Lithium-ion Battery Degradation Research Articles

A Single Particle model with electrolyte and side reactions for degradation of lithium-ion batteries

Battery degradation, which is the reduction of performance over time, is one of the main roadblocks to the wide deployment of lithium-ion batteries. Physics-based models, such as those based on the Doyle-Fuller-Newman model, are invaluable tools to understand and predict such phenomena. However, these models are often too complex for practical applications, so reduced models are needed. In this article we introduce the Single Particle Model with electrolyte and Side Reactions, a reduced model with electrochemical degradation which has been formally derived from the Doyle-Fuller-Newman model with Side Reactions using asymptotic methods. The reduced model has been validated against the full model for three scenarios (solid-electrolyte interphase growth, lithium plating, and both effects combined) showing similar accuracy at a much lower computational cost. The implications of the results are twofold: the reduced model is simple and accurate enough to be used in most real practical applications, and the reduction framework used is robust so it can be extended to account for further degradation effects.

Electrochemical Failure Results Inevitable Capacity Degradation in Li-Ion Batteries—A Review

Lithium-ion batteries (LIBs) have been widely used in mobile devices, energy storage power stations, medical equipment, and other fields, became an indispensable technological product in modern society. However, the capacity degradation of LIBs limits their long-term deployment, which is not conducive to saving resources. What is more, it will lead to safety problems when the capacity of the battery is degraded. Failure of the battery is a key issue in the research and application of LIBs. Faced with the problem of capacity degradation, various aspects of LIBs have been studied. This paper reviews the electrochemical degradation mechanism of LIBs’ life fade, detection technologies for battery failure, methods to regulate battery capacity degradation, and battery lifetime prognostics. Finally, the development trend and potential challenges of battery capacity degradation research are prospected. All the key insights from this review are expected to advance the research on capacity fading and lifetime prediction techniques for LIBs.

Thermal Investigation of Cylindrical Lithium-ion Batteries for Different Loading using Experimental and Numerical Techniques

With growing concerns over climate change due to automotive emissions and fossil fuel depletion, electric vehicles (EVs) have gained more interest as a mode of transportation. Mostly, lithium-ion batteries are being used as an energy storage system in EVs. Temperature plays a major role in lithium-ion battery performance, degradation, and safety. Thermal investigation of cylindrical lithium-ion batteries of different chemistry and shape factors (18650 NMC and 21700 NCA) is conducted for different charging/discharging rates (0.5 C, 1 C, 1.5 C) and surrounding temperatures (26 °C and 45 °C) using numerical and experimental techniques. The numerical technique involves the determination of the surface temperature of the battery by solving the energy balance equation using a suitable numerical method. The numerical technique proposed in this study is validated using the experimental technique and can predict the temperature with at least 90% accuracy during the entire charging and discharging cycle from a lower to higher loading rate and different surrounding temperatures. The impact of different battery internal and external parameters on battery temperature is carried out using validated numerical technique. Thermal investigation conducted in this research can help in deciding the operating strategies of the battery management system and further in designing the proper thermal management system.

Prediction of Battery Cycle Life Using Early-Cycle Data, Machine Learning and Data Management

The prediction of the degradation of lithium-ion batteries is essential for various applications and optimized recycling schemes. In order to address this issue, this study aims to predict the cycle lives of lithium-ion batteries using only data from early cycles. To reach such an objective, experimental raw data for 121 commercial lithium iron phosphate/graphite cells are gathered from the literature. The data are analyzed, and suitable input features are generated for the use of different machine learning algorithms. A final accuracy of 99.81% for the cycle life is obtained with an extremely randomized trees model. This work shows that data-driven models are able to successfully predict the lifetimes of batteries using only early-cycle data. That aside, a considerable reduction in errors is seen by incorporating data management and physical and chemical understanding into the analysis.

