Life Prediction Of Batteries Research Articles

A Recurrent Neural Network Model for Battery Capacity Fade Curve Prediction Using Early Life Data

Batteries are often considered the most performance and reliability-limiting component of any electronic and electrical assembly. Thus their degradation trend characterization through extensive testing and development of life prediction algorithms using testing data have been at the forefront of battery research. However, battery testing is also time-consuming, and it can take at least few months, even years, to test a battery for cycle life performance under normal operating conditions. Hence, there is a need to develop tools for battery life and degradation trend prediction using limited cycle data. Early life prediction tools can facilitate battery qualification, timely replacement of batteries in critical applications, and even the secondary use market. In this study, we present a recurrent neural network-linear network model to predict the entire battery capacity fade curve using less than 100 cycles of data (~ 1 month of testing). The discharge voltage-capacity curves have been used as input to the model. Our approach is also able to predict the rollover point (knee point) in the capacity fade curve of the battery, which indicates the onset of increased capacity fade rate. We use the publicly available MIT dataset to show the validity of the approach and extend the previous work from Severson et al. [1] on the end-of-life prediction to the entire capacity fade trend prediction. The recurrent neural network can capture the sequential changes in discharge voltage-capacity curves with cycles and provide flexibility in terms of the input data size.

Battery cycle life study through relaxation and forecasting the lifetime via machine learning

Battery lifetime modeling and prediction of precise capacity degradation for real-life applications are critical to understanding the complex and non-linear battery behavior. However, the application of accurate and robust aging models on dynamic on-road scenarios is still a challenge. In this work, a comprehensive aging dataset of 40 Nickel Manganese Cobalt (NMC) cells is generated for two years considering distinct relaxation phases in the function of the state of charge (SoC), temperature, and time. A qualitative analysis of the diversified aging parameters along with the sensitivity analysis of the rest criteria is conducted. Taking the discharge capacity as the pivotal predictor, a robust training dataset is built and preliminary fed to common data-driven models. Among them, the Gaussian process regression (GPR) is identified to be the best suit with which a 0.02% root-mean-squared error (RMSE) can be achieved for battery life prediction when tested with a static profile choosing an exponential kernel. Further, to demonstrate a real-life scenario, a worldwide harmonized light-duty test cycle (WLTC) is performed, and the capacity fade percentile can be predicted accurately with a 0.05% RMSE. This research shows that data-driven algorithms like GPR can be a promising online tool that can forecast the entire lifetime with high precision for dynamic profiles.

Solid-State Lithium Battery Cycle Life Prediction Using Machine Learning

Battery lifetime prediction is a promising direction for the development of next-generation smart energy storage systems. However, complicated degradation mechanisms, different assembly processes, and various operation conditions of the batteries bring tremendous challenges to battery life prediction. In this work, charge/discharge data of 12 solid-state lithium polymer batteries were collected with cycle lives ranging from 71 to 213 cycles. The remaining useful life of these batteries was predicted by using a machine learning algorithm, called symbolic regression. After populations of breed, mutation, and evolution training, the test accuracy of the quantitative prediction of cycle life reached 87.9%. This study shows the great prospect of a data-driven machine learning algorithm in the prediction of solid-state battery lifetimes, and it provides a new approach for the batch classification, echelon utilization, and recycling of batteries.

Remaining useful life prediction of lithium-ion batteries based on Monte Carlo Dropout and gated recurrent unit

Lithium-ion batteries have been widely used for energy storage systems and vehicle industries, highly accurate remaining useful life (RUL) prediction of lithium-ion batteries is one of the key technologies on prognostics and health management. However, the uncertainty quantification and reliability of RUL prediction is ignored. To describe the uncertainty of the RUL prediction and avoid the over-fitting phenomenon, a model combining Monte Carlo Dropout (MC_dropout) and gated recurrent unit (GRU) is proposed. Firstly, the indirect health indicator is extracted and gray relation analysis (GRA) is used to analyze the relation with capacity Then, dropout method is combined with GRU model to avoid the phenomenon of gradient disappearance and over-fitting. The probability distribution of the prediction results and 95% confidence interval is obtained through the MC_dropout method, which established the uncertainty of the prediction model. Finally, propagation neural network based on particle swarm optimization (BPNN-PSO), the least-squares support vector machine (LS-SVM) is compared with the proposed method. The results show that the proposed model not only has superiority point prediction of RUL compared with other methods, but also can describe the uncertainty of RUL prediction.

