Battery Aging Research Articles

Predictive State-of-Health (SOH) of lithium-ion batteries throughout their lifecycle based on LSTM neural networks

With the rapid development of new energy sources, the demand for energy storage systems in the power industry has surged. While lithium-ion battery energy storage systems have gained widespread market acceptance due to their performance advantages, concerns regarding battery safety and durability have also escalated. Therefore, accurately predicting the health status of batteries has become a crucial aspect in assessing their safety and durability. State of Health (SOH), as a vital indicator of battery health status, has become a focal point for many researchers aiming to enhance the accuracy of its prediction. Today, owing to the rapid advancements in machine learning and neural networks, data-driven approaches have gradually emerged as one of the primary methods for SOH prediction. This paper proposes a method based on Long Short-Term Memory (LSTM) neural networks, which combines theoretical analysis with neural network models, utilizing authentic battery aging data from NASA to predict the health status of lithium-ion batteries throughout their lifecycle. Experimental results demonstrate that the proposed method exhibits high accuracy.

Predictive State-of-Health (SOH) of lithium-ion batteries throughout their lifecycle based on LSTM neural networks

With the rapid development of new energy sources, the demand for energy storage systems in the power industry has surged. While lithium-ion battery energy storage systems have gained widespread market acceptance due to their performance advantages, concerns regarding battery safety and durability have also escalated. Therefore, accurately predicting the health status of batteries has become a crucial aspect in assessing their safety and durability. State of Health (SOH), as a vital indicator of battery health status, has become a focal point for many researchers aiming to enhance the accuracy of its prediction. Today, owing to the rapid advancements in machine learning and neural networks, data-driven approaches have gradually emerged as one of the primary methods for SOH prediction. This paper proposes a method based on Long Short-Term Memory (LSTM) neural networks, which combines theoretical analysis with neural network models, utilizing authentic battery aging data from NASA to predict the health status of lithium-ion batteries throughout their lifecycle. Experimental results demonstrate that the proposed method exhibits high accuracy.

A Design Tool for Battery/Supercapacitor Hybrid Energy Storage Systems Based on the Physical–Electrochemical Degradation Battery Model BaSiS

A design toolbox has been developed for hybrid energy storage systems (HESSs) that employ both batteries and supercapacitors, primarily focusing on optimizing the system sizing/cost and mitigating battery aging. The toolbox incorporates the BaSiS model, a non-empirical physical–electrochemical degradation model for lithium-ion batteries that enables accurate simulations of battery performance and degradation under realistic operating conditions. The paper presents a detailed description of the parameterization, and validation process for the battery model, emphasizing the high accuracy and strong reliability of the battery aging prediction. The HESS design toolbox can be used to investigate the impact of various battery/supercapacitor configurations and energy management algorithms on the design, battery degradation, and system investment cost of the hybrid storage system. To illustrate the effectiveness of the design toolbox, a case study on Dynamic Moderation frequency support in the UK grid was conducted. For this use case, the application of hybrid storage energy systems is well suited due to the highly dynamic power regulation requirements in island grids with low inertia. By utilizing the fast response of supercapacitors, the stress on the battery caused by short-term high-power peaks can be significantly alleviated. In this way, the hybrid storage system effectively reduces either the battery size or the battery aging rate. In summary, this research highlights the crucial role of a comprehensive analysis in the design of hybrid energy storage systems, addressing both battery aging and overall system costs. The design toolbox can provide transparency regarding the design space and assist in determining the most suitable HESS configuration for a given application.

