Life Of Lithium-ion Batteries Research Articles

A multi-scale learning approach for remaining useful life prediction of lithium-ion batteries based on variational mode decomposition and Monte Carlo sampling

Reliable and accurate prediction of remaining useful life for lithium-ion batteries has tremendous significance, since they can alleviate users' anxiety about mileage and safety. However, accuracy and reliability of remaining useful life prediction are affected by capacity regeneration and uncertainty quantification. In this study, we propose an approach to predict the remaining useful life of lithium-ion batteries, where multi-scale learning is developed to catch the uncertainty and capacity regeneration. Specifically, the multi-scale learning approach based on Gaussian process regression and dropout-Monte Carlo gated recurrent unit is applied to establish accurate prediction model with uncertainty quantification. Meanwhile, the optimal charging voltage interval is extracted with a high correlation coefficient. The variational mode decomposition is selected to multi-scale decompose the proposed health indicator as intrinsic mode functions and residual term. Finally, the observed data has been selected to verify the accuracy and robustness of the proposed method. Compared to the existing single data-driven methods, the proposed method can obtain high accuracy and strong robustness for RUL prediction with root mean square error limited below 3%.

Remaining useful life prediction of lithium-ion batteries based on the Support Vector Regression Optimized and Enhanced Whale Optimization Algorithm

Accurate and reliable predictions of the remaining useful life (RUL) of lithium-ion batteries are crucial for ensuring the proper functioning and longevity of lithium-ion battery systems. Predicting the RUL in its early stages can be challenging and difficult for conventional RUL prediction techniques. To tackle this challenge, this paper proposes a novel approach that integrates five carefully chosen health features with a Support Vector Regression Optimized model and Enhanced Whale Optimization Algorithm (EWOA-SVR) to predict the future capacity trend and subsequently estimate the RUL using XGBoost method according to the predicted capacity. The proposed approach aims to achieve early and accurate capacity estimation and RUL prediction for lithium-ion batteries. To begin, this method extracts some health features during the relaxation process and during the constant-current and constant-voltage charge (CC-CV) process. To demonstrate the strength of the correlation between the health features and capacity, both Pearson and Spearman rank correlation analysis are employed in this study. Then based on the measurable health features and faded capacity data, Ant Lion Optimization (ALO), Whale Optimization Algorithm (WOA) and EWOA are used to select suitable penalty factor and kernel parameter in the SVR algorithm to improve the performance of SVR. With the well-trained model, the future capacity can be predicted with an accuracy of MAPE within 1%, and the RUL prediction error is lower than 11 cycles. The results suggest that the proposed model can achieve a precise and dependable early estimation of the future capacity and RUL prediction.

Remaining useful life prediction method of lithium-ion batteries is based on variational modal decomposition and deep learning integrated approach

A common method based on variational modal decomposition (VMD) and an integrated depth model is proposed to address the problem that it is difficult to precisely anticipate the remaining useful life (RUL) of lithium-ion batteries (LIBs). Initially, VMD is employed to decompose the LIBs capacity data in multiple scales to obtain the signal's global degradation tendency and local random fluctuation components. Then, the global degradation trend and each fluctuation component are modeled using an echo state network (ESN) and a Bayesian optimized long short-term memory (LSTM) network, respectively. The final LIBs RUL prediction results are obtained by integrating the prediction outcomes. On three public LIBs datasets with distinct degradation characteristics, the performance of the proposed model is tested, and alternative prediction algorithms are compared. The results of the experiments show that the proposed model's maximum average absolute percentage error does not exceed 0.43%. The average relative error not more than 0.5%, indicating great accuracy and stability in prediction.

Remaining Useful Life Prediction of Lithium-Ion Batteries by Using a Denoising Transformer-Based Neural Network

In this study, we introduce a novel denoising transformer-based neural network (DTNN) model for predicting the remaining useful life (RUL) of lithium-ion batteries. The proposed DTNN model significantly outperforms traditional machine learning models and other deep learning architectures in terms of accuracy and reliability. Specifically, the DTNN achieved an R2 value of 0.991, a mean absolute percentage error (MAPE) of 0.632%, and an absolute RUL error of 3.2, which are superior to other models such as Random Forest (RF), Decision Trees (DT), Multilayer Perceptron (MLP), Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Dual-LSTM, and DeTransformer. These results highlight the efficacy of the DTNN model in providing precise and reliable predictions for battery RUL, making it a promising tool for battery management systems in various applications.