Prediction of Remaining Useful Life of Lithium Batteries Based on WOA-VMD and LSTM

The remaining useful life (RUL) of a lithium-ion battery is directly related to the safety and reliability of the electric system powered by a lithium-ion battery. Accurate prediction of RUL can ensure timely replacement and maintenance of the batteries of the power supply system, and avoid potential safety hazards in the lithium-ion battery power supply system. In order to solve the problem that the prediction accuracy of the RUL of lithium-ion batteries is reduced due to the local capacity recovery phenomenon in the process of the capacity degradation of lithium-ion batteries, a prediction model based on the combination of the whale optimization algorithm (WOA)-variational mode decomposition (VMD) and short-term memory neural network (LSTM) was proposed. First, WOA was used to optimize the VMD parameters, so that the WOA-VMD could fully decompose the capacity signal of the lithium-ion battery and separate the dual component with global attenuation trend and a series of fluctuating components representing the capacity recovery from the capacity signal of the lithium-ion battery. Then, LSTML was used to predict the dual component and fluctuation components, so that LSTM could avoid the interference of the capacity recovery to the prediction. Finally, the RUL prediction results were obtained by stacking and reconstructing the component prediction results. The experimental results show that WOA-VMD-LSTM can effectively improve the prediction accuracy of the RUL of lithium-ion batteries. The average cycle error was one cycle, the average RMSE was less than 0.69%, and the average MAPE was less than 0.43%.

The Dilemma of C-Rate and Cycle Life for Lithium-Ion Batteries under Low Temperature Fast Charging

Electric vehicles (EVs) in severe cold regions face the real demand for fast charging under low temperatures, but low-temperature environments with high C-rate fast charging can lead to severe lithium plating of the anode material, resulting in rapid degradation of the lithium-ion battery (LIB). In this paper, by constructing an electrode–thermal model coupling solid electrolyte interphase (SEI) growth and lithium plating, the competition among different factors of capacity degradation under various ambient temperatures and C-rates are systematically analyzed. In addition, the most important cause of rapid degradation of LIBs under low temperatures are investigated, which reveal the change pattern of lithium plating with temperature and C-rate. The threshold value and kinetic law of lithium plating are determined, and a method of lithium-free control under high C-rate is proposed. Finally, by studying the average aging rate of LIBs, the reasons for the abnormal attenuation of cycle life at lower C-rates are ascertained. Through the chromaticity diagram of the expected life of LIBs under various conditions, the optimal fast strategy is explored, and its practical application in EVs is also discussed. This study can provide a useful reference for the development of high-performance and high-safety battery management systems to achieve fine management.

Deformation measurement within lithium-ion battery using sparse-view computed tomography and digital image correlation

Electrode deformation can cause high local strain and serious capacity degradation in lithium-ion batteries (LIBs) during cycling. Risk reduction in many applications requires an understanding of the effects of the charging/discharging rate on the electrode structure during the battery life cycle. Cyclic charging/discharging experiments of wound 18 650 cylindrical LIBs were conducted at different charging/discharging rates (1C and 2C) to determine the effect of rate on electrode deformation. The charging/discharging capacity as well as battery voltage and time were analyzed during cycling. To acquire electrode deformation images and meet the requirements of computed tomography (CT) within 2 min during the charging/discharging process, sparse-view CT was performed at fixed cycle intervals. Subsequently, a sparse-view reconstruction network was proposed to generate a slice image. Finally, the electrode displacement and strain fields were calculated using the augmented Lagrangian digital image correlation algorithm. The causes of electrode deformation were analyzed and discussed from the perspective of molecular and macroscopic structure. Experimental results show that the structural similarity, peak signal-to-noise ratio and root mean square error in the reconstructed image of the axial section within the battery obtained via the proposed network were 0.9616, 38.7411 dB and 0.0108, respectively, which were better than the other methods of comparison. After 100 cycles, the capacity decay of the battery at 2C was 9.23-fold higher than that at 1C. After 100 cycles at 2C, the maximum displacement of the electrode reached 0.46 mm along the x-direction. The electrode structural deformation of the battery can be intuitively understood at different rates, which facilitates reasonable utilization and structural optimization of the battery.