A novel LFSSA-LSTM algorithm for the remaining useful life prediction of lithium-ion batteries in whole life cycle

The accurate prediction of the remaining useful life (RUL) of lithium-ion batteries is essential to improve the safety of battery systems and optimize battery management strategies. Therefore, a new algorithm combining Sparrow search algorithm (SSA) improved by the Levy flight method and long short-term memory neural network (LSTM) is proposed in this paper. By extracting the National Aeronautics and Space Administration (NASA) battery aging datasets, the battery correlated health characteristics are obtained, and the deterioration trend of battery performance is effectively described. The sparrow search algorithm improved by the Levy flight method enhances the ability of the algorithm to jump out of the optimal solution. By using the improved algorithm, the hyper-parameter optimization problem of the long short-term memory network is solved, and the accuracy of the algorithm is improved. Finally, the probability density function is used to predict the remaining useful life of lithium-ion batteries, which effectively prevents the uncertainty caused by a single prediction result. Compared with long short-term memory neural network and the one optimized by a traditional sparrow search algorithm, the mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE) of the proposed algorithm are kept at 1.77 %, 2.24 %, and 1.25 %, respectively. The results indicate that the algorithm has better accuracy and fitting performance.

Enhanced SOC estimation in Lithium-ion Batteries using GRU Neural Networks with GSA

State of charge (SOC) is one of the most important metrics to evaluate the performance of the battery management system (BMS) in electric vehicles (EVs). Lithium-ion batteries have been utilized often for SOC estimates in EV applications because of its attractive characteristics such as extended lifespans, high voltages, energy, and capacities. However, dynamic charging and discharging profiles, material degradation, battery aging, chemical reaction, and temperature fluctuations would diverge the accuracy of SOC estimation. Deep learning has proven to become an efficient method for SOC estimation in EVs due to its strong computational capabilities, enhanced generalization performance, and excellent accuracy under dynamic loading profiles. Therefore, this study proposes the use of the gated recurrent unit neural network (GRUNN) for estimating SOC in lithium-ion batteries. The best hyperparameters of GRUNN that enable improving SOC accuracy are found using gravitational search optimization (GSA). To evaluate the functional sustainability of the suggested method, a full battery test bench model is created and accordingly, robustness of the proposed approach is verified under diverse EV drive cycles and aging impacts. The effectiveness of the proposed approach is evaluated by comparing with several existing approaches. The results demonstrate that the suggested method archives SOC error and RMSE in EV driving cycles and ageing effects below 5% and 2%, respectively. The excellent outcomes obtained by the proposed method would certainly enhance the battery charging and discharging profile and EV performance. GUB JOURNAL OF SCIENCE AND ENGINEERING, Vol 9(1), 2022 P 1-12

Comprehensive aging model coupling chemical and mechanical degradation mechanisms for NCM/C6-Si lithium-ion batteries

The aging of lithium-ion batteries (LIBs) is synergistically influenced by multiple chemical/mechanical degradation mechanisms. Therefore, conventional models that incorporate only partial mechanisms exhibit limited predictive accuracy and applicability, failing to fully reflect the effects of chemical/mechanical degradation under complex operating conditions. Here, we propose an aging model for NCM/C6-Si LIBs coupled with comprehensive chemical/mechanical degradation mechanisms. The model includes chemical mechanisms at the C6-Si anode solid electrolyte interface (SEI), Li plating, and NCM cathode electrolyte interface (CEI), as well as mechanical mechanisms of loss of active material (LAM) for C6, Si, and NCM. Based on this model, we comprehensively investigate the effect of capacity loss by (dis)charge rates and ambient temperatures, obtaining the aging characteristics and the contribution of each mechanism to loss under different variables. Furthermore, we quantitatively analyze the sensitivity and response characteristics of the degradation sub-mechanism to (dis)charge rate and temperature. This study introduces an advanced aging analysis model for NCM/C6-Si LIBs, which can effectively decouple the operational characteristics of the degradation mechanism and provide guidance for developing next-generation high-energy LIBs.

