Data-driven state-of-health estimation for lithium-ion battery based on aging features

Li, Xining; Ju, Lingling; Geng, Guangchao; Jiang, Quanyuan

doi:10.1016/j.energy.2023.127378

Cited by 43 publications

(7 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, SOH C represents a hybrid definition of SOH based on battery health factors. 40 Figure 2 presents the model parameters of battery C1 at SOC of 0.6, obtained through online identification using the Kalman filter method. Figure 2 shows that capacity and internal resistance increase during the first six cycles of the battery.…”

Section: Battery Degradation Datasets and Methodsmentioning

confidence: 99%

A Novel State-of-Health Prediction and Assessment Strategies for High-Capacity Mining Lithium-Ion Batteries Based on Multi-Indicator

Feng,

Cai,

Zhan

et al. 2024

J. Electrochem. Soc.

View full text Add to dashboard Cite

Accurately assessing battery state of health (SOH) is crucial for ensuring the safety of lithium-ion batteries. However, current SOH evaluation methods suffer from inconsistent criteria and limited accuracy in prediction models. This paper introduces a novel SOH prediction and assessment strategy that relies on multiple indicators to address these challenges. First, multifaceted health factors are extracted based on charge cycle data, including battery charging time, incremental capacity, and dV/dt curve. Subsequently, a support vector regression model optimized by the sparrow search algorithm is proposed to predict SOH. The results show that MAE, RMSE, and MAPE are less than 0.037%, 0.047%, and 0.04%, respectively. Meanwhile, the Kalman filtering method is used to identify the second-order RC model online, and the relative SOH curves are obtained by defining the SOH through the internal resistance. Finally, by analyzing the effects of capacity and internal resistance changes on SOH, a new strategy for SOH assessment is proposed, which considers various factors and selects an appropriate judgment mechanism according to the characteristics exhibited by the battery at different life stages. The strategy is more conservative and reliable, providing a solid guarantee for the safe operation of mining equipment.

show abstract

Section: Battery Degradation Datasets and Methodsmentioning

confidence: 99%

A Novel State-of-Health Prediction and Assessment Strategies for High-Capacity Mining Lithium-Ion Batteries Based on Multi-Indicator

Feng,

Cai,

Zhan

et al. 2024

J. Electrochem. Soc.

View full text Add to dashboard Cite

show abstract

“…High-quality data can provide more accurate estimation results, while effective feature extraction can capture key features of battery health status. Reference [19] is based on incremental capacity (IC) analysis and battery operating characteristics combined with a regression model to correct for the bias caused by individual batteries. The method was validated on laboratory and EV datasets, with average absolute percentage errors of 0.29% and 3.20%, respectively.…”

Section: Introductionmentioning

confidence: 99%

Joint Estimation of State of Charge and State of Health of Lithium-Ion Batteries Based on Stacking Machine Learning Algorithm

Dong,

Chen,

Zhang

et al. 2024

WEVJ

View full text Add to dashboard Cite

Conducting online estimation studies of the SOH of lithium-ion batteries is indispensable for extending the cycle life of energy storage batteries. Data-driven methods are efficient, accurate, and do not depend on accurate battery models, which is an important direction for battery state estimation research. However, the relationships between variables in lithium-ion battery datasets are mostly nonlinear, and a single data-driven algorithm is susceptible to a weak generalization ability affected by the dataset itself. Meanwhile, most of the related studies on battery health estimation are offline estimation, and the inability for online estimation is also a problem to be solved. In this study, an integrated learning method based on a stacking algorithm is proposed. In this study, the end voltage and discharge temperature were selected as the characteristics based on the sample data of NASA batteries, and the B0005 battery was used as the training set. After training on the dataset and parameter optimization using a Bayesian algorithm, the trained model was used to predict the SOH of B0007 and B0018 models. After comparative analysis, it was found that the prediction results obtained based on the proposed model not only have high accuracy and a short running time, but also have a strong generalization ability, which has a great potential to achieve online estimation.

show abstract

“…Shallow learning methods encompass neural networks like multilayer perceptron (MLP) and extreme learning machine (ELM), support vector machines (SVM) and their variants, as well as stochastic techniques such as Gaussian and Wiener processes. Li et al [17] extracted internal and external health features based on electrochemical models (EM) and voltage and temperature curves, and constructed SOH estimation models using a back-propagation neural network (BPNN). This approach effectively enhances the estimation accuracy across various operating conditions and charge/discharge patterns.…”

Section: Introductionmentioning

confidence: 99%

A State-of-Health Estimation Method for Lithium Batteries Based on Fennec Fox Optimization Algorithm–Mixed Extreme Learning Machine

Sun,

Qin,

Yun

2024

Batteries

View full text Add to dashboard Cite

A reliable and accurate estimation of the state-of-health (SOH) of lithium batteries is critical to safely operating electric vehicles and other equipment. This paper proposes a state-of-health estimation method based on fennec fox optimization algorithm–mixed extreme learning machine (FFA-MELM). Firstly, health indicators are extracted from lithium-battery-charging data, and grey relational analysis (GRA) is employed to identify highly correlated features with the state-of-health of the battery. Subsequently, a state-of-health estimation model based on mixed extreme learning machine is constructed, and the hyperparameters of the model are optimized using the fennec fox optimization algorithm to improve estimation accuracy and convergence speed. The experimental results demonstrate that the proposed method has significantly improved the accuracy of the state-of-health estimation for lithium batteries compared to the extreme learning machine. Furthermore, it can achieve precise state-of-health estimation results for multiple batteries, even under complex operating conditions and with limited charge/discharge cycle data.

show abstract

Data-driven state-of-health estimation for lithium-ion battery based on aging features

Cited by 43 publications

References 29 publications

A Novel State-of-Health Prediction and Assessment Strategies for High-Capacity Mining Lithium-Ion Batteries Based on Multi-Indicator

A Novel State-of-Health Prediction and Assessment Strategies for High-Capacity Mining Lithium-Ion Batteries Based on Multi-Indicator

Joint Estimation of State of Charge and State of Health of Lithium-Ion Batteries Based on Stacking Machine Learning Algorithm

A State-of-Health Estimation Method for Lithium Batteries Based on Fennec Fox Optimization Algorithm–Mixed Extreme Learning Machine

Contact Info

Product

Resources

About