CoCopula: Conformal Calibration and Copula Aggregation for Reliable Renewable Energy Forecasting

CoCopula is a framework that transforms site-level probabilistic forecasts into reliable fleet-level forecasts by combining Copula-based dependence modeling with Context-Aware Conformal Prediction. It captures cross-site correlations and mitigates aggregation-induced miscalibration, producing sharp prediction intervals with near-nominal empirical coverage. Experiments on large-scale solar generation data from MISO, ERCOT, and SPP show lower Winkler scores than competing baselines and performance comparable to directly calibrated system-level forecasts.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.