Application of improved graph convolutional networks in daily-ahead carbon emission prediction

Autor: Feng Pan, Yuyao Yang, Yilin Ji, Jinli Li, Jun Zhang, Lihua Zhong
Jazyk: angličtina
Rok vydání: 2024
Předmět:
Zdroj: Frontiers in Energy Research, Vol 12 (2024)
Druh dokumentu: article
ISSN: 2296-598X
DOI: 10.3389/fenrg.2024.1371507
Popis: With the increasing complexity of power systems and the proliferation of renewable energy sources, the task of calculating carbon emissions has become increasingly challenging. To address these challenges, we developed a new method for predicting carbon emission factors. Bayesian optimization technique graphical convolutional networks with long- and short-term network (BO-TGNN) is used to predict the carbon emissions of the power system. The method aims to quickly predict the day-ahead carbon emissions of power system nodes with enhanced feature extraction and optimized network training hyperparameters. The effectiveness of the proposed method is demonstrated through simulation tests on three different power systems using four deep learning algorithms. The method provides a tailored solution to the evolving needs of carbon reduction efforts and is a significant step forward in addressing the complexity of carbon emission calculations for modern power systems.
Databáze: Directory of Open Access Journals