Precise remaining useful life (RUL) estimation for lithium-ion batteries is essential for improving the safety, reliability, and maintenance of electric vehicles (EVs). This study proposes a random forest (RF)-based ensemble learning framework using the publicly available Hawaii Natural Energy Institute (HNEI) dataset containing 15,064 charge-discharge cycles. Seven degradation-related features, including cycle index, discharge time, voltage decrement, maximum discharge voltage, minimum charging voltage, time at 4.15 V, and constant-current charging duration, are extracted to characterize battery aging. The proposed RF model is compared with linear regression (LR), long short-term memory (LSTM), and attention-LSTM models using MAE, root mean square error (RMSE), mean absolute percentage error (MAPE), and coefficient of determination (R²). RF demonstrates superior prediction performance, achieving MAE of 5.20, RMSE of 6.83, MAPE of 1.33%, and R² of 0.997. Parity and residual analyses further confirm its strong predictive consistency. The proposed approach provides an accurate, computationally efficient, and interpretable solution for BMS applications, enabling effective battery health monitoring, predictive maintenance, charging optimization, and timely replacement.
Ponkumar Ganesapandiyan, P. Hemachandu, N. Rajavinu et al.· International Journal of Pow...· 0 citations
The prediction of lithium-ion battery capacity degradation plays a vital role in ensuring safe and efficient operation in electric mobility and renewable energy applications. This paper evaluates standalone machine learning, deep learning, and hybrid models for battery capacity estimation. The evaluated ML models include random forest, gradient boosting, and extreme gradient boosting (XGBoost), while the DL model employs a multilayer perceptron. The hybrid framework combines DL based feature extraction with ensemble ML regression or classification. A real-world dataset comprising temperature, resistance, reactance, and battery type was preprocessed, scaled, and divided into training and testing subsets. Hyperparameter tuning, k-fold cross-validation, and uncertainty quantification were incorporated to improve reliability and reproducibility. Model performance was assessed using RMSE, MAE, and R² for regression and receiver operating characteristic–area under the curve (ROC-AUC) and F1-score for classification. ROC curves, calibration curves, metric-comparison charts, cycle-wise degradation plots, and residual analyses were used for evaluation. Results demonstrate that the hybrid model outperforms standalone approaches by reducing RMSE and improving calibration, reliability, uncertainty alignment, and interpretability. This also establishes its novelty over existing state of health (SOH) models and highlights future extensions involving LSTM-based temporal modeling and chemistry-adaptive transfer learning. Overall, hybrid modeling provides a promising solution for reliable predictive battery maintenance.
Shobana Devendiren, A. Muthuraman, M. Vanitha et al.· International Journal of Pow...· 0 citations
This paper presents an integrated internet of things (IoT) and machine learning-based framework for secure and efficient demand-side management (DSM) in modern smart grids. The proposed approach combines long short-term memory (LSTM) networks for accurate load forecasting, federated learning (FL) for decentralized privacy-preserving model training, and blockchain technology for secure and tamper-proof communication. In addition, a digital twin (DT)-assisted architecture is incorporated to enable predictive decision-making and system-level optimization. The framework explicitly considers renewable energy integration, electric vehicle (EV) charging loads, distributed energy storage, and power electronic converter constraints. Simulation results demonstrate a reduction in forecasting error by 15-25%, a 25% decrease in daily energy cost, and a 23.7% reduction in peak demand. The proposed system achieves improved voltage stability with reduced deviation and enhances overall efficiency up to 88%. Furthermore, cybersecurity performance is validated with an anomaly detection AUC of 0.95 and reduced data transmission through FL by 87%. The results confirm that the proposed framework provides a scalable, secure, and intelligent solution for next-generation smart grid applications.
S. Pushpa, Jonnadula Narasimharao, K. L. Kishore et al.· International Journal of Pow...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