Remaining Useful Life Prediction of Lithium Battery Based on Sequential CNN–LSTM Method

Abstract Among various methods for remaining useful life (RUL) prediction of lithium batteries, the data-driven approach shows the most attractive character for non-linear relation learning and accurate prediction. However, the existing neural network models for RUL prediction not only lack accuracy but also are time-consuming in model training. In this paper, the sequential convolutional neural network–long short-term memory (CNN–LSTM) method is proposed for accurate RUL prediction of lithium batteries. First, degradation trajectories are analyzed, and six features are adopted for RUL prediction. Then, the CNN model is introduced for filtering the data features of degradation characters. And the orthogonal experiment is studied for optimizing the hyperparameters of the CNN model. Furthermore, by inputting the time-series features flattened by CNN and non-time series feature, the LSTM is reconstructed for memorizing the long-term degradation data of lithium battery. Finally, the proposed method is validated by four cells under different aging conditions. Comparing with the isolated models, the RUL prediction of sequential CNN–LSTM method has higher accuracy.

Comparative analysis of hybrid vehicle energy management strategies with optimization of fuel economy and battery life

Hybrid Electric Vehicles (HEV) form an important category of the automotive segment and most importantly fill the transition between internal combustion engines powered conventional vehicles and electric motor-powered vehicles. One of the main propagandas and sales factor of HEV is improved fuel economy over conventional vehicles. Over the recent years, considering the longevity of HEV usage, battery life evaluation has been brought to the forefront of research along with fuel economy and the key method to slow down the battery aging is through Energy Management Strategy (EMS). This research paper presents the comparative analysis of battery life optimization with different control strategies in a parallel hybrid vehicle. In the available research work, EMS considering battery aging is still lacking. This research work considers the impact of multiple parameters, including temperature and current on battery aging, providing more accurate battery life prediction. Four different control strategies are analyzed including, Thermostat, Fuzzy logic, Adaptive Equivalent Consumption Minimization Strategy (A-ECMS) and Q-learning considering battery aging. Results are compared concerning battery aging and fuel economy. In this research, ECMS results show a 25% improved fuel economy compared to the rule-based strategy. Also, a market cost-analysis is depicted to show the monetary savings for each of the energy management strategy.

Remaining Useful Life Prediction of Lithium-ion Batteries Based on Wiener Process Under Time-Varying Temperature Condition

Time-varying temperature condition has a significant impact on discharge capacity and aging law of lithium-ion battery. Consequently, a novel remaining useful life (RUL) prediction method for lithium-ion battery under time-varying temperature condition is developed in this paper. Firstly, a stochastic degradation rate model based on Arrhenius temperature model is proposed, and an interesting battery capacity conversion path from random temperature condition to reference temperature condition is established. Secondly, the aging model of lithium-ion battery under time-varying temperature condition is developed based on Wiener process, and a two-step unbiased estimation method based on maximum likelihood estimation (MLE) combined with genetic algorithm (GA) is proposed. Next, the random parameter is online updated under Bayesian framework. Then the probability density function (PDF) of the RUL for lithium-ion battery under time-varying temperature condition is derived. Finally, a case study is implemented to verify the effectiveness, and the results show that the proposed prediction method has higher accuracy and smaller uncertainty.

Remaining useful life prediction of lithium-ion battery based on auto-regression and particle filter

PurposeWith the rapid development and stable operated application of lithium-ion batteries used in uninterruptible power supply (UPS), the prediction of remaining useful life (RUL) for lithium-ion battery played an important role. More and more researchers paid more attentions on the reliability and safety for lithium-ion batteries based on prediction of RUL. The purpose of this paper is to predict the life of lithium-ion battery based on auto regression and particle filter method.Design/methodology/approachIn this paper, a simple and effective RUL prediction method based on the combination method of auto-regression (AR) time-series model and particle filter (PF) was proposed for lithium-ion battery. The proposed method deformed the double-exponential empirical degradation model and reduced the number of parameters for such model to improve the efficiency of training. By using the PF algorithm to track the process of lithium-ion battery capacity decline and modified observations of the state space equations, the proposed PF + AR model fully considered the declined process of batteries to meet more accurate prediction of RUL.FindingsExperiments on CALCE dataset have fully compared the conventional PF algorithm and the AR + PF algorithm both on original exponential empirical degradation model and the deformed double-exponential one. Experimental results have shown that the proposed PF + AR method improved the prediction accuracy, decreases the error rate and reduces the uncertainty ranges of RUL, which was more suitable for the deformed double-exponential empirical degradation model.Originality/valueIn the running of UPS device based on lithium-ion battery, the proposed AR + PF combination algorithm will quickly, accurately and robustly predict the RUL of lithium-ion batteries, which had a strong application value in the stable operation of laboratory and other application scenarios.

