Feature representation optimization and dynamic hierarchical correlation modeling based ultra-short-term probabilistic power forecasting for distributed PV

doi:10.1016/j.enbuild.2025.116308

Feature representation optimization and dynamic hierarchical correlation modeling based ultra-short-term probabilistic power forecasting for distributed PV

2025-08-29

0

PRE

AI

DOI：10.1016/j.enbuild.2025.116308

Original

Original request for help

Save

Abstract

En 中文

• The hierarchical graph method can model correlations within and between subregions. • The feature representation optimization can exploit the deep features of power. • The distributed PV clustering realizes the effective division of subregions. • The probabilistic forecasting ensures the accuracy under all kinds of volatility.

Feature representation optimization and dynamic hierarchical correlation modeling based ultra-short-term probabilistic power forecasting for distributed PV

Abstract

Journal

Organization