Lithium-Ion Battery Degradation: Measuring Rapid Loss of Active Silicon in Silicon-Graphite Composite Electrodes.

To increase the specific energy of commercial lithium-ion batteries, silicon is often blended into the graphite negative electrode. However, due to large volumetric expansion of silicon upon lithiation, these silicon-graphite (Si-Gr) composites are prone to faster rates of degradation than conventional graphite electrodes. Understanding the effect of this difference is key to controlling degradation and improving cell lifetimes. Here, the effects of state-of-charge and temperature on the aging of a commercial cylindrical cell with a Si-Gr electrode (LG M50T) are investigated. The use of degradation mode analysis enables quantification of separate rates of degradation for silicon and graphite and requires only simple in situ electrochemical data, removing the need for destructive cell teardown analyses. Loss of active silicon is shown to be worse than graphite under all operating conditions, especially at low state-of-charge and high temperature. Cycling the cell over 0-30% state-of-charge at 40 °C resulted in an 80% loss in silicon capacity after 4 kA h of charge throughput (∼400 equiv full cycles) compared to just a 10% loss in graphite capacity. The results indicate that the additional capacity conferred by silicon comes at the expense of reduced lifetime. Conversely, reducing the utilization of silicon by limiting the depth-of-discharge of cells containing Si-Gr will extend their lifetime. The degradation mode analysis methods described here provide valuable insight into the causes of cell aging by separately quantifying capacity loss for the two active materials in the composite electrode. These methods provide a suitable framework for any experimental investigations involving composite electrodes.

Scenario analysis, management, and optimization of a new Vehicle-to-Micro-Grid (V2μG) network based on off-grid renewable building energy systems

To fully exploit the potential of decarburization in the transport sector (e.g., electric vehicles (EV)) and energy sector (e.g., building energy system), this paper proposes a new concept of ‘Vehicle-to-Micro-Grid (V2μG) network’ that incorporates the off-grid building energy system with flexible power storage/supply provided by battery EVs (BEVs) and fuel cell EVs (FCEVs). The work is conducted with three main contributions: 1) a rule-based energy management strategy is proposed to study the impact of the V2μG interactions on the EV battery degradation; 2) a scenario analysis based on four working modes is conducted to evaluate the energy efficiency, costing, environmental impacts, and component ageing of the proposed V2μG network; 3) the optimum settings for system configuration, capacity, and operation strategy of the V2μG network are obtained by the NSGA-II algorithm. The study suggested that the degradation of lithium-ion batteries in BEV can be reduced by 13% compared to the network without FCEVs. In a community with 160 households and 200 EVs, the optimal V2μG network can reduce carbon dioxide emissions by 515.56 tons annually compared to the conventional off-grid building energy system powered by internal combustion engines.

Simultaneous neutron and X-ray tomography for visualization of graphite electrode degradation in fast-charged lithium-ion batteries

Advanced battery characterization using neutron and X-ray-based imaging modalities is crucial to reveal fundamental degradation modes of lithium-ion batteries (LIBs). Taking advantage of the sensitivity of neutrons to some low-Z (Li) and X-rays to high-Z materials (Cu), here we demonstrate the viability of simultaneous neutron- and X-ray-based tomography (NeXT) as a non-destructive imaging platform for ex situ 3D visualization of graphite electrode degradation following extreme fast charging (XFC). In addition, we underscore the benefits of the simultaneous nature of NeXT by combining the neutron and X-ray data from the same sample location for material identification and segmentation of one pristine and two XFC-cycled graphite electrodes (9C charge for 450 cycles). Our ex situ results and methodology development pave the way for the design of NeXT-friendly LIB geometries that will allow operando and/or in situ three-dimensional (3D) visualization of electrode degradation during XFC. • Simultaneous neutron and X-ray tomography (NeXT) is a non-invasive imaging method • NeXT enables visualization of graphite electrode degradation after fast charging • NeXT overlaps neutron and X-ray data from same sample area to identify materials • NeXT shows qualitative correlation between electrode degradation and plated Li Advanced battery characterization is critical to reveal fundamental degradation modes of lithium-ion batteries. Using the sensitivity of neutrons to lithium and X-rays to copper, Yusuf et al. demonstrate the viability of simultaneous neutron and X-ray tomography as a non-invasive imaging modality for visualization of graphite electrode degradation following fast charging.

