相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Engineering, Electrical & Electronic
Liyuan Shen et al.
Summary: SOC estimation is crucial for the safety of batteries in battery management systems. However, existing data-driven methods for Li-ion battery SOC estimation require labeled data that share the same distribution as the testing data, which is not realistic in the real world. To solve this problem, a temperature adaptive transfer network (TATN) is proposed, which can adaptively mitigate distribution discrepancy. The TATN achieves state-of-the-art performance in practical applications, reducing errors significantly compared to other methods.
IEEE TRANSACTIONS ON POWER ELECTRONICS
(2023)
Article
Energy & Fuels
Isaiah Oyewole et al.
Summary: In this paper, a Controllable Deep Transfer Learning (CDTL) network is proposed for short and long-term state-of-charge (SOC) estimation. The CDTL utilizes shared knowledge between target cells and historical source cells using controllable Multiple Domain Adaptation (MDA). Experimental results demonstrate that the CDTL outperforms existing benchmarks in deep learning and transfer learning.
Article
Automation & Control Systems
Farshid Naseri et al.
Summary: This article introduces an efficient modeling approach based on the Wiener structure to enhance the capacity of classical equivalent circuit models in capturing the nonlinearities of Li-ion batteries. The proposed method combines a linear ECM with a static output nonlinearity block to achieve superior nonlinear mapping property while maintaining real-time efficiency. The observability of the battery model is analytically proven, and an efficient parameter estimator is introduced for real-time estimation of the Wiener model's parameters.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2022)
Article
Engineering, Electrical & Electronic
Zhekai Du et al.
Summary: The article proposes a unified deep learning method for the estimation of RUL and SoC in rechargeable lithium-ion batteries. The method leverages LSTM recurrent neural networks to achieve accurate capacity estimation under complex operating conditions. Experimental results show that the method can increase estimation accuracy for LIBs by capturing long-term dependencies in battery degradation data.
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION
(2022)
Article
Computer Science, Artificial Intelligence
Jingjing Li et al.
Summary: This paper addresses the problem of accelerating machine learning in situations where there is a lack of training data and limited computing power. By proposing the Faster Domain Adaptation (FDA) protocol and two paradigms, the method achieves comparable or even better accuracy while using fewer computing resources.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Jingjing Li et al.
Summary: This paper discusses the generalization problem of conventional machine learning algorithms when dealing with data from different distributions and proposes an unsupervised domain adaptation method. The authors consider the scenario where either the source domain data or the target domain data is unknown, and investigate how to handle the divergence-agnostic adaptive learning problem from the perspective of adversarial attacks. Experimental results show that the proposed method outperforms previous ones.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Engineering, Electrical & Electronic
Amin Bavand et al.
Summary: Estimating the state of health (SOH) and state of charge (SOC) of lithium-ion batteries is crucial for increasing battery lifetime and performance. A fast online SOH estimation method is introduced in this article, which uses partial charge/discharge and updates the OCV reference curve based on temperature readings for more accurate results. Experimental results validate the proposed method with low reported errors for both SOH and SOC.
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION
(2022)
Article
Computer Science, Artificial Intelligence
Elias Galiounas et al.
Summary: The potential of using acoustic signatures for State-of-Charge (SoC) estimation is demonstrated through artificial neural network regression models. A more streamlined approach of processing the entire acoustic waveform is used instead of manual waveform peak selection. Simple metrics of statistical significance are used to identify the most informative waveform features for SoC inference. It is shown that early and late interfacial reflections can highly correlate with SoC, challenging the common peak selection methods. Transformation into the frequency domain helps reduce dimensionality of the problem and improve estimation accuracy. Furthermore, it is shown that acoustic signatures alone can be used to produce highly accurate SoC estimates without voltage information.
Article
Automation & Control Systems
Yan Qin et al.
Summary: The proposed new method reduces prediction errors at fixed temperatures and improves prediction accuracies at new temperatures by utilizing temporal dynamics of measurements and transferring consistent estimation ability among different temperatures.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Computer Science, Artificial Intelligence
Jingjing Li et al.
