Wind power prediction method and system based on LSTM-CNN joint model

Abstract

The embodiment of the invention discloses a wind power prediction method and system based on an LSTM-CNN joint model. The method comprises the steps that: the power data of N wind turbines are input into N LSTMs, and N first time correlation output values can be obtained; the N first time correlation output values are preset and then input into a two-dimensional matrix, the structure of the two-dimensional matrix is matched with the position information of the wind turbines in a wind field; and spatial correlation data in the two-dimensional matrix is extracted from a CNN model, and the spatial correlation data is processed, so that a one-dimensional model output value is obtained and is adopted as a wind power prediction result of the N wind turbines. The model trained by the method can extract high-level spatial-temporal characteristics from historical power data of the wind turbine generators, so that the purpose of simultaneously predicting the power of the wind turbines at different positions is achieved. With the wind power prediction method and system adopted, the power under of the wind turbines under different wind conditions can be predicted more accurately, so that more accurate power scheduling is realized, and finally the purpose of stable grid connection is achieved.

Description

technical field [0001] The present invention relates to the technical field of wind speed forecasting in wind farms, and more specifically, relates to a wind power forecasting method based on combined forecasting theory and based on LSTM-CNN joint model. Background technique [0002] Wind energy is a renewable and clean energy. In the current global energy crisis and environmental crisis, wind power generation has been widely valued and promoted. [0003] Due to the randomness and uncertainty of wind energy itself, the wind turbines cannot output stably, which increases the difficulty of wind farm control. In large wind farms, the wind captured by wind turbines at different positions in the wind farm is different. There are scheduling schemes (equal distribution method and proportional distribution method) that will lead to performance degradation or even instability, so the power of each fan needs to be predicted. [0004] However, most of the current research focuses on m...

AI Technical Summary

Problems solved by technology

[0003] Due to the randomness and uncertainty of wind energy itself, the wind turbines cannot output stably, which increases the difficulty of controlling the wind farm. In large wind farms, the wind captured by the wind turbines at different positions in the wind farm is different. There are scheduling schemes (equal distribution method and proportional distribution method) that will lead to performance degradation or even instability, so the power of each fan needs to be predicted

Images