Online Aging Diagnostics Using Optimally Designed Experiments

Degradation of lithium-ion batteries is the result of many complex phenomena occurring simultaneously at varying time and length scales. The underlying electrochemical and mechanical phenomena have received much attention from researchers [1]. Physics-based models of these effects support the mechanistic understanding of degradation modes and can thereby help reduce their severity. Few studies target changing electrochemical parameters such as diffusion coefficients or reaction rate constants that have a direct impact on model accuracy and manifest themselves in observable aging. Lyu et al. [2] used a simplified electrochemical model and monitored battery degradation by following changes in diffusion time constants, electrode balancing, reaction rate coefficients, and ohmic resistance. However, several of the parameters they attempted to track could only be identified with low accuracy as they used the same data-set to identify all parameters. In this study, we investigate parameters of a full order Newman-type model [3] over the course of a batteries lifetime under real-world load-cycles. To ensure parameter identifiability, optimally designed experiments are used for parameter estimation.In a previous study [4] the feasibility of optimal experiment design for parametrization of electrochemical battery models was demonstrated. We now extend this work and re-evaluate key parameters over the course of an aging study on commercial, nickel-rich 18650 lithium-ion batteries. We highlight how quantifying changes in physical battery parameters can extend standard performance metrics for a batteries state-of-health by, e.g., including degradation in rate-capability. Additionally, the importance of battery usage conditions such as C-rate or state-of-charge window on model parameter trajectories is investigated and their relationship with conventional performance metrics such as the bulk cell resistance or rate-capability determined.Quantifying how specific mechanisms contribute to apparent capacity or power fade is a major step towards battery lifetime optimization. This could enable designs more tailored for specific applications and significantly extend batteries useful lifetime. Furthermore, updating parameters is essential for electrochemical control strategies relying on accurate model predictions of battery states as illustrated in Figure 1. This re-calibration would make a battery management system aging-sensitive and enable more efficient utilization and a physics-informed state-of-health.Figure 1: The central plot shows how parameters change during aging. If this change is not considered, model performance deteriorates between beginning-of-life (BOL) and end-of-life (EOL) (in blue, right-hand side). This is normally handled by using conservative battery management systems and over-sizing systems. The proposed strategy (orange) achieves higher model accuracy during the entire useful life and the parameter estimates can be used to formulate an extended state-of-health.

(Invited) Beyond Single Cell Characterization: Impacts of Module Configuration on Lithium-Ion Battery Performance and Degradation

Batteries are commonly connected in series and parallel to form modules that meet the power and energy requirements of specific applications. However, conclusions about battery performance and degradation under different conditions are often based on single cell cycling results. In this study, we evaluate the performance of lithium-ion eight-cell modules in six different series-parallel configurations over the course of hundreds of cycles. Each cell was individually instrumented for voltage, current, and temperature monitoring. Current data were collected using closed-loop Hall effect sensors to provide accurate measurements of parallel branch currents without the influence of in-circuit shunt sensing elements. This cell-level data collection enabled calculation of the underutilization of individual cells and assessment of differences in energy lost to cell balancing for different configurations. We also evaluated current redistribution in parallel cells, which in some cases yielded transients with peak magnitudes more than twice the average expected cell current value. The capacity and energy fade of the full modules was compared to that of individual cells cycled under the same conditions. This module cycling study, one of the broadest reported to date, allows us to systematically evaluate the performance trade-offs and safety implications of different module configurations.Sandia National Laboratories is a multi-mission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525. SAND2022-5209 A

