|
| 1 | +# Energy Consumption Prediction |
| 2 | + |
| 3 | +## Description |
| 4 | +A machine learning model to predict energy consumption patterns for buildings, households, or industrial facilities. This project helps optimize energy usage and reduce costs through accurate forecasting. |
| 5 | + |
| 6 | +## Project Structure |
| 7 | +``` |
| 8 | +Energy-Consumption-Prediction/ |
| 9 | +├── data/ # Dataset files |
| 10 | +├── notebooks/ # Jupyter notebooks |
| 11 | +├── src/ # Source code |
| 12 | +├── models/ # Saved models |
| 13 | +├── requirements.txt # Dependencies |
| 14 | +└── README.md # Project documentation |
| 15 | +``` |
| 16 | + |
| 17 | +## Dataset |
| 18 | +The dataset includes energy consumption data with features such as: |
| 19 | +- Temporal features (hour, day, month, season) |
| 20 | +- Weather conditions (temperature, humidity, wind speed) |
| 21 | +- Building characteristics (size, type, occupancy) |
| 22 | +- Historical consumption patterns |
| 23 | +- Holiday and weekend indicators |
| 24 | + |
| 25 | +## Installation |
| 26 | +```bash |
| 27 | +pip install -r requirements.txt |
| 28 | +``` |
| 29 | + |
| 30 | +## Usage |
| 31 | +```python |
| 32 | +from src.model import EnergyPredictor |
| 33 | + |
| 34 | +predictor = EnergyPredictor() |
| 35 | +predictor.load_model('models/energy_model.pkl') |
| 36 | +prediction = predictor.predict(input_features) |
| 37 | +``` |
| 38 | + |
| 39 | +## Model Details |
| 40 | +- **Algorithm**: LSTM, XGBoost, Random Forest, Prophet |
| 41 | +- **Features**: 25+ engineered features including lag variables |
| 42 | +- **Metrics**: MAE, RMSE, MAPE, R-squared |
| 43 | + |
| 44 | +## Results |
| 45 | +| Model | MAE | RMSE | MAPE | R-squared | |
| 46 | +|-------|-----|------|------|----------| |
| 47 | +| LSTM | 45.2 | 62.3 | 8.5% | 0.92 | |
| 48 | +| XGBoost | 48.1 | 65.7 | 9.1% | 0.90 | |
| 49 | +| Random Forest | 51.3 | 68.9 | 9.8% | 0.88 | |
| 50 | +| Prophet | 52.8 | 71.2 | 10.2% | 0.86 | |
| 51 | + |
| 52 | +## Applications |
| 53 | +- Smart grid optimization |
| 54 | +- Building energy management |
| 55 | +- Cost forecasting for utilities |
| 56 | +- Demand response planning |
| 57 | + |
| 58 | +## Contributing |
| 59 | +Contributions are welcome! Please read the contributing guidelines before submitting a pull request. |
| 60 | + |
| 61 | +## License |
| 62 | +MIT License |
0 commit comments