(Invited) Improving Calendar Life of Si-Based Lithium-Ion Batteries By FEC/VC-Free Electrolytes

Silicon (Si) has been regarded as one of the most promising anode materials for the next generation lithium-ion batteries (LIBs) with high energy density because it has 10 times higher theoretical specific capacity (4200 mAh/g) than that of graphite. However, severe volume change (~300%) of Si during lithiation and delithiation hinders the practical application of Si anode. In the last decade, significant progresses have been made on the specific energy and cycle life of Si based LIBs. However, calendar-life of these batteries is still far less than 10-year target required by DOE’s electrical vehicle (EV) project.In the current work, the role of FEC/VC in conventional carbonate electrolytes has been investigated. It has been found that FEC/VC causes large impedance growth on both Si and graphite anodes at 55 oC. The cut-off voltage of NMC622 has been found to significantly affect its impedance. NMC622 at 4.2V shows no impedance growth at 55 °C for 8 days, while severe impedance growth is observed at 4.4V. To address the instability of both SEI and CEI at elevated temperature, a new Localized High Concentration Electrolyte (LHCE) has been developed to improve the calendar life of Si based LIBs by effectively limiting impedance growth of Si/NMC622 full cells. In addition, the cycle life of the full cell is improved. The mechanism of Si full cells with improved calendar life will also be discussed.

Scalable Direct Recycling of Cathode Black Mass from Spent Lithium-Ion Batteries

End of life (EoL) lithium-ion batteries (LIBs) are piling up at an intimidating rate which is alarming for the environmental health and balance. The rise of renewable technologies such as electric vehicles (EV) needs to be balanced with LIB recycling to reduce the carbon footprint, ensure sustainable development, and create a circular supply chain. LiNixCoyMnzO2 (NCM) cathode materials being a dominant chemistry in high energy LIBs makes a huge portion of this waste accumulation. Direct recycling is one of the most promising ways to turn this waste to wealth but has been limited to lab-scale due to lack of robustness, namely the tedious pretreatment required which involve toxic organic solvents. Herein, we have demonstrated a process which integrates the pretreatment and relithiation of the cathode black. Cathode material from EoL EV batteries is treated in 100g per batch operation and the regenerated cathode active material demonstrates 100% electrochemical performance recovery, with 91% yield rate, and showed promise to further scale up. No use of toxic organic solvents makes our process unique with minimal environmental impact. This process has the advantages of integration, scalability, and universality, which clears the barricade for the direct recycling to move from lab to industry scale with considerable profitability.

CFD simulation of effect spacing between lithium-ion batteries by using flow air inside the cooling pack

The CFD simulation of this study shows the impact of airflow with varying Reynolds numbers on heat transfer improvement with cooling lithium-ion batteries at varied battery row spacing. Air was used as a cooling fluid to remove heat from lithium-ion batteries by flowing within the cooling pack during testing of four different spacing ranges (S = 1–4 mm). The Reynolds numbers vary from15,000 to 30,000 in the current analysis. The average Nu numbers are augmented through increasing Reynolds numbers for all types of cooling fluids. The CFD results show that the Nu number increases as spacing increases. The temperature of the single cell drops with increasing the Re or the flow rate since the liquid in the pack exchange with a new cold HTF at a faster rate. The average Nu number is directly proportional to the Re number and the spacing between the neighbor cells. Further, the temperature of the cells located closer to the inlet section is always lower than those closer to the outlet section because of the temperature differences between the cells and the HTF. This study has confirmed that the flowing air with high Re and highest spacing S = 4 mm has a major impact on thermal dissipation and rapid heat transfer improvement from lithium-ion batteries in the cooling pack, which leads to improved electrical performance and increase the life of lithium-ion batteries.

A Method for Predicting the Life of Lithium-Ion Batteries Based on Successive Variational Mode Decomposition and Optimized Long Short-Term Memory

Accurately predicting the remaining lifespan of lithium-ion batteries is critical for the efficient and safe use of these devices. Predicting a lithium-ion battery’s remaining lifespan is challenging due to the non-linear changes in capacity that occur throughout the battery’s life. This study proposes a fused prediction model that employs a multimodal decomposition approach to address the problem of non-linear fluctuations during the degradation process of lithium-ion batteries. Specifically, the capacity attenuation signal is decomposed into multiple mode functions using successive variational mode decomposition (SVMD), which captures capacity fluctuations and a primary attenuation mode function to account for the degradation of lithium-ion batteries. The hyperparameters of the long short-term memory network (LSTM) are optimized using the tuna swarm optimization (TSO) technique. Subsequently, the trained prediction model is used to forecast various mode functions, which are then successfully integrated to obtain the capacity prediction result. The predictions show that the maximum percentage error for the projected results of five unique lithium-ion batteries, each with varying capacities and discharge rates, did not exceed 1%. Additionally, the average relative error remained within 2.1%. The fused lifespan prediction model, which integrates SVMD and the optimized LSTM, exhibited robustness, high predictive accuracy, and a degree of generalizability.