Linking Thermal Runaway and Cycling Degradation of Lithium-Ion Batteries Using Multi-Scale X-Ray Imaging

To meet the increasing energy demands of portable devices and electric vehicles, high-nickel lithium-ion cathode materials with the general formula Li(NixMnyCoz)O2 (NMCXYZ) have been extensively researched. Currently, NMC811 is used commercially due to its high capacity and low cobalt content. However, capacity fade is still a prominent issue, with a variety of degradation mechanisms responsible. Of particular interest is secondary particle cracking, where the stress-strain effect of c-parameter expansion in the NMC crystal lattice leads to fractures. Boundaries between the primary particle grain that make up the secondary particle structure act as nucleation points for particle cracking. The exposed surface then reacts with the liquid electrolyte, leading to oxygen evolution and capacity loss. Single-crystal materials have shown promise as a method of improving cycle life, as the lack of agglomerated primary grains appear to make the monocrystalline particles more resilient to the mechanical strain of expansion during charging1.In parallel with the cycle life concerns, attention over battery safety continues to increase as instances of catastrophic battery failure are publicised across the world. While the specific steps that lead to battery thermal runaway have been extensively researched, the safety and thermal stability of degraded cells remains a scarcely investigated topic2. To that end, the goal of this work is to characterise the safety properties of different NMC morphologies, and study how charge-discharge cycling affects the thermal stability of these materials.In this work, we have combined accelerating rate calorimetry (ARC) with lab-based, post-mortem macro, micro and nano X-ray CT to observe the influence of cell structure on thermal runaway of NMC811 pouch cells with single-crystal and polycrystalline cathode materials in both the pristine and aged state. The results presented show that when pristine, the larger surface area of single-crystal materials provides more reaction sites and leads to a lower self-heating onset and thermal runaway initiation temperature compared to polycrystalline, suggesting a lower thermal stability. However, as polycrystalline particles break up during thermal runaway, the fragments create fresh surface to react with the electrolyte and continue the series of exothermic reactions, leading to a higher peak temperature. Macro-CT has revealed how the wound jellyroll structure of the cell focuses the force of the explosion through the flat sides of the cells, possibly increasing the likelihood of failure propagation. It has also been observed that the polycrystalline materials show a greater material ejection out from the centre of the cell, reinforcing the findings from the ARC investigations that although polycrystalline cells are more thermally stable initially, the peak temperatures produced during catastrophic failure are higher due to particle fracture. Micro and nano-CT are used to probe this idea further, revealing that polycrystalline particles show significant damage and fracturing during failure, whereas single-crystal materials remain largely intact.Both polycrystalline and single-crystal morphologies were cycled to 80% capacity retention, with EIS and diagnostic cycles used to identify prevalent degradation modes for both particle structures. As previously found, the single-crystal cells showed superior cycle life to the polycrystalline materials3. The same combination of ARC and X-ray CT was then applied to aged cells, where it was found that aged cells show lower thermal stability than their pristine counterparts.Overall, this work aims to build understanding of the interplay between material structure, safety and degradation of nickel-rich cathodes, with the goal of informing future material development to produce batteries that are safer throughout their lifetimes. G. Qian et al., Energy Storage Mater., 27, 140–149 (2020) https://doi.org/10.1016/j.ensm.2020.01.027.X. Feng et al., Energy Storage Mater., 10, 246–267 (2018) https://doi.org/10.1016/j.ensm.2017.05.013.D. Ren et al., eTransportation, 2, 100034 (2019) https://doi.org/10.1016/j.etran.2019.100034.