Degradation of lithium-ion batteries under automotive-like conditions: aging tests, capacity loss and q-OCP interpretation

Battery electric vehicles are spreading worldwide as a relevant solution for the decarbonization of the transportation sector, ensuring high volume and weight-based energy density, high efficiency and low cost. Nevertheless, batteries are known to age in a rather complex and conditions-dependent way. This work aims at investigating battery aging resulting from close-to-real world conditions, highlighting single stressors role. Hence, aiming at representativeness for automotive application, an extensive literature review is performed, identifying a wide set of representative conditions together with their specific variations to be investigated. Realistic driving schedules like WLTP is identified and continuously applied in cycling on commercial samples, investigating the capacity loss from a q-OCP perspective with an equilibrium model. In general, loss of lithium inventory is detected as the main degradation parameter, likely related to SEI growth. Recharge C-rate and load profile appear as poorly-affecting degradation, while a dominant role is associated with operating temperature. Interestingly, temperature and cycling-related degradation appears to be independent and their effects can be effectively superimposed. Loss of active positive electrode material seems particularly affected by cycling depth of discharge, likely having mechanical origin as particle cracking.

Parametric Correlation Analysis between Equivalent Electric Circuit Model and Mechanistic Model Interpretation for Battery Internal Aging

In modern electric vehicle applications, understanding the evolution of the internal electrochemical reaction throughout the aging of batteries is as relevant as knowing their state of health. This article demonstrates the feasibility of correlating a mechanistic model of the battery internal electrochemical reactions with an equivalent electrical circuit (EEC) model, providing a practical and understandable interpretation of the internal reactions for electrical specialists. By way of electrochemical impedance spectroscopy analysis and automatic control theory, a methodology for correlating the resistance and capacitance variations of the EEC model and how they reflect the electrochemical reaction changes is proposed. These changes are represented through the time constants of the three RC parallel arrays from an EEC model. PS-260 lead–acid batteries were analyzed throughout the SOC and their useful life to validate this methodology. The result analysis allows us to establish that the first RC array corresponds to the negative electrode reactions in the range of 1.48 Hz to 10 kHz, the second RC array to the positive electrode reactions and generation of sulfates in the range of 0.5 to 1.48 Hz, and the third RC array to the generation of sulfates and their diffusion in the range of 0.01 to 0.5 Hz.

Lithium battery state of health estimation using real-world vehicle data and an interpretable hybrid framework

Accurate point estimation and uncertainty estimation of the state of health (SOH) of battery systems are crucial for alleviating user range anxiety and preventing battery safety incidents. On top of approximately 28 million real-world electric vehicle operation data samples, this paper embeds the categorical boosting algorithm as the base learner of the natural gradient boosting, proposing a novel interpretable N-CatBoost hybrid framework to achieve precise point estimation and uncertainty estimation of the battery SOH. Deep charging segments are selected to calculate the initial capacity, and the nonlinear decay trend of the capacity is derived through the Savitzky-Golay filter. Based on the influencing factors of capacity degradation, health features characterizing battery aging are extracted from electric vehicle data as model inputs. The model's hyperparameters are optimized using the particle swarm optimization algorithm, and it is compared with seven other popular machine learning algorithms. The results indicate that the proposed N-CatBoost model achieves the highest estimation accuracy, with mean absolute percentage error and root mean square error of 0.817 % and 1.156 Ah, respectively. In addition, the Shapley additive explanation method is utilized to make the model interpretable, providing full transparency of all predicted values. More importantly, the developed N-CatBoost model can achieve uncertainty estimation of its predictions, with 100 % of the actual capacity values successfully falling within the model's 99 % prediction interval. Therefore, the proposed N-CatBoost model is a reliable and effective method with great potential for deployment on the cloud side for the SOH estimation of batteries in large-scale vehicles.

Digital Twin‐Assisted Degradation Diagnosis and Quantification of NMC Battery Aging Effects During Fast Charging

AbstractA comprehensive understanding of aging mechanisms and degradation diagnosis under fast charging in battery electric vehicles is needed. However, there is a lack of tools to capture both macroscopic and microscopic parameters, still focusing on experiment results analysis. To get a comprehensive degradation insight for improved service life, this study explores aging mechanisms in LiNi0.5Co0.2Mn0.3O2 graphite batteries under different fast charging conditions (0.6C to 2C) for up to 1000 equivalent full cycles (EFCs). An improved digital twin is used to quantify aging effects and identify aging modes, combining electrochemical techniques with post‐mortem analysis for assessing chemical and structural degradation. After 1000 EFCs, the dominant aging modes are loss of active material (LAM) in the negative electrode and loss of lithium inventory (LLI), surpassing LAM in the positive electrode. Compared to constant current charging, multistep fast charging effectively mitigates Li plating effects and graphite cracking, resulting in a 13% reduction in LAM. Additionally, optimizing the depth of discharge leads to at least 4% reductions in LLI and 16% in LAMpos. This study proves the electrochemical digital twin's potential for quantifying aging effects and serves as a basis for future physics‐informed machine learning to predict aging behavior and modes.

