Hierarchical Probabilistic Conformal Prediction for Distributed Energy Resources Adoption
Researchers have developed a new framework for predicting the adoption of distributed energy resources (DERs) by incorporating hierarchical probabilistic conformal prediction. This method addresses the challenges of uncertainty and spatial disparity in DER growth, ensuring statistical guarantees at both circuit and substation levels. By utilizing a multivariate Hawkes process and a tailored split conformal prediction algorithm, the approach aims to improve accuracy and calibration in forecasting for grid management and infrastructure planning. AI
IMPACT Provides a novel statistical framework for improving predictions in energy resource management.