Blockchain-Based Interpretable Electric Vehicle Battery Life Prediction in IoV
- 14 September 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (4) , 7214-7227
Abstract
The remarkable success of deep learning (DL) in predicting battery health has prompted interest in its application in recent years. While state-of-the-art DL models have achieved high accuracy in battery health prediction, they have not been widely adopted in industrial workflows, primarily due to their lack of interpretability and security. To address this issue, we propose a blockchain-based interpretable prediction algorithm for battery health prediction in electric vehicles (EVs) within the Internet of Vehicles (IoV). Specifically, the proposed method includes a platform architecture for a blockchain-based DL system, ensuring secure storage of user data during the prediction process. Notably, we develop a novel battery life prediction algorithm called BLP-Transformer, which leverages short-term relationships between degraded data and explains the impact of feature extraction on predicted results through the contribution of aggregated features based on a feature focusing mechanism. Experimental results demonstrate that the system is feasible for security and can provide accurate battery life prediction. In addition, the comparison study further highlights the superiority of the proposed algorithm in terms of robustness, prediction accuracy, and model interpretability.Keywords
This publication has 43 references indexed in Scilit:
- Bayesian Inference for Kendall’s Rank Correlation CoefficientThe American Statistician, 2018
- XGBoost2016
- Mean Absolute Percentage Error for regression modelsNeurocomputing, 2016
- State-of-charge estimation for battery management system using optimized support vector machine for regressionJournal of Power Sources, 2014
- Health diagnosis and remaining useful life prognostics of lithium-ion batteries using data-driven methodsJournal of Power Sources, 2013
- Behavior and state-of-health monitoring of Li-ion batteries using impedance spectroscopy and recurrent neural networksInternational Journal of Electrical Power & Energy Systems, 2012
- Remaining useful life estimation – A review on the statistical data driven approachesEuropean Journal of Operational Research, 2011
- Emerging Energy-Efficient Technologies for Hybrid Electric VehiclesProceedings of the IEEE, 2007
- Another look at measures of forecast accuracyInternational Journal of Forecasting, 2006
- Random ForestsMachine Learning, 2001