Summary: This paper proposes a new domain adaptation method named ATM, which aims to reduce domain divergence by minimizing inter-domain divergence and maximizing intra-class density, combining adversarial training and metric learning. Experimental results show that the proposed ATM achieves state-of-the-art performance on multiple benchmark datasets.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2021)
Article
Energy & Fuels
Ji'ang Zhang et al.
Summary: This paper investigates the use of electrochemical impedance spectroscopy theory to guide and improve the equivalent circuit model, proposing a high accuracy and adaptability SOC online estimation method.
Review
Computer Science, Artificial Intelligence
Wouter M. Kouw et al.
Summary: This review categorizes approaches in domain adaptation into sample-based, feature-based, and inference-based methods, highlighting the importance of conditions for formulating bounds on cross-domain generalization error.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2021)
Article
Thermodynamics
Guodong Fan et al.
Summary: This paper presents a robust estimator based on a reduced-order electrochemical model for accurate state-of-charge estimation of lithium ion batteries at different ambient temperatures. A global sensitivity analysis investigates the impact of temperatures on model performance, and a robust adaptive extended Kalman filter is designed to actively compensate potential model uncertainties. Results show the proposed algorithm's robustness against initial estimation errors, driving cycle inputs, and ambient temperatures, with extensive comparative analysis with a conventional extended Kalman filter.
Article
Thermodynamics
Iman Babaeiyazdi et al.
Summary: This paper utilizes electrochemical impedance spectroscopy (EIS) and machine learning models to accurately predict the state of charge (SOC) of li-ion batteries, demonstrating the effectiveness and accuracy of the method in feature selection and model training.
Article
Engineering, Multidisciplinary
Mahammad A. Hannan et al.
Summary: This study compares commonly used deep learning models and proposes a two-hidden layer stacked gated recurrent unit model trained with a one-cycle policy learning rate scheduler, achieving a minimum RMSE of 0.52% on the train dataset and 0.65% on the test dataset while maintaining a relatively low computation cost.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2021)
Article
Energy & Fuels
Zhong Ren et al.
Summary: This paper investigates the effects of different current rates on OCV tests and their impact on SoC estimation algorithms. The results show that the OCV-SoC curve obtained from the 0.5C incremental OCV test performs best at 25 degrees C and 45 degrees C, while the 0.2C incremental OCV test performs best at 0 degrees C. Additionally, a 3-hour relaxation time is optimal at higher temperatures, while a 4-hour relaxation time is recommended at 0 degrees C.
INTERNATIONAL JOURNAL OF ENERGY RESEARCH
(2021)
Article
Computer Science, Artificial Intelligence
J. Ashok Kumar et al.
Summary: This paper proposes a method using bidirectional LSTM and GRU networks in combination with stratified sampling technique to handle imbalanced aspect categories. By utilizing corpus-specific word embeddings and pre-trained word embeddings, the performance is improved.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Energy & Fuels
Kiarash Movassagh et al.
Summary: This paper investigates the error in state-of-charge estimation during Coulomb counting, caused by various sources such as current measurement, current integration approximation, battery capacity uncertainty, and timing oscillator error/drift. The study shows that the errors can be either time-cumulative or state-of-charge-proportional, highlighting the importance of reducing these mistakes through simulation analysis.
Article
Engineering, Electrical & Electronic
M. A. Hannan et al.
Summary: In this study, a deep fully convolutional network model was proposed to accurately estimate the state-of-charge (SOC) of lithium-ion batteries, achieving impressive performance under different temperature conditions.
IEEE TRANSACTIONS ON POWER ELECTRONICS
(2021)
Article
Engineering, Electrical & Electronic
Chong Bian et al.
Summary: The research introduces a deep transfer neural network (DTNN) with multiscale distribution adaptation (MDA) for cross-domain SOC estimation, which learns dynamic features from battery measurements in different domains and minimizes distribution discrepancy through constraint terms. DTNN shows improved performance under specific battery conditions with limited training data, compared to existing transfer learning methods, enhancing generalizability and robustness.
