Ocean wind energy downscaling method based on deep learning neural network

Abstract

The invention discloses an ocean wind energy downscaling method based on a deep learning neural network. The method comprises: collecting for a period; performing 0.2-0.3 degree high-resolution reanalysis on day-by-day sea surface 100m wind field data in the data, performing 1-3 degree low-resolution reanalysis on day-by-day sea surface 10m wind field and sea level air pressure field data in the data, and performing 1-3 degree low-resolution global climate mode data on day-by-day sea surface 10m wind field and sea level air pressure field data in the data; carrying out normalization processingon the collected data; training a deep learning neural network model; and applying a deep learning neural network model to obtain 0.2-0.3 degree high-resolution day-by-day sea surface 100m wind fielddata after downscaling of the target area. According to the method, through the deep learning neural network, intrinsic characteristics and essential rules contained in the data are extracted, downscaling errors are reduced, and prediction of ocean wind energy is more accurate.

Description

technical field [0001] The present invention relates to a method for downscaling, in particular, the present invention relates to a method for downscaling ocean wind energy based on a deep learning neural network. The invention belongs to the technical field of marine wind energy prediction. Background technique [0002] my country's ocean wind energy resources are extremely rich, suitable for large-scale development and utilization. Since wind energy resources are a type of climate resources, climate change has a significant impact on wind energy resources. The study found that the surface wind speed in most areas of China showed a decreasing trend in the past few decades, resulting in a 15-17% decrease in the power generation of wind farms in some areas, which brought large economic losses to the invested wind farms. Therefore, it is necessary to scientifically predict the impact of future climate change on marine wind energy, and provide a feasible reference for the inv...

AI Technical Summary

Problems solved by technology

[0005] In view of the problem of large downscaling errors in the prior art, the purpose of the present invention is to provide a downscaling method suitable for marine wind energy based on deep learning neural network, so as to reduce the downscaling error

Images