Predicting the Cycle Life of Lithium-Ion Batteries Using Data-Driven Machine Learning Based on Discharge Voltage Curves

Battery degradation is a complex nonlinear problem, and it is crucial to accurately predict the cycle life of lithium-ion batteries to optimize the usage of battery systems. However, diverse chemistries, designs, and degradation mechanisms, as well as dynamic cycle conditions, have remained significant challenges. We created 53 features from discharge voltage curves, 18 of which were newly developed. The maximum relevance minimum redundancy (MRMR) algorithm was used for feature selection. Robust linear regression (RLR) and Gaussian process regression (GPR) algorithms were deployed on three different datasets to estimate battery cycle life. The RLR and GPR algorithms achieved high performance, with a root-mean-square error of 6.90% and 6.33% in the worst case, respectively. This work highlights the potential of combining feature engineering and machine learning modeling based only on discharge voltage curves to estimate battery degradation and could be applied to onboard applications that require efficient estimation of battery cycle life in real time.

Optimal sizing and lifetime investigation of second life lithium-ion battery for grid-scale stationary application

The renewable energy sources (RESs) integration into the grid system aims to solve the problem of power shortage and satisfy the increasing demand with the production of surplus energy. However, the intermittent nature of these RESs (solar and wind) is a challenge to integrate with the grid system without the deployment of mitigating solutions. The re-use of first life-end-of-life (FL_EoL) electric vehicle batteries known as second life batteries (SLBs) is therefore proposed as a reliable solution to resolve this problem, satisfying the techno-economic requirements of stationary applications. Though various studies performed on the technical viability evaluation of SLBs, most of them have not considered the field data of existing stationary plants and were found to be limited with simulation-based aging results. On the other hand, few of the studies have performed the second life aging test with limited cycles considering the end of test conditions rather than using the cells reached to their end of first life criteria. Therefore, in this paper, the prolonged cycling aging of SLBs is conducted (both cell-level and module-level aging), focusing on aging characteristics of SLBs by using a real-life renewable power smoothing profile extracted from an existing photovoltaic grid-connected system (PVGCS) installed in Ethiopia. Prior to the aging characterization of SLBs, assessment of the selected stationary application requirement, optimal sizing of the storage battery and cycling profile definition are performed using advanced moving average ramp rate controller (MARRC) algorithm. The proposed method of MARRC is used to determine the optimal SLBs capacity by analyzing the relationship between ramp-rate, initial battery capacity and window sizes of moving average control method in terms of time series. The algorithm addresses the main objectives of reducing unnecessary battery cycling, mitigating ramp-rate violations and meeting minimum storage requirements for the second-life application. From the result of the battery sizing study, the desired optimal battery capacity and the permissible ramp-rate limit below 10 %/minute is achieved. Moreover, the aging characterization and lifetime model validation results with a root mean square error (RMSE) of 1.5 % shows that, the technical performance of the SLBs is found to be promising with an efficiency of >90 % interpreted in terms of the service year that the SLBs can provide. Therefore, SLBs can be regarded as a viable solution for integrating RESs with the grid system for the power variability smoothing application.

A Transferable Prediction Approach for the Remaining Useful Life of Lithium-Ion Batteries Based on Small Samples

Predicting the remaining useful life (RUL) of batteries can help users optimize battery management strategies for better usage planning. However, the RUL prediction accuracy of lithium-ion batteries will face challenges due to fewer data samples available for the new type of battery. This paper proposed a transferable prediction approach for the RUL of lithium-ion batteries based on small samples to reduce time in preparing battery aging data and improve prediction accuracy. This approach, based on improvements from the adaptive boosting algorithm, is called regression tree transfer adaptive boosting (RT-TrAdaBoost). It combines the advantages of ensemble learning and transfer learning and achieves high computational efficiency. The RT-TrAdaBoost approach takes the charging voltage and temperature curve as input and utilizes the classification and regression tree (CART) as the base learner, which has better feature capture ability. In the experiment, the working condition migration experiment and battery type migration experiment are conducted on non-overlapping datasets. The verified results revealed that the RT-TrAdaBoost approach could transfer not only the battery aging knowledge between various working conditions but also realize the RUL migration prediction from lithium iron phosphate battery to lithium cobalt oxide battery. The analysis of error and computation time demonstrates the proposed method’s high efficiency and speed.

