Search

The increasing demand for reliable and safe Lithium-ion (Li-ion) batteries requires more accurate estimation of state of health (SOH) and remaining useful life (RUL) prediction. However, the inherent complexity and non-linear dynamics of Li-ion batteries present specific challenges to traditional methods of SOH modeling. Although particle filter (PF) techniques can handle nonlinear dynamics, they still face challenges, including particle degeneracy and loss of diversity, that reduce their ability to effectively model the nonlinear degradation mechanisms of batteries. To tackle these limitations, this paper presents a novel artificial intelligence-driven PF (AI-PF) technology for battery health modeling and prognosis. The main contributions of the AI-PF technique are as follows: (1) A novel dynamic sample degeneracy detection method is proposed to provide real-time assessment of particle weights so as to promptly identify degeneracy and improve computational efficiency. (2) An adaptive crossover and mutation strategy is proposed to reallocate low-weight particles and maintain particle diversity to improve modeling and RUL forecasting accuracy. The effectiveness of the AI-PF framework is validated through systematic evaluations carried out using benchmark models and well-recognized battery datasets. [ABSTRACT FROM AUTHOR]

This paper presents the development of an advanced battery management system (BMS) for electric vehicles (EVs), designed to enhance battery performance, safety, and longevity. Central to the BMS is its precise monitoring of critical parameters, including voltage, current, and temperature, enabled by dedicated sensors. These sensors facilitate accurate calculations of the state of charge (SOC) and state of health (SOH), with real-time data displayed through an IoT cloud interface. The proposed BMS employs data-driven approaches, like advanced Kalman filters (KF), for battery state estimation, allowing continuous updates to the battery state with improved accuracy and adaptability during each charging cycle. Simulation tests conducted in MATLAB's Simulink across multiple charging and discharging cycles demonstrate the superior accuracy of the advanced Kalman filter (KF), in handling non-linear battery behaviours. Results indicate that the proposed BMS achieves a significantly lower error margin in SOC tracking, ranging from 0.32% to 1%, compared to traditional methods with error margins up to 5%. These findings underscore the importance of integrating robust sensor systems in BMSs to optimise EV battery management, reduce maintenance costs, and improve battery sustainability. [ABSTRACT FROM AUTHOR]

Effective health management and accurate state of charge (SOC) estimation are crucial for the safety and longevity of lithium-ion batteries (LIBs), particularly in electric vehicles. This paper presents a health management system (HMS) that continuously monitors a 4s2p LIB pack's parameters—current, voltage, and temperature—to mitigate risks such as overcurrent and thermal runaway while ensuring balanced charge distribution between cells. An improved online battery model (IOBM) is developed to enhance SOC estimation accuracy. The system utilises forgetting factor recursive least squares (FFRLS) for real-time parameter updates, an adaptive nonlinear sliding mode observer (ANSMO) for SOC estimation, and a long short-term memory (LSTM) network to dynamically adjust capacity based on operating conditions. Validation using the urban dynamometer driving schedule (UDDS) test demonstrated high accuracy, with the proposed battery model achieving a root mean square error (RMSE) of 12.13 mV and the LSTM achieving an RMSE of 0.0118 Ah. Regular updates to the battery's current capacity, along with the proposed IOBM, significantly improved SOC estimation performance, maintaining estimation errors within 1.08%. [ABSTRACT FROM AUTHOR]

Na-ion batteries are growing interest due to their sustainability and low cost. A wide implementation in stationary applications, but also for short range transportation, is envisaged. This is further supported by the recent progress on Na-ion cells with increased energy density. To this regards, the development of procedures for real-time assessment of batteries state of health is of crucial relevance. The present paper provides an innovative procedure to assess sodium-ion battery capacity fading based on the application of discrete wavelet transform to voltage signals, acquired once a certain load pattern is applied at the battery terminals. The procedure development is provided through Na-ion cell aging test. During all the test battery capacity measurements are carried out. Root mean square error (RMSE) between assessed and measured values equals 1.18%. Moreover, during the aging test significant differences between performance evolution of Na-ion and NCR Li-ion cells are highlighted and discussed.

This paper presents the development of an advanced battery management system (BMS) for electric vehicles (EVs), designed to enhance battery performance, safety, and longevity. Central to the BMS is its precise monitoring of critical parameters, including voltage, current, and temperature, enabled by dedicated sensors. These sensors facilitate accurate calculations of the state of charge (SOC) and state of health (SOH), with real-time data displayed through an IoT cloud interface. The proposed BMS employs data-driven approaches, like advanced Kalman filters (KF), for battery state estimation, allowing continuous updates to the battery state with improved accuracy and adaptability during each charging cycle. Simulation tests conducted in MATLAB’s Simulink across multiple charging and discharging cycles demonstrate the superior accuracy of the advanced Kalman filter (KF), in handling non-linear battery behaviours. Results indicate that the proposed BMS achieves a significantly lower error margin in SOC tracking, ranging from 0.32% to 1%, compared to traditional methods with error margins up to 5%. These findings underscore the importance of integrating robust sensor systems in BMSs to optimise EV battery management, reduce maintenance costs, and improve battery sustainability.

As one of the key parameters to characterize the life of lithium‐ion batteries, the state of health (SOH) is of great importance in ensuring the reliability and safety of the battery system. Considering the complexity of practical application scenarios, a novel method based on mechanical‐electrical feature extraction and machine learning is proposed to accurately estimate the SOH. A series of degradation experiments are designed to generate battery aging datasets, including the stress and voltage changes. Health features are directly extracted from the stress‐voltage profile and the mechanical‐electrical health feature factors are obtained through correlation analysis. The long short‐term memory (LSTM) network is introduced to map the relationship between mechanical‐electrical responses and the SOH, where the health feature factors are selected as input vectors. The effectiveness of the proposed method is demonstrated by battery datasets under different conditions, from which the estimated errors are less than 1.5 %. This work demonstrates that the analysis and utilization of mechanical‐electrical parameters can not only realize accurate SOH estimation, but also provide a broader field for battery energy management. [ABSTRACT FROM AUTHOR]

