Wind power prediction method combining WRF-PALM coupling model and CNN-LSTM algorithm and related device

By combining the WRF-PALM coupling model and the CNN-LSTM algorithm, the accuracy problem of the prior art stroke power prediction in complex mountainous conditions is solved, and higher wind field details capture and wind power prediction accuracy are achieved.

CN119578655APending Publication Date: 2025-03-07DATANG HYDROPOWER SCI & TECH RES INST CO LTD +2
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411781230.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-05
Publication Date
2025-03-07

AI Technical Summary

Technical Problem

The accuracy of wind power prediction in the prior art is not high, especially in complex mountainous conditions, it is difficult to accurately capture local micrometeorological details and rapid changes in wind fields.

Method used

Combining the WRF-PALM coupling model and the CNN-LSTM algorithm, by obtaining historical meteorological data and terrain data, WRF and PALM models are constructed, meteorological element prediction and wind field simulation are carried out, and the training set is then constructed using the pre-processed data, the CNN-LSTM model is trained, and the prediction results are finally input into the CNN-LSTM model for wind power prediction.

Benefits of technology

It improves the ability to capture wind farm details, reduces the prediction error of meteorological elements under complex terrain conditions, and improves the accuracy of wind power prediction, especially when the data volume is insufficient or the meteorological changes are severe.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119578655A_ABST
    Figure CN119578655A_ABST
Patent Text Reader

Abstract

The invention provides a wind power prediction method combining a WRF-PALM coupling model and a CNN-LSTM algorithm and a related device, and belongs to the technical field of wind power prediction. The method comprises the following steps: constructing a PALM model, and inputting an obtained dynamic input file and a constructed static input file into the PALM model to obtain a wind field and meteorological element simulation result; constructing a training set, a verification set and a test set by using wind turbine generator information, the preprocessed historical fan output power data and the preprocessed wind field and meteorological element simulation result; constructing a CNN-LSTM model, training the CNN-LSTM model based on the constructed training set, verifying the CNN-LSTM model based on the constructed verification set, and testing the training result of the CNN-LSTM model based on the constructed test set to obtain a trained CNN-LSTM model; and inputting a wind field and meteorological element prediction result at a future to-be-predicted moment into the trained CNN-LSTM model to carry out wind power prediction so as to obtain a wind power prediction result. According to the invention, the problem of low accuracy of wind power prediction is solved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Cited By

  • AI-NWP three-dimensional closed-loop bidirectional dynamic feedback coupling method, system and program product for extreme rainfall event area simulation

    CN120317160A

  • An AI-NWP three-dimensional closed-loop bidirectional dynamic feedback coupling method, system and program product for regional simulation of extreme precipitation events

    CN120317160B