Lithium Battery State-of-Charge Estimation Based on AdaBoost.Rt-RNN
Abstract
:1. Introduction
2. AdaBoost.Rt-RNN Algorithm Principle
2.1. Recurrent Neural Network
SOC Estimation of Lithium Battery Based on Recurrent Neural Network
- Input layer: open-circuit voltage at time t, high-frequency resistance (real part) , ohmic resistance , time constant , voltage change rate and the differential value of specific capacity as the input to the model, among them, the open-circuit voltage of the battery , the high-frequency resistance (real part) , ohmic resistance and time constant are tested through the HPPC working condition test, and ;
- Hidden layer: The hidden layer node at time t, where n represents the number of hidden layer nodes:
- Output layer: The SOC at time t is treated as the output of the model.
2.2. AdaBoost.Rt Ensemble Algorithm
2.2.1. Ensemble Learning
2.2.2. AdaBoost.Rt Ensemble Algorithm
- Given a training set of m samples, determine the number of iterations and the threshold . According to relevant research, when > 0.38, the prediction error begins to shake violently, and the performance of the strong learning machine is gradually unstable. Therefore, the selected threshold should not exceed 0.38;
- Initialize the sample weight , where is the number of training sets, is the current number of iterations; the error rate is ;
- Train a weak predictor and calculate the relative error of each sample ;
- Calculate the error rate ;
- Update sample weights ;Among them, is the normalization factor, which means the sum of the updated weights of all samples is 1.
- Determine whether it is established. If so, let , and continue to iterate; if , build a strong predictor and output.
2.3. AdaBoost.Rt-RNN Ensemble Algorithm
- Input the sample training set , where , , and represent the ith feature vector of the voltage, current, and capacity [16] in the lithium battery training samples, respectively; represents the number of training samples;
- Initialize the initial weight of the sample training set , train the weak predictor RNN1, calculate the relative error, update the weight of the training sample according to the error result, and iterate continuously until the weight of the K weak predictor RNNs is determined. The calculation method is expressed as Formula (9) below;
- In combination with the independent evaluation data set , K RNN models are assembled through AdaBoost.Rt. Finally, the AdaBoost.Rt-RNN model is constructed.
- Input the test data set ;
- Use the K RNN models formed by the AdaBoost.Rt-RNN model to predict the test data set, and obtain the prediction result ;
- Finally, the prediction result of the final lithium battery SOC is obtained.
3. Simulation Experiment and Result Analysis
3.1. Experimental Data
3.1.1. Experimental Data Selection
- Allow the lithium battery to stand for 60 min;
- Charging at 1 C constant current to 3.65 V, keep the voltage constant at 3.65 V, and stop charging when the current is less than 0.05 C;
- After standing for 60 min, use 1 C rate to discharge 10% SOC.
3.1.2. Data Normalization
3.2. Algorithm Evaluation Index
3.3. Simulation and Results Analysis
4. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
Abbreviations
SOC | State-of-charge |
RNN | Recurrent neural network |
BMS | Battery management system |
RBF | Radial-based function neural network |
BP | Back Propagation |
FCL | Fuzzy logic control |
GBDT | Gradient Boosting Decision Tree |
XGBoost | eXtreme Gradient Boosting |
OCV | Open-circuit voltage |
HPPC | Hybrid Pulse Power Characteristic |
MAE | Mean absolute error |
