Optimizing Charging Control for Fast and Efficient Electric Vehicle Charging

Background: Electric vehicles (EVs) are at the heart of sustainable transportation, and the need for effective charging systems is steadily increasing. The difficulty is to optimize charging tactics that reduce charging time while retaining grid stability. To maximize performance and energy utilization, charging must be managed effectively using variables like battery levels, grid load, and environmental conditions. Objectives: The objective of this research is to create a sophisticated machine-learning technique for predicting the optimal charging control for electric vehicles. Particularly, it aims to use a Dual-Level Voting Boost (DLVB) algorithm to determine whether an EV charging session will be optimized or not, using input features such as battery level, charging rate, ambient temperature, and charging mode. Methodology: This paper proposes the Dual-Level Voting Boost (DLVB) algorithm, which is a two-stage ensemble learning method. The first level employs fundamental classifiers like Decision Trees, Logistic Regression, and K-Nearest Neighbors, whereas the second level employs more sophisticated models like Random Forest, Support Vector Machines, and Naive Bayes. This dual-layer structure enables a more precise classification of optimized and non-optimized charging sessions. The model's efficacy is assessed utilizing metrics like Accuracy, Precision, Recall, F1-Score, and the Matthews Correlation Coefficient (MCC). Results: On the test dataset, the DLVB algorithm obtained 95% accuracy, 94% precision, 93% recall, 94% F1 score, and 92% MCC. These findings show the efficiency of the dual-level voting system in correctly predicting optimized charging sessions, substantially surpassing conventional single-model methods. Conclusion: The DLVB algorithm has significant potential for improving EV charging control by correctly predicting optimization results. This approach provides a potential strategy for enhancing charging effectiveness, decreasing energy waste, and maintaining grid stability.