Ciaran Gilbert | arXiv Science

arXiv2026

Leveraging Continuous Trade Data for Intraday Imbalance Price Forecasting

Forecasts of imbalance prices are essential for participants in short-term electricity markets. This study proposes an approach to integrate trade data from continuous intraday markets into forecasting methods and considers the production of both parametric and non-parametric density forecasts. Extensive feature engineering is required to convert irregular trade data into a suitable tabular structure. Our findings demonstrate that intraday trade data can enhance state-of-the-art forecasts based on quantile regression and LASSO. We also propose a model based on Natural Gradient Boosting to predict the means and variances of a bimodal Gaussian mixture, which offers both improved predictive performance and computational efficiency. In a case study based on multiple years of data from Great Britain, our proposed approach reduced the Pinball Loss of state-of-the-art models by 20% for forecasts made 30 seconds before gate closure, and by 7.5% for forecasts made one hour before delivery.

stat.AP↗

arXiv2023

Probabilistic load forecasting for the low voltage network: forecast fusion and daily peaks

Short-term forecasts of energy consumption are invaluable for the operation of energy systems, including low voltage electricity networks. However, network loads are challenging to predict when highly desegregated to small numbers of customers, which may be dominated by individual behaviours rather than the smooth profiles associated with aggregate consumption. Furthermore, distribution networks are challenged almost entirely by peak loads, and tasks such as scheduling storage and/or demand flexibility maybe be driven by predicted peak demand, a feature that is often poorly characterised by general-purpose forecasting methods. Here we propose an approach to predict the timing and level of daily peak demand, and a data fusion procedure for combining conventional and peak forecasts to produce a general-purpose probabilistic forecast with improved performance during peaks. The proposed approach is demonstrated using real smart meter data and a hypothetical low voltage network hierarchy comprising feeders, secondary and primary substations. Fusing state-of-the-art probabilistic load forecasts with peak forecasts is found to improve performance overall, particularly at smart-meter and feeder levels and during peak hours, where improvement in terms of CRPS exceeds 10%.

stat.AP↗