Novel Practical Life Cycle Prediction Method by Entropy Estimation of Li-Ion Battery

The growth of the lithium-ion battery market is accelerating. Although they are widely used in various fields, ranging from mobile devices to large-capacity energy storage devices, stability has always been a problem, which is a critical disadvantage of lithium-ion batteries. If the battery is unstable, which usually occurs at the end of its life, problems such as overheating and overcurrent during charge-discharge increase. In this paper, we propose a method to accurately predict battery life in order to secure battery stability. Unlike the existing methods, we propose a method of assessing the life of a battery by estimating the irreversible energy from the basic law of entropy using voltage, current, and time in a realistic dimension. The life estimation accuracy using the proposed method was at least 91.6%, and the accuracy was higher than 94% when considering the actual used range. The experimental results proved that the proposed method is a practical and effective method for estimating the life of lithium-ion batteries.

Life Prediction of Battery Using a Neural Gaussian Process with Early Discharge Characteristics.

The state of health (SOH) prediction of lithium-ion batteries (LIBs) is of crucial importance for the normal operation of the battery system. In this paper, a new method for cycle life and full life cycle capacity prediction is proposed, which combines the early discharge characteristics with the neural Gaussian process (NGP) model. The cycle data sets of commercial LiFePO4(LFP)/graphite cells generated under different operating conditions are analyzed, and the power characteristic P is extracted from the voltage and current curves of the early cycles. A Pearson correlation analysis shows that there is a strong correlation between P and cycle life. Our model achieves 8.8% test error for predicting cycle life using degradation data for the 20th to 110th cycles. Based on the predicted cycle life, capacity degradation curves for the whole life cycle of the cells are predicted. In addition, the NGP method, combined with power characteristics, is compared with other classical methods for predicting the remaining useful life (RUL) of LIBs. The results demonstrate that the proposed prediction method of cycle life and capacity has better battery life and capacity prediction. This work highlights the use of early discharge characteristics to predict battery performance, and shows the application prospect in accelerating the development of electrode materials and optimizing battery management systems (BMS).

Remaining Useful Life Prediction of Lithium Batteries Based on Extended Kalman Particle Filter

The prognosis of time‐to‐failure for a battery can avoid the failure caused by battery performance loss. In this paper, a novel and effective algorithm is proposed to predict the remaining useful life of lithium‐ion batteries. The extended Kalman particle filter is used to improve particle degradation problem existing in standard particle filter algorithm. In order to fit battery capacity degradation, a transformed model is proposed based on double exponential empirical degradation model. It can reduce the number of parameters and the training difficulty of parameters; it also matches the form of state transfer equation. In order to improve prediction accuracy, the auto regression model is introduced to correct observation values produced by observation equation. Experimental results show that the proposed algorithm can effectively improve the accuracy of prediction compared with other algorithms. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

A review on state of health estimations and remaining useful life prognostics of lithium-ion batteries

Lithium-ion batteries have been generally used in industrial applications. In order to ensure the safety of the power system and reduce the operation cost, it is particularly important to accurately and timely estimate the state of health (SOH) and predict the remaining useful life (RUL) of lithium-ion batteries. With the development of intelligent tools such as artificial intelligence, big data analysis and the Internet of Things, the methods of battery health assessment have been gradually diversified. Here, we have compiled four publicly available battery datasets. The SOH estimations and RUL prognostics of lithium-ion batteries are reviewed by analyzing the research status. To this end, after studying different scientific and technical literatures, the respective methods are divided into specific groups, and the advantages and limitations of the battery management system application are discussed. At the end, the future development trend and research challenges are analyzed. All key insights in this review will hopefully drive the development of battery health estimation and life prediction techniques.

Remaining Useful Life Prediction of Lithium-Ion Batteries Based on a Mixture of Ensemble Empirical Mode Decomposition and GWO-SVR Model

The remaining useful life (RUL) prediction of lithium-ion batteries (LIBs) plays a vital role in their prognostics and health management (PHM). A battery degradation model is of great significance to maintain and replace the batteries avoiding the hazards in advance to ensure the safety and reliability of an energy storage system. In this article, a novel model is developed based on an integration of ensemble empirical mode decomposition (EEMD), gray wolf optimization, and support vector regression (GWO-SVR) to predict the RUL of LIBs. A GWO-SVR model is proposed to predict the RUL of LIBs, where the GWO algorithm is utilized to optimize the SVR kernel parameters. The EEMD is employed to decouple global degradation and local regeneration in battery capacity time series to improve prediction accuracy. This design scheme captures the global degradation behavior and local regeneration phenomenon in LIBs. The experimental results on LIB from the NASA Ames Prognostics Center of Excellence (PCoE) verify that the proposed method effectively improves the accuracy of RUL prediction of LIBs.

