AUTHOR=Liu Yang , Xie Pingping , Yang Yinguo , Lu Qiuyu , Ma Xiyuan , Zhou Changcheng , Wu Guobing , Hu Xudong 

TITLE=Wind power output prediction in complex terrain based on modal decomposition attentional convolutional network

JOURNAL=Frontiers in Energy Research

VOLUME=Volume 11 - 2023

YEAR=2024

URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2023.1236597

DOI=10.3389/fenrg.2023.1236597

ISSN=2296-598X

ABSTRACT=In this work, modal decomposition is employed to generate more data for matching scenarios with more complex topography for wind farm output prediction in the case of complex terrain. The existing literature shows that a single wind power output prediction model is difficult to cope with complex terrain and thus the accuracy of wind power output prediction is limited. This work combines the advantages of attention mechanism and convolutional neural network for a novel network based on modal decomposition of historical data for wind power output prediction on complex terrain. The proposed novel network can break through the limitations of a single wind power output prediction model. In addition, the signals that are modally decomposed can be predicted more accurately. The proposed method is compared with various other algorithms for the wind power output prediction problem in complex terrain. Comparative experiments show that the proposed network achieves a higher accuracy rate.Keywords Complex terrain, modal decomposition, attention mechanism, convolutional neural network, wind power output prediction output in complex terrain is an important basis for ensuring the safe and stable operation of new power systems [4].Wind power output is affected by complex terrain and climate. The main existing popular datadriven wind power output prediction methods can be classified into the following five categories: time