Probabilistic Load Forecasting via Point Forecast Feature Integration

TL;DR

This paper introduces a two-stage probabilistic load forecasting framework that integrates point forecast features to enhance accuracy and uncertainty quantification in short-term power load predictions.

Contribution

It presents a novel approach combining point forecast features with probabilistic models, improving load forecasting accuracy and reliability.

Findings

Effective in hour-ahead load forecasting

Utilizes gradient boosting for point prediction

Employs quantile regression neural networks for probabilistic forecasts

Abstract

Short-term load forecasting is a critical element of power systems energy management systems. In recent years, probabilistic load forecasting (PLF) has gained increased attention for its ability to provide uncertainty information that helps to improve the reliability and economics of system operation performances. This paper proposes a two-stage probabilistic load forecasting framework by integrating point forecast as a key probabilistic forecasting feature into PLF. In the first stage, all related features are utilized to train a point forecast model and also obtain the feature importance. In the second stage the forecasting model is trained, taking into consideration point forecast features, as well as selected feature subsets. During the testing period of the forecast model, the final probabilistic load forecast results are leveraged to obtain both point forecasting and probabilistic forecasting. Numerical results obtained from ISO New England demand data demonstrate the effectiveness of the proposed approach in the hour-ahead load forecasting, which uses the gradient boosting regression for the point forecasting and quantile regression neural networks for the probabilistic forecasting.