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION
(2021)
Article
Engineering, Electrical & Electronic
Xing Shu et al.
Summary: The accurate estimation of SOC of lithium-ion battery packs is achieved by constructing a model based on LSTM RNN and improving SRCKF algorithm, along with rolling learning method and multi-innovation technique. The combination of SOC estimation from maximum and minimum cells with smoothing method also contributes to the precise SOC estimation.
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION
(2021)
Article
Engineering, Electrical & Electronic
Xiaolei Bian et al.
Summary: The article proposes a novel method for online SOC estimation using particle swarm optimization (PSO) algorithm to optimize model parameters for high accuracy.
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION
(2021)
Article
Energy & Fuels
Xinyou Lin et al.
Summary: This study proposed a SOC estimation method based on an accurate equivalent circuit model, and demonstrated the accuracy advantage of the n-RC model and its increasing accuracy with the increase of n through experiments.
JOURNAL OF ENERGY STORAGE
(2021)
Proceedings Paper
Computer Science, Theory & Methods
Himadri Sekhar Bhattacharyya et al.
Summary: This paper proposed a Neural Network approach known as Convolutional Neural Network (CNN) for State of Charge (SOC) estimation in Batter Management System (BMS), using the New York City Cycle (NYCC) driving cycle dataset for evaluating the model's performance. By validating the model against a different number of hyperparameters, the proposed CNN showed competitive estimation performance compared to other model-based methods for SOC estimation.
2021 5TH INTERNATIONAL CONFERENCE ON ELECTRICAL, ELECTRONICS, COMMUNICATION, COMPUTER TECHNOLOGIES AND OPTIMIZATION TECHNIQUES (ICEECCOT)
(2021)
Article
Computer Science, Artificial Intelligence
Jie Li et al.
Summary: This study focuses on optimizing the training dataset by using simple measurable data sets to improve prediction accuracy, reduce training time, and apply to online estimation. It was found that randomly generated data sets can be successfully applied to various dynamic battery operating conditions. The XGBoost method shows excellent performance for SOC prediction, with fast learning time, short running time, and accurate results.
Article
Computer Science, Information Systems
Yuefeng Liu et al.
Summary: The article introduces a method using Temporal Convolutional Network (TCN) to estimate the SOC of lithium-ion batteries, which can directly map measured values such as voltage, current, and temperature to accurate SOC without using a battery model. The proposed method can accurately estimate SOC under different conditions and achieve an average MAE of 0.67% under various ambient temperature conditions.
Article
Energy & Fuels
Sheng Shen et al.
Article
Energy & Fuels
Yong Tian et al.
Article
Energy & Fuels
Alejandro Gismero et al.
Proceedings Paper
Automation & Control Systems
Gelareh Javid et al.
IECON 2020: THE 46TH ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY
(2020)
Article
Automation & Control Systems
Jingjing Li et al.
IEEE TRANSACTIONS ON CYBERNETICS
(2019)
Review
Energy & Fuels
Muhammad Umair Ali et al.
Article
Computer Science, Artificial Intelligence
Jingjing Li et al.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2019)
Review
Green & Sustainable Science & Technology
Prashant Shrivastava et al.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2019)
Proceedings Paper
Energy & Fuels
Woo-Yong Kim et al.
2019 IEEE ENERGY CONVERSION CONGRESS AND EXPOSITION (ECCE)
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Chaoran Li et al.
PROCEEDINGS OF 2019 IEEE 8TH JOINT INTERNATIONAL INFORMATION TECHNOLOGY AND ARTIFICIAL INTELLIGENCE CONFERENCE (ITAIC 2019)
(2019)
Proceedings Paper
Engineering, Electrical & Electronic
Carlos Vidal et al.
2019 IEEE TRANSPORTATION ELECTRIFICATION CONFERENCE AND EXPO (ITEC)
(2019)
Review
Computer Science, Information Systems
Rui Xiong et al.
Article
Thermodynamics
Fengchun Sun et al.
Article
Energy & Fuels
Kong Soon Ng et al.
Article
Biochemical Research Methods
Karsten M. Borgwardt et al.