RMSE | Root mean square error |
References
- Yang, B.; Wang, Y.; Zhan, Y. Lithium Battery State-of-Charge Estimation Based on a Bayesian Optimization Bidirectional Long Short-Term Memory Neural Network. Energies 2022, 15, 4670. [Google Scholar] [CrossRef]
- Tan, Z.; Sun, R.; Yang, R. Overview of Battery Management System. J. Chongqing Univ. Technol. Nat. Sci. 2019, 33, 40–45. [Google Scholar]
- Zhang, D.; Couto, L.D.; Moura, S.J. Electrode-Level state estimation in lithium-ion batteries via kalman decomposition. J. IEEE Control Syst. Lett. 2021, 5, 1657–1662. [Google Scholar] [CrossRef]
- Ma, Y.; Chen, Q.; Qi, Z. Research on the soc definition and measurement method of batteries used in EVs. J. Tsinghua Univ. Sci. Tech. 2001, 11, 96–98. [Google Scholar]
- Fan, W.; Xu, G.; Yu, B.; Zhang, Z.; Lei, W.; Ren, M.; Dong, M. On-line Estimation Method for Internal Temperature of Lithium-ion Battery Based on Electrochemical Impedance Spectroscopy. J. Proc. CSEE 2021, 41, 3283–3293. [Google Scholar]
- Yang, N.; Zhang, X.; Shang, B.; Li, G. Unbalanced discharging and aging due to temperature differences among the cells in a lithium-ion battery pack with parallel combination. J. Power Sources 2016, 306, 733–741. [Google Scholar] [CrossRef]
- Tang, X.; Wang, Y.; Zou, C.; Yao, K.; Xia, Y.; Gao, F. A novel framework for Lithium-ion battery modeling considering uncertainties of temperature and aging. Energy Convers. Manag. 2018, 180, 162–170. [Google Scholar] [CrossRef]
- Xueyuan, W.; Xuezhe, W.; Jiangong, Z.; Haifeng, D.; Yuejiu, Z.; Xiaoming, X.; Qijun, C. A review of modeling, acquisition, and application of lithium-ion battery impedance for onboard battery management. J. eTransp. 2021, 7, 100093. [Google Scholar]
- Tang, X.; Gao, F.; Zou, C.; Yao, K.; Hu, W. Load-responsive model switching estimation for state of charge of lithium-ion batteries. Appl. Energy 2019, 238, 423–434. [Google Scholar] [CrossRef]
- Tang, H.; Zhang, Z.; Wu, B. Low Temperature Estimation of Battery SOC Based on BP Neural Network under HPPC Conditions. J. Comput. Syst. Appl. 2021, 30, 293–299. [Google Scholar]
- Lei, X.; Chen, Q.; Liu, K.; Ma, L. Radial-Based-Function Neural Network Based SOC Estimation. J. Trans. China Electrotech. Soc. 2008, 5, 81–87. [Google Scholar]
- Liu, X.; Dai, Y. Energy storage battery SOC estimate based on improved BP neural network. J. Phys. Conf. Ser. 2022, 2187, 012042. [Google Scholar] [CrossRef]
- Guan, K.; Wei, Z.; Yin, B. SOC Prediction Method of a New Lithium Battery Based on GA-BP Neural Network. In Lecture Notes in Electrical Engineering, Proceedings of the 4th International Conference on Computer Engineering and Networks, Shanghai, China, 19–20 July 2014; Springer International Publishing: Berlin/Heidelberg, Germany, 2015; pp. 141–153. [Google Scholar]
- Lei, X.; Chen, Q.; Liu, K.; Ma, L. Support Vector Machine Based SOC Estimation for Electric Vehicles. J. Proc. CSEE 2008, 18, 114–118. [Google Scholar]
- Liu, S.; Cui, Y.; Ma, Y.; Peng, L. Short-term Load Forecasting Based on GBDT Combinatorial Optimization. In Proceedings of the 2018 2nd IEEE Conference on Energy Internet and Energy System Integration (EI2), Beijing, China, 20–22 October 2018; pp. 1–5. [Google Scholar]
- Zhang, S.; Yang, L.; Zhao, X.; Qiang, J. A GA optimization for lithium-ion battery aqualization based on SOC estimation by NN and FCL. Int. J. Electr. Power Energy Syst. 2015, 73, 318–328. [Google Scholar] [CrossRef]
- Freund, Y.; Schapire, R.E. A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting. J. Comput. Syst. Sci. 1999, 55, 119–139. [Google Scholar] [CrossRef] [Green Version]