As the low-carbon economy continues to advance, New Energy Vehicles (NEVs) have risen to prominence in the automotive industry. The design and utilization of lithium-ion batteries (LIBs), which are core component of NEVs, are directly related to the safety and range performance of electric vehicles. The requirements for a refined design of lithium-ion battery electrode structures and the intelligent adjustment of charging modes have attracted extensive research from both academia and industry. LIB models can be divided into mechanism-based models and data-driven models; however, the distinctions and connections between these two kinds of models have not been systematically reviewed as yet. Therefore, this work provides an overview and perspectives on LIB modeling from both mechanism-based and data-driven perspectives. Meanwhile, the potential fusion modeling frameworks including mechanism information and a data-driven method are also summarized. An introduction to LIB modeling technologies is presented, along with the current challenges and opportunities. From the mechanism-based perspective of LIB structure design, we further explore how electrode morphology and aging-related side reactions impact battery performance. Furthermore, within the realm of battery operation, the utilization of data-driven models that leverage machine learning techniques to estimate battery health status is investigated. The bottlenecks for the design, state estimation, and operational optimization of LIBs and potential prospects for mechanism-data hybrid modeling are highlighted at the end. This work is expected to assist researchers and engineers in uncovering the potential value of mechanism information and operation data, thereby facilitating the intelligent transformation of the lithium-ion battery industry towards energy conservation and efficiency enhancement. [ABSTRACT FROM AUTHOR]

The accurate prediction of a lithium-ion battery's State of Health is of critical importance for efficient battery health management. Existing data-driven estimation methodologies grapple with issues such as high model complexity and a dearth of guidance from prior knowledge, which impose constraints on their efficacy. This work introduces a novel cross-modal distillation network for battery State of Health estimation, structured around a TransformerEncoder as the teacher network and a Convolutional Neural Network as the student network. Initially, the teacher model is pre-trained offline using State of Health degradation data to learn the degradation patterns. The directly measurable feature data (such as voltage, temperature, and current) is subsequently fed into the student network for online training and computation of a hard loss. the student network's output is then directed into the pre-trained the teacher network to compute a soft loss, thereby offering prior knowledge of degradation laws and steering the optimization process of the student network. Rigorous experiments are conducted utilizing various datasets, with the outcomes validating the superior estimation accuracy and degradation rule adherence of the model. Notably, among five different models, this model demonstrates the best performance on almost all datasets, achieving an RMSE of 0.0097 and an MAE of 0.0065 on Cell1 of the Oxford dataset. Moreover, the model also demonstrates robust performance across different usage scenarios, inclusive of multi-battery estimation. Furthermore, this paper also introduces a fine tuning method for State of Health predictions only using the first half of the data. Comparative analysis with other models underscores the competitiveness of the proposed model, showcasing its potential for broader application. [ABSTRACT FROM AUTHOR]

The widespread adoption of electric vehicles (EVs) underscores the urgent need for innovative approaches to estimate their lithium-ion batteries’ state of health (SOH), which is crucial for ensuring safety and efficiency. This study introduces SOH-TEC, a transformer encoder-based model that processes raw time-series battery and vehicle-related data from a single EV trip to estimate the SOH. Unlike conventional methods that rely on lab-experimented battery cycle data, SOH-TEC utilizes real-world EV operation data, enhancing practical application. The model is trained and evaluated on a real-world dataset collected over nearly three years from three EVs. This dataset includes reliable SOH labels obtained through periodic constant-current full-discharge tests using a chassis dynamometer. Despite the challenges posed by noisy EV real-world data, the model shows high accuracy, with a mean absolute error of 0.72% and a root mean square error of 1.17%. Moreover, our proposed pre-training strategies with unlabeled data, particularly SOH ordinal comparison, significantly enhance the model’s performance; using only 50% of the labeled data achieves results nearly identical to those obtained with the full dataset. Self-attention map analysis reveals that the model primarily focuses on stationary or consistent driving periods to estimate SOH. While the study is constrained by a dataset featuring repetitive driving patterns, it highlights the significant potential of transformer for SOH estimation in EVs and offers valuable insights for future data collection and model development.

Abstract The accurate prediction of a lithium-ion battery’s State of Health is of critical importance for efficient battery health management. Existing data-driven estimation methodologies grapple with issues such as high model complexity and a dearth of guidance from prior knowledge, which impose constraints on their efficacy. This work introduces a novel cross-modal distillation network for battery State of Health estimation, structured around a TransformerEncoder as the teacher network and a Convolutional Neural Network as the student network. Initially, the teacher model is pre-trained offline using State of Health degradation data to learn the degradation patterns. The directly measurable feature data (such as voltage, temperature, and current) is subsequently fed into the student network for online training and computation of a hard loss. the student network’s output is then directed into the pre-trained the teacher network to compute a soft loss, thereby offering prior knowledge of degradation laws and steering the optimization process of the student network. Rigorous experiments are conducted utilizing various datasets, with the outcomes validating the superior estimation accuracy and degradation rule adherence of the model. Notably, among five different models, this model demonstrates the best performance on almost all datasets, achieving an RMSE of 0.0097 and an MAE of 0.0065 on Cell1 of the Oxford dataset. Moreover, the model also demonstrates robust performance across different usage scenarios, inclusive of multi-battery estimation. Furthermore, this paper also introduces a fine tuning method for State of Health predictions only using the first half of the data. Comparative analysis with other models underscores the competitiveness of the proposed model, showcasing its potential for broader application.

