Method and system for wind speed prediction based on atmospheric disturbance

Abstract

The invention discloses a method and a system for wind speed prediction based on atmospheric disturbance. The prediction method comprises: acquiring a trained wind speed prediction model based on a long-short time memory network; obtaining a preliminary prediction sequence of wind speed according to the wind speed prediction model, wherein each time in the preliminary prediction sequence corresponds to a preliminary prediction value of wind speed; obtaining initial conditions and parameters of a Lorentz equation; obtaining a wind speed disturbance sequence according to the initial condition and the parameter, the wind speed disturbance sequence being a numerical solution of the Lorentz equation, the wind speed disturbance sequence comprising a wind speed disturbance value corresponding toeach time; modifying the preliminary prediction sequence according to the wind speed disturbance sequence, to obtain a final prediction sequence of the wind speed. By adopting the wind speed prediction method or the wind speed prediction system, a prediction result is closer to a real wind speed sequence and has smaller prediction error.

Description

technical field [0001] The invention relates to the field of wind speed prediction, in particular to a wind speed prediction method and system based on atmospheric disturbance. Background technique [0002] Since the end of the 20th century, the continuous increase in world energy consumption has led to continuous warming of the global climate, while further destroying the balance of natural ecosystems, threatening human food supply and living environment. According to data released by the International Energy Agency on March 22, 2018, global carbon emissions reached 32.5 billion tons in 2017, an increase of 1.4% over 2016 (a net increase of 460 million tons), a record high. The reason is strong global energy demand. At present, the development and utilization of renewable energy has become the strategic choice of most countries in the world. Many countries regard the development of renewable energy as an important measure to alleviate energy supply contradictions and addre...

AI Technical Summary

Problems solved by technology

Since wind energy may change significantly within a few hours, the biggest problem with wind power generation is its dependence on wind energy volatility, which further leads to wind power volatility, which is not conducive to the grid-connected utilization of wind power

Therefore, accurate wind speed prediction is the key to large-scale development and utilization of wind energy resources, but it is also a very difficult link to control

[0003] In recent years, researchers at home and abroad have carried out in-depth research on wind speed prediction, and proposed various wind speed prediction methods, such as wind speed prediction methods based on meteorological data, wind speed prediction methods based on least squares method, and statistical regression analysis based and neural network prediction methods, wind speed prediction methods based on artificial intelligence algorithms, etc., but they almost all stay on the improvement of the algorithm, while ignoring the nonlinear characteristics of wind energy in the atmospheric dynamic system

Therefore, in the traditional wind speed prediction method, the accuracy of the general prediction is not high

Images