State of Health Diagnosis and Remaining Useful Life Prediction of Lithium-Ion Batteries Based on Multi-Feature Data and Mechanism Fusion

State of Health (SOH) Diagnosis and Remaining Useful Life (RUL) Prediction of lithium-ion batteries (LIBs) are subject to low accuracy due to the complicated aging mechanism of LIBs. This paper investigates a SOH diagnosis and RUL prediction method to improve prediction accuracy by combining multi-feature data with mechanism fusion. With the approach of the normal particle swarm optimization, a support vector regression (SVR)-based SOH diagnosis model is developed. Compared with existing works, more comprehensive features are utilized as the feature variables, including three aspects: the representative feature during a constant-voltage protocol; the capacity; internal resistance. Further, the optimized regularized particle filter (ORPF) model with uncertainty expression is integrated to obtain more accurate RUL prediction and SOH diagnosis. Experiments validate the effectiveness of the proposed method. Results show that the proposed SOH diagnosis and RUL prediction method has higher accuracy and better stability compared with the traditional methods, which help to realize the decision of the maintenance process.

Comparing Single and Hybrid methods of Deep Learning for Remaining Useful Life Prediction of Lithium-ion Batteries

The prediction lifetime of a Lithium-ion battery is able to be utilized as an early warning system to prevent the battery’s failure that makes it very significant for assuring safety and reliability. This paper represents a benchmark study that compares its RUL prediction results of single and hybrid methods with similar articles. We suggest a hybrid method, named the CNN-LSTM, which is a combination of Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM), for predicting and improving the accuracy of the remaining useful life (RUL) of Lithium-ion battery. We selected three statistical indicators (MAE, R², and RMSE) to assess the results of performance prediction. Experimental validation is performed using the lithium-ion battery dataset from the NASA and results reveal that the effectiveness of the suggested hybrid method in reducing the prediction error and in achieving better RUL prediction performance compared to the other algorithms.

Study on Remaining Useful Life Prediction of Lithium-ion Batteries Based on Charge Transfer Resistance

Study on Remaining Useful Life Prediction of Lithium-ion Batteries Based on Charge Transfer Resistance

Remaining Useful Cycle Life Prediction of Lithium-Ion Battery Based on Differential Voltage Analysis

Remaining Useful Cycle Life Prediction of Lithium-Ion Battery Based on Differential Voltage Analysis

An Improved Rainflow Algorithm Combined with Linear Criterion for the Accurate Li-ion Battery Residual Life Prediction

Li-ion battery health assessment has been widely used in electric vehicles, unmanned aerial vehicle and other fields. In this paper, a new linear prediction method is proposed. By weakening the sensitivity of the Rainflow algorithm to the peak data, it can be applied to the field of battery, and can accurately count the number of Li-ion battery cycles, and skip the cumbersome link of parameter identification. Then, a linear criterion is proposed based on the idea of proportion, which makes the life prediction of Li-ion battery linear. Under the verification of multiple sets of data, the prediction error of this method is kept within 2.53%. This method has the advantages of high operation efficiency and simple operation, which provides a new idea for battery life prediction in the field of electric vehicles and aerospace.

Roughsets-based Approach for Predicting Battery Life in IoT

Internet of Things (IoT) and related applications have successfully contributed towards enhancing the value of life in this planet. The advanced wireless sensor networks and its revolutionary computational capabilities have enabled various IoT applications become the next frontier, touching almost all domains of life. With this enormous progress, energy optimization has also become a primary concern with the need to attend to green technologies. The present study focuses on the predictions pertinent to the sustainability of battery life in IoT frameworks in the marine environment. The data used is a publicly available dataset collected from the Chicago district beach water. Firstly, the missing values in the data are replaced with the attribute mean. Later, one-hot encoding technique is applied for achieving data homogeneity followed by the standard scalar technique to normalize the data. Then, rough set theory is used for feature extraction, and the resultant data is fed into a Deep Neural Network (DNN) model for the optimized prediction results. The proposed model is then compared with the state of the art machine learning models and the results justify its superiority on the basis of performance metrics such as Mean Squared Error, Mean Absolute Error, Root Mean Squared Error, and Test Variance Score.

Deep Neural Networks Based Approach for Battery Life Prediction

The Internet of Things (IoT) and related applications have witnessed enormous growth since its inception. The diversity of connecting devices and relevant applications have enabled the use of IoT devices in every domain. Although the applicability of these applications are predominant, battery life remains to be a major challenge for IoT devices, wherein unreliability and shortened life would make an IoT application completely useless. In this work, an optimized deep neural networks based model is used to predict the battery life of the IoT systems. The present study uses the Chicago Park Beach dataset collected from the publicly available data repository for the experimentation of the proposed methodology. The dataset is pre-processed using the attribute mean technique eliminating the missing values and then One-Hot encoding technique is implemented to convert it to numerical format. This processed data is normalized using the Standard Scaler technique. Moth Flame Optimization (MFO) Algorithm is then implemented for selecting the optimal features in the dataset. These optimal features are finally fed into the DNN model and the results generated are evaluated against the state-of-the-art models, which justify the superiority of the proposed MFO-DNN model.