- Solomatine, D.P.; Shrestha, D.L. AdaBoost.RT: A boosting algorithm for regression problems. In Proceedings of the 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No. 04CH37541), Budapest, Hungary, 25–29 July 2004; pp. 1163–1168. [Google Scholar]
- Li, H. Areview of natural language processing based on RNN and Transformer models. J. Inf. Rec. Mater. 2021, 22, 7–10. [Google Scholar]
- Schuster, M.; Paliwal, K.K. Bidirectional recurrent neural networks. J. IEEE Trans. Signal Process. 1997, 45, 2673–2681. [Google Scholar] [CrossRef] [Green Version]
- Wei, Z.; Song, S.; Xia, H. State-of-charge Estimation Using Random Forest for Lithium Ion Battery. J. Guangxi Norm. Univ. Nat. Sci. Ed. 2018, 36, 27–33. [Google Scholar]
- Kurzweil, P.; Scheuerpflug, W.; Frenzel, B.; Schell, C.; Schottenbauer, J. Differential Capacity as a Tool for SOC and SOH Estimation of Lithium Ion Batteries Using Charge/Discharge Curves, Cyclic Voltammetry, Impedance Spectroscopy, and Heat Events: A Tutorial. Energies 2022, 15, 4520. [Google Scholar] [CrossRef]
- Shieh, Y.-T.; Wu, C.-C.; Liu, C.-Y.; Chieng, W.-H.; Su, Y.-S.; Jeng, S.-L.; Chang, E.-Y. Lithium Battery Model and Its Application to Parallel Charging. Energies 2022, 15, 4767. [Google Scholar] [CrossRef]
- El-Sehiemy, R.; Hamida, M.A.; Elattar, E.; Shaheen, A.; Ginidi, A. Nonlinear Dynamic Model for Parameter Estimation of Li-Ion Batteries Using Supply–Demand Algorithm. Energies 2022, 15, 4556. [Google Scholar] [CrossRef]
- Luo, Y.; Qian, Y.; Zeng, Z.; Zhang, Z. Simulation and Analysis of Operating Characteristics of Power Battery for Flying Car Utilization. J. eTransp. 2021, 8, 100111. [Google Scholar] [CrossRef]
- Yu, M.; Shi, W.; Jiang, L.; Wang, H. SOC estimation of power lithium battery based on GA-BP neural network. J. Appl. Electron. Tech. 2020, 46, 104–107112. [Google Scholar]
- Jiang, F.; Yang, J.; Cheng, Y.; Zhang, X.; Huang, Z. An Aging-Aware SOC Estimation Method for Lithium-Ion Batteries using XGBoost Algorithm. In Proceedings of the 2019 IEEE International Conference on Prognostics and Health Management (ICPHM), San Francisco, CA, USA, 17–20 June 2019. [Google Scholar]
- Li, J.; Liu, M. State-of-charge estimation of lithium-ion batteries using composite multi-dimensional features and a neural network. J. IET Power Electron. 2019, 12, 1470–1478. [Google Scholar] [CrossRef]
Parameter | Number |
---|---|
Nominal voltage | 3.7 V |
Discharge termination condition | Voltage < 2 V |
Charge termination condition | Current < 0.05 C |
Ambient temperature | 24 °C |
Quality | 1080 ± 10 g |
Model | MAE | RMSE |
---|---|---|
BP | 0.0426 | 0.0526 |
RNN | 0.0302 | 0.0378 |
XGBoost | 0.0235 | 0.0304 |
0.0366 | 0.0173 | |
UKF | 0.0439 | 0.0162 |
AdaBoost.Rt-RNN | 0.0158 | 0.0205 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Li, R.; Sun, H.; Wei, X.; Ta, W.; Wang, H. Lithium Battery State-of-Charge Estimation Based on AdaBoost.Rt-RNN. Energies 2022, 15, 6056. https://doi.org/10.3390/en15166056
Li R, Sun H, Wei X, Ta W, Wang H. Lithium Battery State-of-Charge Estimation Based on AdaBoost.Rt-RNN. Energies. 2022; 15(16):6056. https://doi.org/10.3390/en15166056
Chicago/Turabian StyleLi, Ran, Hui Sun, Xue Wei, Weiwen Ta, and Haiying Wang. 2022. "Lithium Battery State-of-Charge Estimation Based on AdaBoost.Rt-RNN" Energies 15, no. 16: 6056. https://doi.org/10.3390/en15166056
APA StyleLi, R., Sun, H., Wei, X., Ta, W., & Wang, H. (2022). Lithium Battery State-of-Charge Estimation Based on AdaBoost.Rt-RNN. Energies, 15(16), 6056. https://doi.org/10.3390/en15166056