之一。當行駛工況復雜、溫度變化區間較大時，所提方法的均方根誤差小于1%，最大誤差小于3%。當存在初始誤差與環境噪聲時，所提方法顯示出了優越的魯棒性。
- 鋰離子電池 /
- 荷電狀態估計 /
- 極限學習機 /
- 灰狼優化 /
- 卡爾曼濾波 /
- 魯棒性估計
Abstract: Lithium-ion batteries are widely used in electric vehicles and energy storage systems. As a prerequisite for the safe and efficient application of lithium-ion batteries, battery management systems have received extensive attention worldwide. Among these prerequisites, the state of charge (SOC), as the basic parameter of battery management system online application, is crucial for the safe and efficient operation of battery management systems. However, measurement noise decreases the accuracy and robustness of the state of charge estimation. To reduce the impact of noise on the state of charge estimation of lithium-ion batteries, a novel SOC estimation method based on an extreme learning machine and a maximum correlation entropy square root volumetric Kalman filter is proposed in this paper. First, the extreme learning machine is used as the measurement equations of the Kalman filter because of its good generalization and fast running speed, and the voltage and current are selected as the model input; second, on the basis of the gray wolf optimization algorithm, the extreme learning machine hyperparameters are thoroughly optimized to improve the accuracy of the state of charge estimation for lithium-ion batteries; finally, on the basis of the framework of the maximum correlation entropy square root volumetric Kalman filter, a closed-loop estimation is realized to further reduce the state of charge estimation error caused by the measurement noise of voltage and current. The proposed method can simplify the time-consuming parameter adjustment of an extreme learning machine and show superior robustness under low-quality measurement. The proposed method is validated under multiple drive cycles and a wide temperature range to verify its generalization performance. The test results show that the proposed method substantially improves the accuracy of lithium-ion battery state of charge estimation. At the same time, the average running time of the proposed method is only one-third of that of long short memory neural networks and gate recurrent unit neural networks. Under complex driving conditions and a large temperature range, the root mean square error of the proposed method is less than 1%, and the maximum error is less than 3%. Furthermore, two case experiments are performed to evaluate the robustness of the proposed closed-loop estimation approach, and the results obtained when data have an initial state of charge error and measurement noise verify the superior robustness of the proposed approach compared with long short memory neural networks and gate recurrent unit neural networks.
- lithium-ion battery /
- state of charge /
- extreme learning machine /
- grey wolf optimizer /
- Kalman filter /
- closed-loop estimation

HTML全文

圖 1 極限學習機拓撲結構

Figure 1. Topology of an extreme learning machine

下載: 全尺寸圖片幻燈片

圖 2 基于MCSCKF的SOC閉環估計

Figure 2. Closed-loop SOC estimation realized by MCSCKF

下載: 全尺寸圖片幻燈片

圖 3 實驗測試平臺

Figure 3. Experimental test platform

下載: 全尺寸圖片幻燈片

圖 4 部分循環下被測樣本的電流電壓曲線. (a) DST工況；(b) US06工況；(c) FUDS工況；(d)隨機混合工況-1

Figure 4. Current and voltage curves of the tested samples under partial cycles: (a) DST cycle; (b) US06 cycle; (c) FUDS cycle; (d) Mix-1

下載: 全尺寸圖片幻燈片

圖 5 混合工況測試結果. (a)混合工況-1測試結果；(b)混合工況-2測試結果

Figure 5. SOC estimation results under a mixed drive cycle: (a) results under mix-1 cycle; (b) results under mix-2 cycle

下載: 全尺寸圖片幻燈片

圖 6 初始誤差校正測試結果. (a)混合工況-1測試結果；(b)混合工況-2測試結果

Figure 6. Initial SOC error correction test results: (a) results under mix-1 cycle; (b) results under mix-2 cycle

下載: 全尺寸圖片幻燈片

圖 7 隨機噪聲魯棒性測試結果. (a)混合工況-1測試結果；(b)混合工況-2測試結果

Figure 7. SOC estimation in the case of random noises: (a) results under mix-1 cycle; (b) results under mix-2 cycle