Development of self-adaptive digital twin for battery monitoring and management system

The application of digital twin (DT) on battery energy storage systems (BESS) has attracted increasing attention in the last decade. However, existing studies usually focus on building pre-calibrated DT for state estimation and prediction. These DTs lack the ability for dynamic adaptation to changes in battery aging and evolving operating environment, which thus limits their effectiveness in intelligent decision-making for system performance enhancement. Therefore, this work develops a self-adaptive DT for battery monitoring and management system (DT-BMMS). The proposed self-adaptive algorithm ensures accurate long-term mapping between the physical entity and the digital model. Additionally, a model predictive control-based state-of-charge (SOC) balancing method is deployed. Simulation results demonstrate the capability of the developed DT-BMMS to adaptively adjust the DT as the system evolves, which allows the maintenance of SOC balancing under different scenarios.

Sizing up sustainability: Influence of battery size and cell chemistry on battery-electric trucks’ life-cycle carbon emissions

Customers and legislators will increasingly demand monitoring and reducing the carbon footprint of logistic companies in the coming years. Besides the fundamental decision for a battery-electric truck, the choice of battery size and cell chemistry should be approached with the goal of minimizing the carbon footprint of the supply chain. Therefore, this paper analyzes the impact of battery size and cell chemistry on the carbon footprint and which factors influence this the most. To quantify this footprint, emissions from production, potential replacements based on battery aging, energy generation, and reuse in second-life applications for both LFP and NMC cell chemistries are considered. The results show that use phase emissions from energy generation have the largest leverage on life-cycle emissions, whereas production emissions play a minor role. If electricity with lower emissions at the home depot compared to the public electricity is available, oversizing the battery capacity tends to be the emission optimal choice. If more than 8% additional renewable energy at the home base for NMC and more than 38% for LFP is available, the largest possible battery is emission-optimal. For smaller shares, the smallest possible battery should be selected. Significant dependence on purchase year and share of second-life usage can be observed.

Robust Online Estimation of State of Health for Lithium-Ion Batteries Based on Capacities under Dynamical Operation Conditions

Lithium-ion batteries, as the main energy storage component of electric vehicles (EVs), play a crucial role in ensuring the safe and reliable operation of the battery systems through monitoring their state of health (SOH). However, temperature variations and battery aging have significant impacts on the internal parameters of lithium-ion batteries, and these changes directly correlate with the accuracy of battery SOH estimation. To address these issues, this paper proposes an estimation method for lithium-ion battery SOH that considers the impact of temperature. The method begins with reconstructing a second-order hybrid equivalent circuit model for lithium-ion batteries, through which an adaptive update rate for battery model parameters is designed. On this basis, a nonlinear observer for battery states is introduced by integrating filters to estimate SOH. The proposed method considers the impact of capacity in the design of the parameter adaptive update rate, enabling the capacity to be dynamically adjusted based on the actual state of the batteries. This reduces the cumulative error in the SOC observer and improves the modeling accuracy of battery models. Experimental results demonstrate that the method proposed in this paper exhibits exceptional performance in SOH estimation under different temperature conditions. The mean absolute error for SOH estimation does not exceed 0.5%, and the root mean square error does not exceed 0.2%. This method can significantly improve the estimation accuracy of SOH, providing a more efficient and accurate solution for battery management systems in EVs.