State of Health Prediction for Lithium-Ion Batteries through Curve Compression and CatBoost

This study proposes a novel approach for predicting the State of Health (SoH) and Remaining Useful Life (RUL) of lithium-ion batteries. The low accuracy of SoH and RUL is due to the challenges of establishing effective feature engineering for battery attributes. To address this issue, a SoH and RUL prediction model based on curve compression and CatBoost is proposed. Firstly, an improved threshold selection method based on curvature analysis is introduced to enhance the compression performance of battery attributes under different cycles. Secondly, to ensure that the extracted feature sequences have the same length, spline interpolation and local anomaly factor detection techniques are utilized to fill or eliminate feature points for feature length normalization. Finally, a dynamic time regularization algorithm is applied to calculate the shortest distance between the feature sequence and the original curve to determine the optimal feature length for input into the CatBoost prediction model. The experimental results demonstrate that the proposed approach outperforms other prediction models in the research object dataset, achieving R2 values higher than 0.98 and MSE values around 1 × 10−5. The proposed approach also achieves better prediction results in the validation object dataset, indicating its strong generalization capability. Additionally, the proposed model shows significant robustness by accurately predicting SoH and RUL under noisy environmental conditions. Overall, the proposed model shows significant potential to accurately predict SoH and RUL by efficiently addressing the challenges associated with feature engineering for battery attributes, reducing the impact of background noise on prediction results, and exhibiting strong robustness.

Early prediction of remaining useful life for lithium-ion batteries based on CEEMDAN-transformer-DNN hybrid model

A reliable and safe energy storage system utilizing lithium-ion batteries relies on the early prediction of remaining useful life (RUL). Despite this, accurate capacity prediction can be challenging if little historical capacity data is available due to the capacity regeneration and the complexity of capacity degradation over multiple time scales. In this study, data decomposition, transformers, and deep neural networks (DNNs) are combined to develop a model of RUL prediction for lithium-ion batteries. Complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is used for battery capacity sequential data to account for the capacity regeneration effect. The transformer networks are leveraged to predict each component of capacity regeneration thus improving the model's ability to handle long sequences while reducing the amount of data. The global degradation trend is predicted using a deep neural network. We validated the early prediction performance of the model using two publicly available battery datasets. Results show that the prediction model only uses 25%-30% data to achieve high accuracy. In the two public data sets, the RMSE errors were 0.0208 and 0.0337, respectively. A high level of accuracy is achieved with the model proposed in this study, which is based on fewer capacity data.

Analysis of strategies to maximize the cycle life of lithium-ion batteries based on aging trajectory prediction

Analysis of strategies to maximize the cycle life of lithium-ion batteries based on aging trajectory prediction

Experimental Study on Remaining Useful Life Prediction of Lithium-Ion Batteries Based on Three Regression Models for Electric Vehicle Application

This paper presents three regression models that predict the lithium-ion battery life for electric cars based on a supervised machine learning regression algorithm. The linear regression, bagging regressor, and random forest regressor models will be compared for the capacity prediction of lithium-ion batteries based on voltage-dependent per-cell modeling. When sufficient test data are available, three linear regression learning algorithms will train this model to give a promising battery capacity prediction result. The effectiveness of the three linear regression models will be demonstrated experimentally. The experiment table system is built with an NVIDIA Jetson Nano 4 GB Developer Kit B01, a battery, an Arduino, and a voltage sensor. The random forest regressor model has evaluated the model’s accuracy based on the average of the square of the difference between the initial value and the predicted value in the data set (MSE (mean square error)) and RMSE (root mean squared error), which is smaller than the linear regression model and bagging regressor model (MSE is 516.332762; RMSE is 22.722957). The linear regression model with MSE and RMSE is the biggest (MSE is 22060.500669; RMSE is 148.527777). This result allows the random forest regressor model to remain helpful in predicting the life of lithium-ion batteries. Moreover, this result allows rapid identification of battery manufacturing processes and will enable users to decide to replace defective batteries when deterioration in battery performance and lifespan is identified.