Transportation electrification is accelerating the clean energy transition. Due to high efficiencies and energy density, Li‐ion batteries (LIBs) are used as on‐board energy carrier for battery electric vehicles (BEVs). LIBs are subject to rapid degradation due to fast‐charging, mechanical, electrical and thermal factors. Thus, state‐of‐health (SoH) prediction is required to optimize LIBs exploitation over their lifespan. An online accurate and easy‐of‐implementation battery SoH prediction and monitoring method for BEV applications is here presented. The method implements discrete wavelet transform (DWT) analysis to voltage profiles, measured while driving. Specifically, an extensive cycle aging experimental campaign on NCR 18650 cells was performed, applying two typical US test drives (urban and extra‐urban drive cycle, respectively) to the cells at different SoH. Moreover, tests carried out on LIBs at different temperatures demonstrated that temperature effect on the implemented DWT‐based method can be distinguished and separated from cycle aging effect. The proposed method allows a real‐time SoH estimation showing a good accuracy (MAE, ME and RMSE respectively result in 0.917, 2.897 and 1.32) without requiring high computational efforts. This allows to assess battery SoH during the driving. The method can also be extended to other chemistries requiring a dedicated experimental activity for the parameters tuning. [ABSTRACT FROM AUTHOR]

Data-driven models for battery state estimation require extensive experimental training data, which may not be available or suitable for specific tasks like open-circuit voltage (OCV) reconstruction and subsequent state of health (SOH) estimation. This study addresses this issue by developing a transfer-learning-based OCV reconstruction model using a temporal convolutional long short-term memory (TCN-LSTM) network trained on synthetic data from an automotive nickel cobalt aluminium oxide (NCA) cell generated through a mechanistic model approach. The data consists of voltage curves at constant temperature, C-rates between C/30 to 1C, and a SOH-range from 70% to 100%. The model is refined via Bayesian optimization and then applied to four use cases with reduced experimental nickel manganese cobalt oxide (NMC) cell training data for higher use cases. The TL models’ performances are compared with models trained solely on experimental data, focusing on different C-rates and voltage windows. The results demonstrate that the OCV reconstruction mean absolute error (MAE) within the average battery electric vehicle (BEV) home charging window (30% to 85% state of charge (SOC)) is less than 22mV for the first three use cases across all C-rates. The SOH estimated from the reconstructed OCV exhibits an mean absolute percentage error (MAPE) below 2.2% for these cases. The study further investigates the impact of the source domain on TL by incorporating two additional synthetic datasets, a lithium iron phosphate (LFP) cell and an entirely artificial, non-existing, cell, showing that solely the shifting and scaling of gradient changes in the charging curve suffice to transfer knowledge, even between different cell chemistries. A key limitation with respect to extrapolation capability is identified and evidenced in our fourth use case, where the absence of such comprehensive data hindered the TL process.

Lithium-ion (Li-ion) batteries have become essential in modern industries and domestic applications due to their high energy density and efficiency. However, they experience gradual degradation over time, which presents significant challenges in maintaining optimal battery performance and increases the risk of unexpected system failures. To ensure the reliability and longevity of Li-ion batteries in applications, various methods have been proposed for battery health monitoring and remaining useful life (RUL) prediction. This paper provides a comprehensive review and analysis of the primary approaches employed for battery health monitoring and RUL estimation under the categories of model-based, data-driven, and hybrid methods. Generally speaking, model-based methods use physical or electrochemical models to simulate battery behaviour, which offers valuable insights into the principles that govern battery degradation. Data-driven techniques leverage historical data, AI, and machine learning algorithms to identify degradation trends and predict RUL, which can provide flexible and adaptive solutions. Hybrid approaches integrate multiple methods to enhance predictive accuracy by combining the physical insights of model-based methods with the statistical and analytical strengths of data-driven techniques. This paper thoroughly evaluates these methodologies, focusing on recent advancements along with their respective strengths and limitations. By consolidating current findings and highlighting potential pathways for advancement, this review paper serves as a foundational resource for researchers and practitioners working to advance battery health monitoring and RUL prediction methods across both academic and industrial fields.

The increasing demand for reliable and safe Lithium-ion (Li-ion) batteries requires more accurate estimation of state of health (SOH) and remaining useful life (RUL) prediction. However, the inherent complexity and non-linear dynamics of Li-ion batteries present specific challenges to traditional methods of SOH modeling. Although particle filter (PF) techniques can handle nonlinear dynamics, they still face challenges, including particle degeneracy and loss of diversity, that reduce their ability to effectively model the nonlinear degradation mechanisms of batteries. To tackle these limitations, this paper presents a novel artificial intelligence-driven PF (AI-PF) technology for battery health modeling and prognosis. The main contributions of the AI-PF technique are as follows: (1) A novel dynamic sample degeneracy detection method is proposed to provide real-time assessment of particle weights so as to promptly identify degeneracy and improve computational efficiency. (2) An adaptive crossover and mutation strategy is proposed to reallocate low-weight particles and maintain particle diversity to improve modeling and RUL forecasting accuracy. The effectiveness of the AI-PF framework is validated through systematic evaluations carried out using benchmark models and well-recognized battery datasets.

Electric vehicles (EVs) have created a revolution in sustainable transportation. The number of EV users has increased significantly within a short period globally. EVs running largely on the battery source require large-capacity battery packs to handle the range anxiety. The primary lifetime of such batteries in EV applications is said to end when their capacity drops to 80% of their initial capacity. This is termed as the end of-life of these batteries. These batteries can still be utilized for secondary applications based on their remaining capacity. Batteries undergo many degradations throughout their lifecycle which affects their capacity. This paper carries out a detailed study on the major degradation factors like solid electrolyte interphase and lithium plating which results in loss of lithium inventory. These affect the capacity of the battery in the long run. Remaining useful capacity must be accurately estimated to identify if the cells are useful for the next phase or must be recycled. Many estimation techniques are available with attention rising towards data derivational methods due to their accuracy and their sensitivity towards battery degradation which thereby makes it easy to track them. Incremental capacity analysis is one such method which is discussed in detail in this paper. The method starts from the initial stage of data extraction and extends to the training set of the models. This method is greatly beneficial as it can reveal the deviations in battery behavior with the help of the valley peak locations and alterations in the slope. The quantitative insights make it an advantageous technique in the field of battery health monitoring and diagnostics. These are discussed in detail and validated by experimental analysis and results. This paper also discusses the market prospects, developments, various ageing mechanisms in batteries, applications, comparison with other estimation techniques and challenges related to secondary life applications. The complete analysis of the estimation method along with the detailed steps also aims to serve as a foundation for the upcoming developments and research in this field.