The sensitive detection of the early-stage internal short circuit triggered by lithium plating through the simplified electrochemical model at various working conditions

The internal short circuit (ISC) is one of the main causes of thermal runaway in batteries. Facing the current fast charging scenario of batteries, this paper aims to explore the sensitivity of solid-phase diffusion coefficient to ISC during high current charging. The voltage and current data of the real ISC is input into the simplified pseudo-two-dimensions model to identify the intrinsic parameters. The results indicate that the positive solid-phase diffusion coefficient (Ds,p) has higher sensitivity and stability than the negative solid-phase diffusion coefficient (Ds,n). For real ISC, the sensitivity of the solid-state diffusion coefficient increases as the charging current increases. This is consistent with the trends in ISC that we have derived from our low-temperature ISC extension experiment. For #1 ISC battery at 25 °C (SOC = 0.3–0.85), Ds,p changes by 56 % (3C) and 96 % (5C), and Ds,n changes by 6.87 × 10−14m2/s (166 %) (3C), 8.38 × 10−14m2/s (670 %) (5C). Moreover, the sensitivity of the solid-state diffusion coefficient to ISC decreases after battery aging, but it still has sufficient ability to diagnose ISC. High temperature will reduce the sensitivity of the solid phase diffusion coefficient.

Enhanced energy extraction from wind driven PMSG using digital twin model of battery charging system

The pressing need to meet sustainability and climate objectives has prompted a call for expanding the use of renewable energy sources, improving the intelligence and flexibility of electricity grids, and increasing the adoption of electric vehicles and other electricity-powered products. In light of their numerous benefits, widespread electrification and digital technologies can form the basis of energy and climate policy. This research presents an innovative biased transformer and digital twin battery model to support the wind energy based electric vehicle battery charging system. During periods of low and moderate wind, the biased transformer is utilized to increase the voltage for charging electric vehicles. A digital twin of the battery system creates a window into battery charging and aging levels by evaluating the data gathered based on internal resistance. Furthermore, the State of Charge estimation model and different internal resistance estimation algorithms are also exploited. A novel method for precise State of Charge (SOC) estimation in third-gen battery models is presented, integrating Collaborative Gradient Boosting and Adaptive Extended Kalman Filter (AEKF). This approach, implemented in MATLAB Simulink, combines machine learning and adaptive filtering, enhancing accuracy and adaptability for efficient battery management. The functionalities and resilience of both hardware and software of the proposed system are validated under field operation.

A High-Speed Multichannel Electrochemical Impedance Spectroscopy System Using Broadband Multi-Sine Binary Perturbation for Retired Li-Ion Batteries of Electric Vehicles

Retired electric vehicle (EV) batteries are reused in second-life energy storage applications. However, the overall performance of repurposed energy storage systems (ESSs) is limited by the variability in the individual batteries used. Therefore, battery grading is required for the optimal performance of ESSs. Electrochemical impedance spectroscopy (EIS)-based evaluation of battery aging is a promising way to grade lithium-ion batteries. However, it is not practical to measure the impedance of mass-retired batteries due to their high complexity and slowness. In this paper, a broadband multi-sine binary signal (MSBS) perturbation integrated with a multichannel EIS system is presented to measure the impedance spectra for the high-speed aging evaluation of lithium-ion batteries or modules. The measurement speed is multiple times higher than that of the conventional EIS. The broadband MSBS is validated with a reference sinusoidal sweep perturbation, and the corresponding root-mean-square error (RMSE) analysis is performed. Moreover, the accuracy of the presented multichannel EIS system is validated by impedance spectra measurements of Samsung INR18650-29E batteries and compared with those measured using a commercial EIS instrument. A chi-squared error under 0.641% is obtained for all channels. The non-linearity of batteries has a significant impact on the quality of impedance spectra. Therefore, Kronig–Kramer (KK) transform validation is also performed.

