What question did this study set out to answer?

The aim is to develop a forecasting framework for electricity demand using advanced machine learning models.

April 16, 20260 citationsOpen Access

Electricity Demand Forecasting Using Transformer-Based Model Optimization and Interval Calibration

Key Points

The aim is to develop a forecasting framework for electricity demand using advanced machine learning models.
Utilized multiple machine learning models including Naive 7, XGBoost, Random Forest, and Transformer.
Employed hyperparameter optimization to improve model performance.
Conducted time-series cross-validation for robust evaluation.
Integrated conformal prediction for uncertainty quantification.
The Transformer model outperformed other models in accuracy metrics.
Achieved the lowest Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE).
Recorded the highest R² and Prediction Interval Coverage Probability (PICP) values.

Abstract

This study proposes a power load prediction framework based on multiple machine learning models, including Naive 7, XGBoost, Random Forest, and a deep learning model based on Transformer. The framework integrates hyperparameter optimization, time-series cross-validation, and uncertainty quantification using the conformal prediction (CP) method. The results show that the Transformer model achieves higher accuracy than the other models, with the lowest MAE, RMSE, and MAPE, and the highest R² and PICP. The Transformer model is effective at capturing long-term dependencies and providing a reliable forecast range, making it suitable for dynamic power systems.

اسأل الذكاء الاصطناعي

Bookmark

View Full Paper