Li-ion batteries (LIBs), thanks to high efficiencies and energy density, represent the mainstream technology to replace traditional internal combustion vehicles with electric ones. However, LIBs state of health (SoH) should be investigated to avoid fast degradation due to fast-charging, electrical, mechanical and thermal factors. Therefore, SoH prediction and monitoring for battery electric vehicles is necessary for extending LIB lifespan and avoiding failures. In this paper, an accurate real-time SoH prediction and monitoring method, based on discrete wavelet (DWT) analysis, is investigated through an extensive experimental campaign considering the effect of temperature variation. Specifically, moving from cycle aging performed on Li-ion NCR 18650 cells and applying two typical US test drive cycles at different SoHs, three different operating temperatures (i.e., 0 °C, 20 °C and 30 °C) were investigated. Applying DWT on the gathered LIB voltage profiles, it is demonstrated that temperature effect on the implemented method is easily recognizable from the one of cycle aging. Moreover, suitable linearized functions are identified to refer DWT outcomes assessed at the operative temperature to a reference temperature, at which a suitable equation is previously identified to assess capacity fading. Due to its general validity the method can be extended to stationary applications.

Lithium-ion battery State of Health (SOH) estimation is an essential issue in battery management systems. In order to better estimate battery SOH, Extreme Learning Machine (ELM) is used to establish a model to estimate lithium-ion battery SOH. The Swarm Optimization algorithm (PSO) is used to automatically adjust and optimize the parameters of ELM to improve estimation accuracy. Firstly, collect cyclic aging data of the battery and extract five characteristic quantities related to battery capacity from the battery charging curve and increment capacity curve. Use Grey Relation Analysis (GRA) method to analyze the correlation between battery capacity and five characteristic quantities. Then, an ELM is used to build the capacity estimation model of the lithium-ion battery based on five characteristics, and a PSO is introduced to optimize the parameters of the capacity estimation model. The proposed method is validated by the degradation experiment of the lithium-ion battery under different conditions. The results show that the battery capacity estimation model based on ELM and PSO has better accuracy and stability in capacity estimation, and the average absolute percentage error is less than 1%.

Accurate state of health (SOH) estimation of lithium-ion batteries is critical for maintaining reliable and safe working conditions for electric vehicles (EVs). The machine learning-based method with health features (HFs) is encouraging for health prognostics. However, the machine learning method assumes that the training and testing data have the same distribution, which restricts its application for different types of batteries. Thus, in this paper, a deep learning neural network and fine-tuning-based transfer learning strategy are proposed for accurate and robust SOH estimation toward different types of batteries. First, a universal HF extraction strategy is proposed to obtain four highly related HFs. Second, a deep learning neural network consisting of long short-term memory (LSTM) and fully connected layers is established to model the relationship between the HFs and SOH. Third, the fine-tuning-based transfer learning strategy is exploited for SOH estimation of various types of batteries. The proposed methods are comprehensively verified using three open-source datasets. Experimental results show that the proposed deep learning neural network with the HFs can estimate the SOH accurately in a single dataset without using the transfer learning strategy where the mean absolute error (MAE) and root mean square error (RMSE) are constrained to 1.21% and 1.83%. For the transfer learning between different aging datasets, the overall MAE and RMSE are limited to 1.09% and 1.41%, demonstrating the reliability of the fine-tuning strategy. [ABSTRACT FROM AUTHOR]

Lead–acid batteries are widely used, and their health status estimation is very important. To address the issues of low fitting accuracy and inaccurate prediction of traditional lead–acid battery health estimation, a battery health estimation model is proposed that relies on charging curve analysis using historical degradation data. This model does not require the assistance of battery mechanism models or empirical degradation models, instead, it is combined with improved deep learning algorithms. A long short-term memory (LSTM) regression model was established, and parameter optimization was performed using the bat algorithm (BA). The experimental results show that the proposed model can achieve an accurate capacity estimation of lead–acid batteries. [ABSTRACT FROM AUTHOR]

Due to the complexity of the aging process, maintaining the state of health (SOH) of lithium-ion batteries is a significant challenge that must be overcome. This study presents a new SOH estimation approach based on hybrid Grey Wolf Optimization (GWO) with Bayesian Regularized Neural Networks (BRNN). The approach utilizes health features (HFs) extracted from the battery charging-discharging process. Selected external voltage and current characteristics from the charging-discharging process serve as HFs to explain the aging mechanism of the batteries. The Pearson correlation coefficient, the Kendall rank correlation coefficient, and the Spearman rank correlation coefficient are then employed to select HFs that have a high degree of association with battery capacity. In this paper, GWO is introduced as a method for optimizing and selecting appropriate hyper-p parameters for BRNN. GWO-BRNN updates the population through mutation, crossover, and screening operations to obtain the globally optimal solution and improve the ability to conduct global searches. The validity of the proposed technique was assessed by examining the NASA battery dataset. Based on the simulation results, the presented approach demonstrates a higher level of accuracy. The proposed GWO-BRNN-based SOH estimation achieves estimate assessment indicators of less than 1%, significantly lower than the estimated results obtained by existing approaches. The proposed framework helps develop electric vehicle battery prognostics and health management for the widespread use of eco-friendly and reliable electric transportation. [ABSTRACT FROM AUTHOR]

The algorithms currently applied to state of health(SOH) estimation require numerous data samples for training and the estimation effect is not good. To address this issue, this study proposed a least-squares support vector machine(LSSVM) algorithm based on the grey wolf optimization(GWO) algorithm to estimate the SOH using the grey relational analysis method to choose constant current charging time as the input characteristic. Considering the 18650 lithium cobalt oxide battery charge/discharge cycle test as an example, the established algorithm model was used to estimate the SOH of batteries with different capacity specifications under different proportions of training set samples. The estimated results were compared with those obtained by the LSSVM algorithm based on the grid search method and the LSSVM algorithm based on the particle swarm optimization algorithm. The experimental results showed that the LSSVM algorithm model based on the GWO algorithm is suitable for small-sample data and is characterized by small estimation errors; therefore, it is more effective for battery SOH.