Comprehensive study of the aging knee and second-life potential of the Nissan Leaf e+ batteries

Aging knee may induce rapid degradation of lithium-ion batteries (LIBs) and pose safety risks. It is crucial to study the battery aging knee and find solutions to address it. In this light, this study focuses on the comprehensive aging characterizations of the Nissan Leaf third generation (Gen3) LIBs to elucidate their aging trajectories. Twenty-four battery cells were divided into six groups and cycled at different working conditions, i.e., voltage, current, and temperatures. The test result shows that the batteries tend to reach the aging knee around 75 % state of health (SOH). Then, we improved the working conditions after the knee point to mimic second-life operation. Key findings show that the aging knee can be stopped by improving the working conditions. The battery capacity degradation slowed down to about 4 % per 1000 cycles, projecting a second-life potential of 4000–6000 cycles between 75 % and 50 % SOH. Post-mortem analysis indicated that negative electrode failure is the cause of aging knee and gas generation. The impact of working conditions on the battery aging is analyzed. Based on the analysis, we propose strategies for both the first and second life operations of LIBs to improve battery life and avoid the aging knee.

Online battery scheduling for enhanced profitability and longevity in pay-for-performance frequency regulation markets

Batteries participating in frequency regulation (FR) markets earn revenue at the expense of longevity. The declining reward prices and highly stochastic automatic generation control (AGC) signals with short update intervals necessitate research into aging-aware real-time FR strategies for the battery. To address these challenges, this paper develops an online battery dispatch strategy based on Lyapunov optimization, accounting for Ah-throughput, C-rate, and temperature aging effects to enhance the service profitability of the battery. The strategy operates independently of probabilistic data from random processes, with its optimality gap inversely correlated with the control parameter. A systematic and intuitive approach is established for constructing virtual queues and determining the maximum feasible parameter. The proposed method is validated through extensive numerical simulations using actual lithium-ion battery aging data, along with year-long regulation signal and market price data from Pennsylvania-New Jersey-Maryland (PJM). Results show that the strategy can extend the battery lifespan by 83.41% and increase the net present value (NPV) by 172.67% compared to the conventional Ah-based strategy.

Depletion of Electrolyte Salt Upon Calendaric Aging of Lithium-Ion Batteries and its Effect on Cell Performance

The trend for increased nickel content in layered transition metal oxide cathode active materials and increasing charging cut-off voltages aggravates aging of lithium-ion battery cells at high state of charge (SOC). We investigate the calendaric aging behavior of large-format automotive prototype cells and laboratory single-layer pouch cells at high but realistic cell voltages/SOCs and demonstrate that electrolyte oxidation in combination with follow-up reactions can cause a significant loss of the LiPF6 salt in the electrolyte. For this, we analyze the LiPF6 concentration in aged cells, the generation of H2 upon storage, and the cell resistance for different aging conditions. We show that the LiPF6 loss is a critical aging phenomenon, as it cannot readily be detected by capacity fading measurements at low/medium C-rates or by cell resistance measurements, while it severely reduces rate and fast-charging capability. Under certain circumstances, LiPF6 loss can even lead to a temporary capacity increase due to conversion of the conducting salt in the electrolyte to cyclable lithium in the active material. Finally, we suggest a possible reaction mechanism and a simple accounting model to keep track of how different side reactions involved in LiPF6 loss change the cyclable lithium inventory of a lithium-ion cell.

The aging pattern of lithium-ion batteries based on experiments at different temperatures

Abstract In this paper, we take an 18650-type LiFePO4 battery as the research object to investigate the aging mechanism of Li-ion batteries at different temperatures. The aging tests are conducted at 25°C, 35°C and 45°C. The results show that elevating the operating temperature can improve the cycle life of Li-ion batteries. The aging mechanism was analyzed by incremental capacity and electrochemical impedance spectrum analysis. The following conclusions were obtained: the aging mechanism of lithium-ion batteries mainly includes loss of lithium inventory, loss of active material, and conductivity loss, among which loss of lithium inventory and loss of active material are the main causes of Li-ion battery aging, and the contribution of conductivity loss is the smallest. Further analysis of the causes of battery aging by disassembling and testing the aging electrode sheet SEM and TEM shows that the aging of Li-ion battery is manifested by the rupture of graphite particles, the generation of SEI layer and lithium dendrite, and increasing the temperature can ease the rupture of graphite particles and the generation of lithium dendrite, thus improving the cycle life of the battery.