Optimal configuration and operation for user-side energy storage considering lithium-ion battery degradation

Battery energy storage systems (BESSs) have been widely employed on the user-side such as buildings, residential communities, and industrial sites due to their scalability, quick response, and design flexibility. However, cell degradation is caused by the charging and discharging of batteries, which reduces the economy of BESSs. For the optimal configuration and operation of BESSs, the battery degradation process should be integrated into the optimal problem by considering the equivalent full cycles over the optimization time horizon. This paper proposes a two-layer optimization frame to estimate and improve the net profit of BESSs in the whole life cycle, the outer layer optimizes the rated capacity and power of BESSs, and the inner layer optimizes the daily operation strategy of BESSs. To this end, the semi-empirical degradation model of lithium-ion batteries and economic models of BESSs are embedded into the optimization frame. Particle swarm optimization (PSO) algorithm and fmincon toolbox of MATLAB are adopted to solve the two-layer frame to maximize the net profit of BESSs. Simulation results of the BESS for a typical industrial user in China demonstrate that the proposed frame can effectively improve the net profit of BESSs.

Insights into LiMXO4F (M-X = Al-P and Mg-S) as Cathode Coatings for High-Performance Lithium-Ion Batteries.

Cathode coatings have received extensive attention due to their ability to delay electrochemical performance degradation in lithium-ion batteries. However, the development of cathode coatings possessing high ionic conductivity and good interfacial stability with cathode materials has proven to be a challenge. Here, we performed first-principles computational studies on the phase stability, thermodynamic stability, and ionic transport properties of LiMXO4F (M-X = Al-P and Mg-S) used as cathode coatings. We find that the candidate coatings are thermodynamically metastable and can be synthesized experimentally. The coating materials possess high oxidative stability, with the materials predicted to decompose above 4.2 V, suggesting that they have good electrochemical stability under a high-voltage cathode. In addition, the candidate coatings exhibit significant chemical stability when in contact with oxide cathodes. Finally, we have studied the Li-ion transport paths and migration barriers of LiMXO4F (M-X = Al-P and Mg-S) and calculated the low migration barriers to be 0.19 and 0.09 eV, respectively. Our findings indicate that LiMXO4F (M-X = Al-P and Mg-S) are promising cathode coatings, among which LiAlPO4F has been experimentally confirmed. The theoretical cathode coating computational methods presented here can be extended to the solid-state battery system.

Study on Integration of Retired Lithium-Ion Battery With Photovoltaic for Net-Zero Electricity Residential Homes

Abstract The behavior of a retired lithium-ion battery (LIB) from its first-life in an electric aircraft (EA) to its second-life in a solar photovoltaic (PV) system for a net-zero electricity residential home is studied. The first part of this study presents the design and sizing of a battery energy storage system (BESS), made from retired LIBs, to store a portion of the PV generation for a typical home in Ohio. The home is connected to the grid, but the net electricity usage from the grid in one year is zero. The purpose of the BESS is to peak shaving, power arbitrage, reduce the home dependency on the grid, and increase the economic benefits. The sizing is determined based on the hourly data of the PV system generation, ambient temperature, irradiation, and home demand electricity. In the second part of this study, the retired LIB degradation rate and its remaining useful life in the BESS are estimated using an adopted empirical LIB model. The model includes the capacity-fade for both first-life and second-life of the LIB under various duty cycles. It is shown that the retired LIB from its first-life is still suitable to be used in the PV grid-tied battery (PVGB) system for another 10 years. The results of this study can potentially reduce the LIB cost for electric vehicles (EVs) and EAs because the retired LIBs from these applications still have value to serve for other applications such as PVGB systems for residential homes.

Onset Potential for Electrolyte Oxidation and Ni-Rich Cathode Degradation in Lithium-Ion Batteries.