This article addresses the objective, experimental design and methodology of the tests conducted for battery State of Health (SOH) estimation using an accelerated test method. For this purpose, 25 unused cylindrical cells were aged, by continual electrical cycling using a 0.5C charge and 1C discharge to 5 different SOH breakpoints (80, 85, 90, 95 and 100%). Ageing of the cells to the different SOH values was undertaken at a temperature of 25 °C. A reference performance test (RPT) of C/3 charge-discharge at 25 °C was performed when the cells were new and at each stage of cycling to define the energy capacity reduction due to increased charge-throughput. An electrochemical impedance spectroscopy (EIS) test was performed at 5, 20, 50, 70 and 95% states of charge (SOC) for each cell at temperatures of 15, 25 and 35 °C. The shared data includes the raw data files for the reference test and the measured energy capacity and the measured SOH for each cell. It contains the 360 EIS data files and a file which tabulates the key features of the EIS plot for each test case. The reported data has been used to train a machine-learning model for the rapid estimation of battery SOH discussed in the manuscript co-submitted (MF Niri et al., 2022). The reported data can be used for the creation and validation of battery performance and ageing models to underpin different application studies and the design of control algorithms to be employed in battery management systems (BMS).

Accurate simulation of characteristics performance and state of health (SOH) estimation for lithium-ion batteries are critical for battery management systems (BMS) in electric vehicles. Battery simplified electrochemical model (SEM) can achieve accurate estimation of battery terminal voltage with less computing resources. To ensure the applicability of life-cycle usage, degradation physics need to be involved in SEM models. This work conducts deep analysis on battery degradation physics and develops an aging-effect coupling model based on an existing improved single particle (ISP) model. Firstly, three mechanisms of solid electrolyte interface (SEI) film growth throughout life cycle are analyzed, and an SEI film growth model of lithium-ion battery is built coupled with the ISP model. Then, a series of identification conditions for individual cells are designed to non-destructively determine model parameters. Finally, battery aging experiment is designed to validate the battery performance simulation method and SOH estimation method. The validation results under different aging rates indicate that this method can accurately estimate characteristics performance and SOH for lithium-ion batteries during the whole life cycle.

The health and safety of the battery pack are directly influenced by the state of health of its cells. However, due to the aging inconsistency among cells and the limited measurability of physical quantities for cells within the battery pack, traditional approaches to state of health estimation of cell have significant limitations. This study introduces a machine learning approach for evaluating the state of health of cells within the battery pack. Firstly, a branch charging capacity estimator utilizing BiGRU is formulated, facilitating precise estimation of battery pack branch charging capacity across diverse charging conditions. Then, three categories of features, including aging features, inconsistency features, and operating condition features, are extracted based on aging experimental data at the battery pack level and battery pack branch charging capacity. These features are input into the support vector regression-based generic model, facilitating precise state of health estimation for all cells within the battery pack. The generalization of the model is validated under both five-stage constant current charging conditions and two-stage constant current charging conditions. Additionally, the discussion includes how the choice of model parameters affects the precision of cell state of health estimation. The method proposed enables precise monitoring of cell state of health within the battery pack, offering valuable potential for ensuring overall battery pack safety and issuing safety alerts for cells. • A high-precision branch charging capacity estimator has been developed. • The aging, inconsistency, and operating condition features are extracted. • High-precision estimation of SOH of all cells within battery pack can be achieved. • The model's generalization is validated under different operating conditions. [ABSTRACT FROM AUTHOR]

Accurately monitoring the state of health (SOH) of lithium-ion batteries (LIBs) is crucial for battery management systems (BMS), yet there lack of the possibility to fully use the random charging segments with any length. To this end, a residual convolution and transformer network (R-TNet) is proposed to enable an accurate LIB SOH estimation with the sparse dimension of feature in random segments, where the start and end voltage, the Ampere-hour (Ah) throughput, temperature, and current rate of a charging segment are required for the estimation task. Through the cross-attention mechanism of R-TNet , the operation condition and the position of the partial voltage can be integrated to enable the LIBs SOH estimation within a charging segment. To extend the flexibility with arbitrary charging behaviors, an ElasticNet -based feature transfer strategy is designed to use any charging length. 121 cells with different chemistries and cycling conditions are used to validate the performance of the proposed method. The results of the proposed method show that the root mean square error (RMSE) of SOH estimation can reach 1.6% even for a 50 mV voltage segment. • A residual convolution and transformer network is proposed to ensure SOH estimation of random voltage segments. • An ElasticNet -based feature transfer strategy is designed to use any length of voltage segments. • 121 cells with two chemistries and multiple operating conditions are used to validate the performance of the proposed method. [ABSTRACT FROM AUTHOR]

As an important performance indicator of battery management systems, lithium-ion battery state of health (SOH) information is crucial to ensure battery safety and extend battery lifetime. Aiming at the problems of feature extraction difficulty, low accuracy of long-term prediction, and poor parallel computing capability of general data-driven methods, this paper proposes a SOH estimation method for lithium-ion batteries based on incremental capacity analysis (ICA) and Transformer. First, the original incremental capacity (IC) curve of the battery is extracted based on the ICA method, and the original IC curve is processed using the dual filtering method of moving average smoothing filter plus Gaussian smoothing filter, which in turn extracts the peak features of the curve. Then, the Transformer network model based on the multi-head attention mechanism is built. Finally, the extracted peak features of the IC curve are used as model inputs, and the Transformer model is utilized to realize the SOH estimation of lithium-ion batteries. In this paper, experiments based on different input features, prediction starting points, and ambient temperatures are conducted using experimental data of lithium-ion batteries from three sources and analyzed in comparison with commonly used machine learning methods. The experimental results show that the SOH estimation method proposed in this paper has higher long-term prediction accuracy and better temperature adaptability than commonly used machine learning methods. • The dual filtering method of MASF plus GSF is used to smooth the IC curve. • Transformer model is proposed to estimate the SOH of lithium-ion batteries. • The peak features of the IC curves are used as model inputs. • The advantages of the proposed method were verified by several experiments. [ABSTRACT FROM AUTHOR]