下載: 全尺寸圖片幻燈片

表 1 MCSCKF主要流程

Table 1. Main procedure of MCSCKF

Main procedure of MCSCKF
Step 1: Set the initial value and the corresponding square root covariance matrix and set t=1.
Step 2: Calculate the cubature points and propagation cubature points of the state equation.
Step 3: Calculate the prior estimate and square the root covariance.
Step 4: Evaluate the cubature points and the propagation cubature points of the measurement equation.
Step 5: Define the square root covariance matrix of the updated measurement value with MCC, calculate the average value of the prior 　　　measurement, and get the updated square root covariance matrix.
Step 6: Compare the matrix transformation of the measured value with the threshold: If it is greater than the threshold, return to step 2; 　　　otherwise, go to the next step.
Step 7: Calculate the posterior state estimate and the corresponding square root covariance matrix, t=t+1 and return to step 2.

下載: 導出CSV

表 2 被測電池詳細參數

Table 2. Detailed parameters of tested battery

Test sample	Parameters	Value
	Normal capacity/( mA·h)	2200
	Normal voltage/V	3.7
	Weight/g	43.8
	Internal impedance/ mΩ	≤50

下載: 導出CSV

表 3 混合工況測試結果

Table 3. SOC estimation results under a mixed drive cycle

Methods	Drive cycle	Max error/%	RMSE/%	Average running time/s
LSTM	Mix-1	80.442	1.86	3201.34
LSTM	Mix-2	59.279	1.45	3371.71
GRU	Mix-1	24.204	1.08	2993.41
GRU	Mix-2	24.355	1.11	3227.93
Proposed method	Mix-1	2.299	0.92	970.54
Proposed method	Mix-2	3.054	0.61	1027.67

下載: 導出CSV

www.77susu.com

參考文獻(26)

