On-line conformalized neural networks ensembles for probabilistic forecasting of day-ahead electricity prices

Probabilistic electricity price forecasting (PEPF) is subject of increasing interest, following the demand for proper quantification of prediction uncertainty, to support the operation in complex power markets with increasing share of renewable generation. Distributional neural network ensembles have been recently shown to outperform state of the art PEPF benchmarks. Still, they require critical reliability enhancements, as they fail to pass the coverage tests at various steps on the prediction horizon. In this work, we propose a novel approach to PEPF, extending the state-of-the-art methods based on neural network ensembles through conformal-inference techniques, deployed within an on-line recalibration procedure. Experiments have been conducted on multiple market regions, achieving day-ahead forecasts with improved hourly coverage and stable probabilistic scores. • Novel probabilistic electricity price forecasting approach. • Extending neural network based methods through conformal inference. • Experiments on German market and Italian regional bidding zones. • Improved hourly reliability while providing stable probabilistic scores. • Open GitHub repository provided to fully reproduce paper experiments.