Accurate state of health (SOH) estimation of lithium-ion batteries is critical for maintaining reliable and safe working conditions for electric vehicles (EVs). The machine learning-based method with health features (HFs) is encouraging for health prognostics. However, the machine learning method assumes that the training and testing data have the same distribution, which restricts its application for different types of batteries. Thus, in this paper, a deep learning neural network and fine-tuning-based transfer learning strategy are proposed for accurate and robust SOH estimation toward different types of batteries. First, a universal HF extraction strategy is proposed to obtain four highly related HFs. Second, a deep learning neural network consisting of long short-term memory (LSTM) and fully connected layers is established to model the relationship between the HFs and SOH. Third, the fine-tuning-based transfer learning strategy is exploited for SOH estimation of various types of batteries. The proposed methods are comprehensively verified using three open-source datasets. Experimental results show that the proposed deep learning neural network with the HFs can estimate the SOH accurately in a single dataset without using the transfer learning strategy where the mean absolute error (MAE) and root mean square error (RMSE) are constrained to 1.21% and 1.83%. For the transfer learning between different aging datasets, the overall MAE and RMSE are limited to 1.09% and 1.41%, demonstrating the reliability of the fine-tuning strategy.

Aimed at the uncertainty of charging starting and ending point caused by incomplete charging and discharging in practical applications of lithium-ion battery, an estimation method of battery health based on dual charging state factors is proposed. A battery aging experiment bench is built, and eight nickel-cobalt-manganese lithium-ion batteries are subjected to aging test. Different from the traditional single state factor estimation, the average value of equal time difference current at the front end of constant voltage charging curve and the equal amplitude voltage charging time at the end of constant current charging curve are selected under different aging conditions to construct health factors. The corresponding relationship between state of charge (SOC) and open circuit voltage (OCV) of the experimental battery in different aging states is analyzed and the correctness of health factor is proved by theoretical deduction and experimental results. An improved support vector regression model with a strong generalization ability is established, and the hyperparameters of the model are optimized through the particle swarm optimization algorithm. The results show that the proposed dual-charging health factor is closely related to battery capacity aging and attenuation. The improved support vector regression model can estimate the health status in different aging states in real time, and has the ability to characterize local capacity rebound change, which can be used as an effective method for estimating the state of health of an embedded battery management system.

Accurately estimating the state of health (SOH) of lithium batteries is a critical and challenging task in battery management systems. Data-driven models are widely used for SOH estimation but still suffer from the difficulty of balancing speed, accuracy, and adaptability. Therefore, this study constructs the dung beetle optimization algorithm to optimize the kernel extreme learning machine model. This paper addresses the issues of long iteration time and mismatches in kernel function mapping in data-driven models. To improve the model's generality, an adaptive learning kernel function is designed to complement the polynomial kernel function and form a joint function. This joint function is then introduced into a single implicit-layer extreme learning machine, which achieves fast speed and strong adaptive capability. To enhance the algorithmic parameter search capability, the optimal Latin hypercube idea, and the Osprey algorithm's global exploration strategy are introduced, which effectively improves the algorithm's global search capability. Additionally, it successfully regulated the positional update through the design of the logarithmic weighting factor, which improved the local search and convergence capabilities of the algorithm. The experiment validates the effectiveness and rationality of the proposed model for advancing battery management system applications. • Innovating an Adaptive learning kernel function for union Poly to construct a joint kernel function. • Using joint kernel function for extreme learning machine. • Optimizing the dung beetle optimization algorithm under multiple constraints. • Constructed LOWDBO-PAKELM model for SOH estimation with a high adaptation range. [ABSTRACT FROM AUTHOR]

Accurate and precise online estimation of the state of health (SOH) is crucial when managing lithium-ion batteries. Most existing SOH estimation methods rely on supervised learning algorithms utilizing large amounts of labeled data. However, lithium-ion batteries are typically operated under dynamic conditions, including significant amounts of unlabeled charging or discharging data in online application scenarios. To fully utilize these data, we propose an adaptive semi-supervised self-learning teacher-student model (AS3LTSM) method for online SOH estimation. First, four physically interpretable health indicators (PIHIs) are extracted from the voltage and current data. The Pearson correlation coefficient (PCC) is used to assess significant associations between PIHIs and the SOH. Regressive and autoregressive long short-term memory (LSTM) models are selected as the teacher and student networks. Knowledge is transferred from the teacher to the student through pseudolabels, which guide the updating and evolution of the student network. Furthermore, a self-learning strategy and a retraining process for improving the long-term estimation accuracy are proposed. Two public datasets are used for comparison and ablation experiments. Experimental analysis validates the improved effectiveness and performance of the proposed method, with the RMSE and MAPE of the three experimental groups all within 1.3 % and 1.29 %, respectively. • An adaptive semi-supervised self-learning method is proposed for online SOH estimation. • An amount of online unlabeled data is utilized to improve SOH estimation accuracy. • The self-learning and re-training process are utilized to reduce long-term cumulative error. • An ablation and experimental results validate the performance and robustness. [ABSTRACT FROM AUTHOR]

Frequency response analysis (FRA) methods are commonly used in the field of State of Health (SOH) estimation for Lithium-ion batteries (Libs). However, identifying their appropriate application scenarios can be challenging. This paper presents four FRA techniques, including electrochemical impedance spectra (EIS), mid-frequency and low-frequency domain equivalent circuit model (MLECM), distribution of relaxation time (DRT) and non-linear FRA (NFRA) technique. This paper proposes two estimation frameworks, machine learning and curve fitting, to be applied to each of the four techniques. Eight SOH estimation models are developed by linking the extracted feature parameters to the battery capacity variations. The paper compares the accuracy of estimation, estimation range, and other properties of the eight models. Application scenarios are identified for the techniques by using three classification methods: different estimation frameworks, frequency response linearity, and impedance technique. The results demonstrate that MLF is recommended for scenarios with a large amount of battery data, while CFF is recommended for scenarios with a small amount of data. NFRA could be applied to electric vehicle power batteries, while LFRA is recommended to be used for retired batteries. EIS method is recommended for complex and dynamic scenarios, while non-EIS method is recommended for scenarios that require high accuracy. • Four FRA techniques, Impedance, MLECM, DRT and NFRA-based method are developed. • Two frameworks, MLF and CFF, are applied to each of these four techniques. • The estimation accuracy and estimation range of the eight models are compared. • Application scenarios are identified by using three classification methods. [ABSTRACT FROM AUTHOR]