[1]	Gao H N, Xu J, Pu X H, et al. An online control method for energy recovery of suspension vibration of new energy vehicles. J Xi’an Jiaotong Univ, 2020, 54(4): 19 doi: 10.7652/xjtuxb202004003 郜浩楠, 徐俊, 蒲曉暉, 等. 面向新能源汽車的懸架振動能量回收在線控制方法. 西安交通大學學報, 2020, 54(4):19 doi: 10.7652/xjtuxb202004003
[2]	Shrivastava P, Soon T K, Idris M Y I B, et al. Overview of model-based online state-of-charge estimation using Kalman filter family for lithium-ion batteries. Renew Sustain Energy Rev, 2019, 113: 109233 doi: 10.1016/j.rser.2019.06.040
[3]	Wang Y J, Tian J Q, Sun Z D, et al. A comprehensive review of battery modeling and state estimation approaches for advanced battery management systems. Renew Sustain Energy Rev, 2020, 131: 110015 doi: 10.1016/j.rser.2020.110015
[4]	Xiong R, Cao J Y, Yu Q Q, et al. Critical review on the battery state of charge estimation methods for electric vehicles. IEEE Access, 2017, 6: 1832
[5]	Zhang S Z, Guo X, Dou X X, et al. A data-driven coulomb counting method for state of charge calibration and estimation of lithium-ion battery. Sustain Energy Technol Assess, 2020, 40: 100752
[6]	Xiong R, Yu Q Q, Wang L Y, et al. A novel method to obtain the open circuit voltage for the state of charge of lithium ion batteries in electric vehicles by using H infinity filter. Appl Energy, 2017, 207: 346 doi: 10.1016/j.apenergy.2017.05.136
[7]	Wang X L, Jin H Q, Liu X Y. Online estimation of the state of charge of a lithium-ion battery based on the fusion model. Chin J Eng, 2020, 42(9): 1200 王曉蘭, 靳皓晴, 劉祥遠. 基于融合模型的鋰離子電池荷電狀態在線估計. 工程科學學報, 2020, 42(9):1200
[8]	Zhang Q, Cui N X, Li Y, et al. Fractional calculus based modeling of open circuit voltage of lithium-ion batteries for electric vehicles. J Energy Storage, 2020, 27: 100945 doi: 10.1016/j.est.2019.100945
[9]	Feng F, Teng S L, Liu K L, et al. Co-estimation of lithium-ion battery state of charge and state of temperature based on a hybrid electrochemical-thermal-neural-network model. J Power Sources, 2020, 455: 227935 doi: 10.1016/j.jpowsour.2020.227935
[10]	Zhu R, Duan B, Zhang J M, et al. Co-estimation of model parameters and state-of-charge for lithium-ion batteries with recursive restricted total least squares and unscented Kalman filter. Appl Energy, 2020, 277: 115494 doi: 10.1016/j.apenergy.2020.115494
[11]	Liu X, Qu H, Zhao J H, et al. Maximum correntropy square-root cubature Kalman filter with application to SINS/GPS integrated systems. ISA Trans, 2018, 80: 195 doi: 10.1016/j.isatra.2018.05.001
[12]	Almeida G C S, Souza A C Z, Ribeiro P F. A neural network application for a lithium-ion battery pack state-of-charge estimator with enhanced accuracy. Proceedings, 2020, 58(1): 33
[13]	Wang J, Yang Y Q, Wang T, et al. Big data service architecture: a survey. J Internet Technol, 2020, 21(2): 393
[14]	Reddy G T, Reddy M P K, Lakshmanna K, et al. Analysis of dimensionality reduction techniques on big data. IEEE Access, 2020, 8: 54776 doi: 10.1109/ACCESS.2020.2980942
[15]	Gozde O S, Milutin P, Zafer S, et al. Battery state-of-charge estimation based on regular/recurrent Gaussian process regression. IEEE Trans. Ind. Electron., 2018, 65(5): 4311 doi: 10.1109/TIE.2017.2764869
[16]	Li X Y, Yuan C G, Li X H, et al. State of health estimation for Li-Ion battery using incremental capacity analysis and Gaussian process regression. Energy, 2020, 190: 116467 doi: 10.1016/j.energy.2019.116467
[17]	Ren X Q, Liu S L, Yu X D, et al. A method for state-of-charge estimation of lithium-ion batteries based on PSO-LSTM. Energy, 2021, 234: 121236 doi: 10.1016/j.energy.2021.121236
[18]	Jiao M, Wang D Q, Qiu J L. A GRU-RNN based momentum optimized algorithm for SOC estimation. J Power Sources, 2020, 459: 228051 doi: 10.1016/j.jpowsour.2020.228051
[19]	Huang G B, Zhu Q Y, Siew C K. Extreme learning machine: Theory and applications. Neurocomputing, 2006, 70(1-3): 489 doi: 10.1016/j.neucom.2005.12.126
[20]	Jiao M, Wang D Q, Yang Y, et al. More intelligent and robust estimation of battery state-of-charge with an improved regularized extreme learning machine. Eng Appl Artif Intell, 2021, 104: 104407 doi: 10.1016/j.engappai.2021.104407
[21]	Hossain Lipu M S, Hannan M A, Hussain A, et al. Extreme learning machine model for state-of-charge estimation of lithium-ion battery using gravitational search algorithm. IEEE Trans Ind Appl, 2019, 55(4): 4225 doi: 10.1109/TIA.2019.2902532
[22]	Mirjalili S, Mirjalili S M, Lewis A. Grey wolf optimizer. Adv Eng Softw, 2014, 69: 46 doi: 10.1016/j.advengsoft.2013.12.007
[23]	Deng Z W, Hu X S, Lin X K, et al. Data-driven state of charge estimation for lithium-ion battery packs based on Gaussian process regression. Energy, 2020, 205: 118000 doi: 10.1016/j.energy.2020.118000
[24]	Wei Z B, Hu J, Li Y, et al. Hierarchical soft measurement of load current and state of charge for future smart lithium-ion batteries. Appl Energy, 2022, 307: 118246 doi: 10.1016/j.apenergy.2021.118246
[25]	Wei Z B, Zhao D F, He H W, et al. A noise-tolerant model parameterization method for lithium-ion battery management system. Appl Energy, 2020, 268: 114932 doi: 10.1016/j.apenergy.2020.114932
[26]	Wei Z B, Dong G Z, Zhang X N, et al. Noise-immune model identification and state-of-charge estimation for lithium-ion battery using bilinear parameterization. IEEE Trans Ind Electron, 2021, 68(1): 312 doi: 10.1109/TIE.2019.2962429