High-capacity Ni-rich layered metal oxide cathodes are highly desirable to increase the energy density of lithium-ion batteries. However, these materials suffer from poor cycling performance, which is exacerbated by increased cell voltage. We demonstrate here the detrimental effect of ethylene carbonate (EC), a core component in conventional electrolytes, when NMC811 (LiNi0.8Mn0.1Co0.1O2) is charged above 4.4 V vs Li/Li+—the onset potential for lattice oxygen release. Oxygen loss is enhanced by EC-containing electrolytes—compared to EC-free—and correlates with more electrolyte oxidation/breakdown and cathode surface degradation, which increase concurrently above 4.4 V. In contrast, NMC111 (LiNi0.33Mn0.33Co0.33O2), which does not release oxygen up to 4.6 V, shows a similar extent of degradation irrespective of the electrolyte. This work highlights the incompatibility between conventional EC-based electrolytes and Ni-rich cathodes (more generally, cathodes that release lattice oxygen such as Li-/Mn-rich and disordered rocksalt cathodes) and motivates further work on wider classes of electrolytes and additives.

Lithium-Ion Battery Degradation and Capacity Prediction Model Considering Causal Feature

Accurate life prediction of lithium-ion batteries is essential for the safety and reliability of smart electronic devices, and data-driven methods are one of the mainstream methods nowadays. However, existing prediction methods suffer from the problems such as lack of practical meaning of features and insufficient interpretability. To address this problem, this article proposes a battery degradation and capacity prediction model based on the Granger causality (GC) test and the long short-term memory network. First, initial health indicators are set from the monitoring data of the battery. Second, the vector autoregressive model and the GC test are used to select causal features that are associated with capacity degradation. Then, the impulse response analysis approach is proposed for the first time to analyze the exact influence of the features on capacity degradation and combine the battery aging mechanism, further clarifying the interpretability of the selected features. Finally, using the causal features as model input, a prediction model based on the long short-term memory network is constructed. The experimental results of the two datasets show that the minimum root mean square error is 0.0093 Ah and 0.9635 mAh with the mean relative errors of 0.25% and 0.13%, which verifies the validity and accuracy of the proposed method.

Degradation Modeling for Lithium-Ion Batteries with an Exponential Jump-Diffusion Model

The degradation of Lithium-ion batteries is usually measured by capacity loss. When batteries deteriorate with usage, the capacities would generally have a declining trend. However, occasionally, considerable capacity regeneration may occur during the degradation process. To better capture the coexistence of capacity loss and regeneration, this paper considers a jump-diffusion model with jumps subject to the exponential distribution. For estimation of model parameters, a jump detection test is first adopted to identify jump arrival times and separate observation data into two series, jump series and diffusion series; then, with the help of probabilistic programming, the Markov chain Monte Carlo sampling algorithm is used to estimate the parameters for the jump and diffusion parts of the degradation model, respectively. The distribution functions of failure time and residual useful life are also approximated by the Monte Carlo simulation approach. Simulation results show the feasibility and good performance of the combined estimation method. Finally, real data analysis indicates that the jump-diffusion process model with the combined estimation method could give a more accurate estimation when predicting the failure time of the battery.

Piecewise-linear modelling with automated feature selection for Li-ion battery end-of-life prognosis

The complex nature of lithium-ion battery degradation has led to many machine learning-based approaches for health forecasting being proposed in the literature. However, machine learning using sophisticated models can be computationally expensive, and although linear models are faster they can also be inflexible. Piecewise-linear models offer a compromise—a fast and flexible alternative that is not as computationally expensive as techniques such as neural networks or Gaussian process regression. Here, a piecewise-linear approach for battery health forecasting, including an automated feature selection step, is compared to a Gaussian process regression model and found to perform equally well in terms of the median error on a training dataset, and indeed somewhat better at the 95th percentile of error. The feature selection process demonstrates the benefit of limiting the correlation between inputs. Further trials found that the piecewise-linear approach was robust to changing input size and availability of training data.