Probabilistic TCN-Transformer Network for Source-Load Prediction in Electricity Industry

With the growing penetration of renewable energy sources like photovoltaic (PV) and wind power, the intermittent nature of generation has posed challenges for grid stability and resource management. This paper introduces a hybrid deep learning framework that integrates Temporal Convolutional Networks (TCN) and Transformer models to enhance the accuracy of source-load forecasting in power systems. The model provides both deterministic forecasts, predicting point values for future power generation and load, and probabilistic forecasts, estimating prediction intervals to account for uncertainty using quantile regression. Tested on real-world datasets, the results show that the proposed TCN-Transformer model outperforms traditional methods, providing a robust solution for integrating renewable energy into modern grids.