A method to prolong lithium-ion battery life during the full life cycle

A method to prolong lithium-ion battery life during the full life cycle

RUL prediction of lithium ion battery based on CEEMDAN-CNN BiLSTM model

With the wide application of lithium ion batteries, the importance of life prediction is also highlighted. The prediction of the remaining life of lithium ion battery is an important part of its health management, and accurate prediction can improve the safety of equipment. In this paper, a method for predicting the residual life of lithium ion batteries based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), One-dimensional Convolutional Neural Network (1D CNN) and Bi-directional Long Short-Term Memory (BiLSTM) neural network is proposed. The capacity is selected as the health factor, and then CEEMDAN is used to decompose the complex and unstable data to obtain stable components. One-dimensional Convolutional Neural Network (1D CNN) is used to deeply mine the capacity data of lithium-ion batteries. Finally, BiLSTM neural network modeling is used to predict the remaining useful life (RUL) of lithium-ion batteries. The NASA data set is used for testing and prediction comparison with BiLSTM model and CNN-BiLSTM model. The prediction results show that CEEMDAN-CNN BiLSTM model has higher prediction accuracy.

Performance of Lithium-Ion Pouch Cells with Silicon Composite Anodes under External Mechanical Pressure

The external forces have significant effects on the mechanical and electrochemical properties of lithium-ion pouch cells with silicon composite electrodes, but the behaviors under the constant pressure condition have been lacking for a long time. In this study, on the basis of the in situ testing equipment that can provide good precision and good stability under constant pressure conditions, the mechanical and electrochemical performance of silicon-composite-based batteries is studied in detail. First, the optimal constant pressure for the battery is proposed with the value of 0.02 MPa. Then, the cycle life is started on the basis of this pressure, but it is found that the improvement in the capacity retention rate is limited. Through non-destructive computed tomography detection and morphology characterization, the silicon composite electrode under a constant pressure leads to a change in the battery spacing, which, in turn, increases the polarization and results in lithium precipitation on the surface, and lithium precipitation consumes the electrolyte and causes the gas formation and eventually battery failure. Finally, an improved strategy that can effectively prolong the battery cycle period is proposed. In the experiment, it is observed that, when the battery operates under a constant pressure, its capacity undergoes a slow decline followed by a rapid drop at a specific capacity point known as the “inflection point”. When the inflection point occurs, increasing the external pressure applied to the battery can delay the aging behavior of the battery. This work has certain guidance for improving the cycle life of silicon-based lithium-ion batteries and pushes their engineering applications in battery pack systems.

Health and lifespan prediction considering degradation patterns of lithium-ion batteries based on transferable attention neural network

With the continuous concern on the safety of battery systems, accurate and rapid assessment of battery degradation is essential for practical applications. In this paper, a transferable attention network model based on deep learning is developed to evaluate battery degradation, which can simultaneously predict state of health (SOH) and remaining useful life (RUL) for lithium-ion batteries. First, degradation patterns of the cells are identified by K-means clustering based on the difference of the cells at their early cycles. Secondly, the attention mechanisms are designed to suppress noises in extracted feature maps and trace the interaction between long- and short-term degradation modes. Thirdly, the common knowledge represented by the reference cells and the unique degradation features of the target cell are fused by transfer learning, then SOH and RUL prediction are realized through multi-task learning. Finally, a large-scale battery dataset containing different cycle conditions is used to verify the accuracy and generalization of the developed method. The results show that the developed method achieves accurate SOH and RUL prediction with the average root mean square error within 0.14% and six cycles, respectively.

Li4Ti5O12 nanowires intertwined with carbon nanotubes for ultra-long life and conductive additive-free anodes of lithium-ion batteries

Herein, a novel Li4Ti5O12@carbon nanotubes (CNT) composite, composed of Li4Ti5O12 nanowires intertwined with CNT (4 wt%), is synthesized by a simple method and studied as conductive additive-free anodes of lithium-ion batteries (LIBs). Results indicate that the Li4Ti5O12@CNT shows impressive electrochemical performances, such as high initial coulombic efficiency (ICE, ～94%), acceptable capacity (147.6 mAh g−1) and ultra-long life (3000 cycles). The facile preparation process and the impressive electrochemical performance make Li4Ti5O12@CNT anode promising in LIB application.