Lithium-ion Batteries (LiB) have a wide range of applications in daily life. However, as they get used over time, battery degradation becomes inevitable, which can lead to a drop in performance and a reduction in the battery's cycle life. The State of Health (SoH) is widely regarded as the health indicator for the battery pack. In Electric Vehicle (EV) applications, the EV user defines the lower limit of SoH when they experience that the battery no longer supports the EV; at that point, the battery is said to be translated from first life to second life. The SoH estimations of Second Life Batteries (SLB) have plenty of uncertainties, such as the availability of battery's previous history, non-uniform degradation in the EV application, variations in chemistry, and charging protocols defined by vehicle manufacturers, making the SoH estimation of SLB a challenging task. This paper discusses the equipment, timelines, computational complexity, health indicators, and list of parameters that need to be considered for the SoH estimation of SLB. The SoH estimation methods are classified into direct and indirect techniques. Direct assessment techniques involve cyclic ageing experiments followed by dismantling the battery for microscopic studies performed by previous researchers that were explained. Indirect assessment techniques include physical and chemical based approach, electrical, and Artificial Intelligence (AI)-based methods that estimate SoH indirectly through incremental, differential approaches and other parameters such as Integrated Voltage (IV) and Probability Density Function (PDF). Health indicator identifications play a vital role in indirect assessment methods to gain critical insights regarding battery degradation. The challenges involved in SoH estimation are categorized into equipment requirements, parameters, SoH accuracy and efforts required to compute SoH, which are discussed. Of all the SoH estimation methods, comparison of such methods in First Life Batteries (FLB) and SLB perspectives are discussed. To estimate the SoH of SLB, this paper explains all aspects, such as computational methods, filtering data, data sampling frequency, and the need for a specific algorithm to post-process the battery test data. Equipment availability and timelines are interrelated with the cost incurred in the SoH estimation of SLB. The efficacy and practicality of SoH estimation methods that are proposed for SLB is discussed. Overall, this paper provides necessary insights into the parameters required for SoH estimation and the computational and experimental methods that can be considered for estimating the SoH of SLB while some of the methods are applicable to FLB as well. • Review of State of Health (SoH) estimation methods for lithium-ion battery pack translating from first life to second life. • Critical analysis of equipment's and test protocols subjected to cyclic ageing. • Classification of SoH estimation methods in the form of physical and chemical based approach, electrical and Artificial Intelligence (AI) based techniques. • Listed the parameters acquired from battery to be considered in second life SoH estimation. • Presented the challenges associated with SoH methods for Second Life Batteries (SLB). [ABSTRACT FROM AUTHOR]

With the rapid development of new energy vehicles, it is recognized that predicting the state of health (SoH) of lithium-ion battery is crucial for ensuring the safety of networked vehicles. However, the selection of health indicators greatly influences the accuracy of SoH prognostics. To obtain an accurate estimation of SoH, this paper proposes an SoH estimation model based on incremental capacity features. First, the incremental capacity curve is extracted from battery discharge data and filtered using a Gaussian filtering algorithm to remove noise. Second, statistical features extracted from the incremental capacity curve are considered health factors, and multiple optimal features are selected using Pearson’s correlation coefficient. Finally, the innovative integration of spatiotemporal feature extraction with advanced pattern recognition and nonlinear modeling led to the proposal of a hybrid Convolutional Neural Network–Multi-Layer Perceptron (CNN-MLP) model for estimating the SoH of lithium-ion batteries. To validate the high accuracy of the proposed method, experiments are conducted using the CALCE battery dataset and compared with other popular models. The experimental results indicate that the proposed method can predict the SoH of the battery with superior performance, such as higher speed and accuracy. [ABSTRACT FROM AUTHOR]

The evolution of performance degradation has become a major obstacle to the long-life operation of the Solid Oxide Fuel Cell (SOFC) system. The feasibility of employing degradation resistance to assess the State of Health (SOH) is proposed and verified. In addition, a real-time Unscented Kalman Filter (UKF) based SOH estimation method is further proposed to eliminate the disturbance of calculating the SOH directly utilizing measurement and electric balance model. The results of real-time SOH estimation with an UKF under constant and varying load conditions demonstrate the feasibility and effectiveness of the SOFC performance degradation assessment method. [ABSTRACT FROM AUTHOR]

To solve the problems in estimating the state of health (SOH) of Li-ion batteries due to real-time estimation difficulty and low precision under various operating conditions, the variations of the SOH caused by increases of the internal resistance have been analyzed. Based on the second-order RC equivalent circuit model, the short-term effect of the state of charge (SOC) on the internal resistance was considered, which was set under the discharge condition. In addition, the variation of the internal resistance was analyzed in two intervals of 0–1 s and 1–10 s. The extended Kalman filter (EKF) algorithm was improved to present a novel improved Kalman filter (IKF) algorithm to accurately predict the long-term internal resistance under different operating conditions. A computational formula based on the internal-resistance increasing was established and the SOH was estimated. The error of the calculated result when compared with the forgetting factor least square method based on the internal-resistance increasing was controlled to within 4.0% under the HPPC condition, 3.0% under the BBDST condition, and 6.0% under the DST condition. The proposed algorithm has good convergence, helps improve the SOH estimation, and encourages the application of Li-ion batteries. [ABSTRACT FROM AUTHOR]

