Electric Vehicle Charging Pattern Prediction Using Machine Learning
The Electric Vehicle Charging Pattern Prediction Using Machine Learning system is a comprehensive machine learning solution that predicts EV charging patterns, distinguishing between Standard and High-Intensity charging behaviors. This project is ideal for B.Tech, MCA, CSE, and engineering students as their final year project. It implements Logistic Regression and Random Forest with leak-free SMOTE, achieving 92.7% accuracy and 0.938 ROC-AUC.
The system utilizes a comprehensive dataset of 10,000 charging events containing 16 features including vehicle specifications, charging parameters, and environmental conditions. It provides a user-friendly web interface for data upload, exploratory data analysis, model training, and real-time charging pattern prediction, enabling utility companies and charging station operators to optimize grid load management and implement dynamic pricing strategies.
- Multi-Model Charging Pattern Classification
- Random Forest (92.7% Accuracy)
- Logistic Regression (88.4% Accuracy)
- Feature Engineering (charging intensity, SOC change)
- Leak-free SMOTE Oversampling
- Interactive Visualizations
- Real-time Pattern Predictions
- Model Performance Comparison
- Feature Importance Analysis
- Flask Web Application