Adaptive Gradient-Weighted Stacking Ensemble for Electric Vehicle Energy Efficiency Prediction: A Novel Meta-Learning Approach
Abstract
One of the most important and complex issues of sustainable transportation is the energy efficiency prediction of electric vehicles (EVs), which is associated with intricate interdependences between the vehicles characteristics, motor capacity, charging ability, and performance efficiency. Conventional methods of machine learning do not always explain the complex non-linear dynamics and multi-scale relationships of EV energy consumption. This study presents a new meta-learning algorithm called Adaptive Gradient-Weighted Stacking (AGWS) Ensemble that combines both the gradient-based optimization and adaptive weight learning to achieve better energy efficiency prediction. The suggested system works with detailed EV data covering model year, manufacturer, vehicle type, motor power (kW), recharge time (hours) as well as car energy efficiency $(\text{km} / \text{kWh})$. AGWS has three levels of architecture (1) the heterogeneous base learners that render various facets of the efficiency landscape, (2) the adaptive optimization of weights depending on the prediction confidence, and the error pattern, and (3) Ridge regression meta-learning to combine the ensembles effectively. The comprehensive comparison of the results of AGWS with seven state-of-the-art algorithms indicates that it has an outstanding predictive accuracy with $\mathrm{R} 2=0.9153, \text{RMSE}=0.2456, \text{MAE}{=}$ 0.1694, and MAPE $=4.0055 \%$ indicating a 0.21 % improvement over the best model. The algorithm has…
Read the paper
Related studies
- Chaos and quantum thermalization
- The Dalton quantum chemistry program system
- Cord Blood Leptin and Adiponectin as Predictors of Adiposity in Children at 3 Years of Age: A Prospective Cohort Study
- Quantitative metabolomics by H-NMR and LC-MS/MS confirms altered metabolic pathways in diabetes
- Stable adiabatic quantum batteries
- A novel self-lipid antigen targets human T cells against CD1c(+) leukemias
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.