The prognostic and health management of the batteries continued to attract interest from automobile manufacturers as the key for lowering life-cycle costs, reducing unexpected power outages, and one of the most important and efficient ways for energy storage for electric vehicle applications. Indeed, an effective battery health monitoring depends on accurate estimation of state of health (SOH). However, the SOH cannot be directly measured by sensors in the battery management system. Moreover, the SOH estimation based on a standard resistor–capacitor (RC) battery model is not so accurate because a RC model is obtained with some approximations and without taking into account more detailed knowledge about the chemical reactions happening inside the battery. In this paper, a combined battery modeling and SOH estimation method over the lifespan of a nickel–metal hydride (Ni–MH) battery is proposed. First, a fuzzy c-regression model based on Euclidean particle swarm optimization is applied to modeling a Ni–MH battery. Second, the SOH monitoring is determined according to the discharge rate of the battery model. The performance of the proposed method has been analyzed through the modeling and the estimation of the SOH using a real data set of the Ni–MH battery. [ABSTRACT FROM AUTHOR]

Within battery management systems, the state of health of lithium-ion batteries is a key and vital enabler to ensure battery safety and efficiency. However, the accurate state of health estimation is still a critical but challenging task, and the complex electrochemical attributes underlying the degradation processes of lithium-ion batteries are not directly available. In response to this challenge, this study proposes a charging-feature-based model to realize state of health estimation by Gaussian process regression. In this approach, two features are extracted only from the monitoring parameters obtained from charging current and voltage curves. These extracted features have been demonstrated to be correlated with the state of health. Subsequently, a regression model with a 2-dimensional linear mean function and a new double-covariance function is developed to improve estimation performance. Consequently, the proposed model effectively tracks both global and local degradation trends synchronously. Finally, the reliability and accuracy of the proposed model are verified using two different batteries datasets. The results illustrate that the proposed model is capable of realizing accurate batteries' state of health estimation, thereby outperforming other counterparts in uncertainty representation and estimation errors, whether under static profiles or dynamic profiles. [ABSTRACT FROM AUTHOR]

Accurate estimation of State of Health (SOH) is crucial to ensure optimal performance and safe operation of lithium-ion battery. This paper proposes a Stacking ensemble learning paradigm for SOH estimation. The Stacking ensemble learning increases adaptability to different features by using base learners with different structures, reducing the risk of overfitting. The model utilizes random vector functional link (RVFL) and active state tracking long-short-term memory network (AST-LSTM) as base learners, where AST-LSTM actively tracks long-term information of lithium-ion battery, and RVFL acts as the meta-learner for stacking. The random vector functional link network helps to avoid the problem of gradient vanishing that is commonly encountered in neural networks due to the gradient descent principle. To further improve estimation accuracy, Singer initialization method and dimension learning method are employed to enhance the Heap-based optimization (HBO) algorithm. In this study, the IHBO algorithm is used to optimize the hyperparameters of the model. Comparing with other methods, the hybrid model proposed in this paper demonstrates superior estimation performance under different operating conditions: at a temperature of 24 °C with a discharge current of 1 A, at a temperature of 4 °C with a discharge current of 1 A, and at a temperature of 4 °C with a discharge current of 2 A. The highest RMSE of the proposed method for the three working conditions are 0.006, 0.01, and 0.017, respectively. Therefore, the proposed Stacking ensemble learning is feasible for SOH estimation of lithium-ion battery and can better adapt to lithium-ion battery data under different operating conditions. • A Stacking ensmeble learning strategy is proposed for SOH estimation. • The improved HBO algorithm is introduced to optimize the Stacking model. • SOH for lithium batteries under three different operating conditions is estimated. • Six benchmark models are used to verify the performance of the proposed model. [ABSTRACT FROM AUTHOR]

The rapid development of lithium-ion battery (LIB) technology promotes its wide application in electric vehicle (EV), aerospace, and mobile electronic equipment. During application, state of health (SOH) of LIB is crucial to enhance stable and reliable operation of the battery system. However, accurate estimation of SOH is a tough task, especially in its large-scale application. Thus far, a variety of works on the estimation of SOH of LIB have been proposed, along with several review studies that aim to summarize the current research status. However, there are some deficiencies in prior reviews, such as unclear classification, incomplete summary, and insufficient evaluation of estimation methods. Thus, to resolve the shortcomings, the enumeration method is used to fully screen published works related to SOH estimation, and a total of one hundred and ninety relevant studies are investigated for a thorough review and discussion. Besides, the definition of SOH from different perspectives and three representative battery models are summarized, respectively. Meanwhile, twenty commonly used evaluation criteria and two explicit SOH estimation schemes are comprehensively introduced, which all are tabulated in detail for systematic evaluation and fair comparison. Finally, the main problems and challenges in SOH estimation are fully discussed, meanwhile, three promising future development trends are proposed and some essential SOH public datasets are summarized. In general, this review is envisioned to offer insightful guidance to researchers or engineers working on SOH estimation and related research, thus further promoting the development of SOH estimation technology and exploration of potential research direction. [Display omitted] • Three typical battery modeling methods in SOH estimation are summarized in detail. • Summarize a variety of evaluation criteria applied to SOH estimation. • Various SOH estimation methods are divided into two general groups. • The characteristics of SOH estimation methods are illustrated and compared. • Suggestions and prospects for the further development of SOH estimation of LIB are put forward. [ABSTRACT FROM AUTHOR]

Summary: The state of charge and state of health estimations are two of the most crucial functions of a battery management system, which are the quantified evaluation of driving mileage and remaining useful life of electric vehicles. This paper investigates a novel data‐driven–enabled battery states estimation method by combining recurrent neural network modeling and particle‐filtering–based errors redress. First, a recurrent neural network with long‐short time memory is employed to learn the long‐term nonlinear relation between batteries states and measurable signals of lithium‐ion batteries, such as current, voltage, and temperature. Second, to denoise the estimation errors of the neural network model, particle filtering is employed to smooth the state of charge estimation results. Third, the terminal voltage difference of battery is highly related to the internal resistance of the battery, which is thus taken as a new input to track the internal resistance of the battery. The performance of the proposed method is verified by multiple comparisons with conventional techniques under randomized loading profiles and different temperatures. [ABSTRACT FROM AUTHOR]