A Hybrid Model Based on Complete Ensemble Empirical Mode Decomposition With Adaptive Noise, GRU Network and Whale Optimization Algorithm for Wind Power Prediction
To ensure the safe and stable operation of power systems, accurate prediction of wind power generation is particularly important. However, due to the randomness, fluctuation, and intermittency of wind energy, as well as the challenges in determining the hyperparameters of the gated recurrent unit (G...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2023-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10155125/ |
_version_ | 1797794191628042240 |
---|---|
author | Andi Sheng Lewei Xie Yixiang Zhou Zhen Wang Yuechao Liu |
author_facet | Andi Sheng Lewei Xie Yixiang Zhou Zhen Wang Yuechao Liu |
author_sort | Andi Sheng |
collection | DOAJ |
description | To ensure the safe and stable operation of power systems, accurate prediction of wind power generation is particularly important. However, due to the randomness, fluctuation, and intermittency of wind energy, as well as the challenges in determining the hyperparameters of the gated recurrent unit (GRU) network, this paper proposes an innovative “decomposition-optimization-reconstruction” prediction method. To enhance the accuracy of wind power prediction, a sophisticated wind power hybrid prediction model based on the GRU network, combined with complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and the whale optimization algorithm (WOA) has been proposed in this paper. Among the efficient decomposition algorithms are successive variational mode decomposition (SVMD) and CEEMDAN. Taking into account adaptability issues, in the data decomposition stage, the CEEMDAN method is chosen to decompose the wind power time series into different sub-sequence components. This approach allows for a better representation of the fluctuation characteristics of each sub-sequence at various time scales. In the prediction optimization stage, the GRU prediction method is used to predict each sub-sequence component and the WOA algorithm is combined to optimize the hyperparameters of each GRU. This approach demonstrates significant advantages in improving wind power prediction accuracy, enhancing generalization capability, and strengthening adaptability. In the prediction reconstruction stage, the prediction results of each sub-sequence component are superimposed to obtain the final wind power prediction value. Finally, the model is verified using actual wind power data from a power plant in the northwest, and the simulation results show that the wind power prediction model based on CEEMDAN-WOA-GRU has significant advantages in prediction accuracy and stability compared to other models. This model provides strong support for optimizing wind energy grid integration and ensuring a stable power supply. |
first_indexed | 2024-03-13T02:58:09Z |
format | Article |
id | doaj.art-f99cd00e02384a7eaef1329ca204a6b6 |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-03-13T02:58:09Z |
publishDate | 2023-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-f99cd00e02384a7eaef1329ca204a6b62023-06-27T23:00:56ZengIEEEIEEE Access2169-35362023-01-0111628406285410.1109/ACCESS.2023.328731910155125A Hybrid Model Based on Complete Ensemble Empirical Mode Decomposition With Adaptive Noise, GRU Network and Whale Optimization Algorithm for Wind Power PredictionAndi Sheng0https://orcid.org/0009-0001-6601-9578Lewei Xie1https://orcid.org/0009-0001-1707-2866Yixiang Zhou2https://orcid.org/0009-0004-0894-2040Zhen Wang3https://orcid.org/0009-0000-8832-2407Yuechao Liu4https://orcid.org/0000-0002-1328-196XComputer Department, North China Electric Power University, Baoding, Hebei, ChinaCollege of Science, Huazhong Agricultural University, Wuhan, Hubei, ChinaSchool of Health Administration, Anhui Medical University, Hefei, Anhui, ChinaHotel Management Department, Zhejiang Yuexiu University, Shaoxing, Zhejiang, ChinaDepartment of Mathematics and Physics, North China Electric Power University, Baoding, Hebei, ChinaTo ensure the safe and stable operation of power systems, accurate prediction of wind power generation is particularly important. However, due to the randomness, fluctuation, and intermittency of wind energy, as well as the challenges in determining the hyperparameters of the gated recurrent unit (GRU) network, this paper proposes an innovative “decomposition-optimization-reconstruction” prediction method. To enhance the accuracy of wind power prediction, a sophisticated wind power hybrid prediction model based on the GRU network, combined with complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and the whale optimization algorithm (WOA) has been proposed in this paper. Among the efficient decomposition algorithms are successive variational mode decomposition (SVMD) and CEEMDAN. Taking into account adaptability issues, in the data decomposition stage, the CEEMDAN method is chosen to decompose the wind power time series into different sub-sequence components. This approach allows for a better representation of the fluctuation characteristics of each sub-sequence at various time scales. In the prediction optimization stage, the GRU prediction method is used to predict each sub-sequence component and the WOA algorithm is combined to optimize the hyperparameters of each GRU. This approach demonstrates significant advantages in improving wind power prediction accuracy, enhancing generalization capability, and strengthening adaptability. In the prediction reconstruction stage, the prediction results of each sub-sequence component are superimposed to obtain the final wind power prediction value. Finally, the model is verified using actual wind power data from a power plant in the northwest, and the simulation results show that the wind power prediction model based on CEEMDAN-WOA-GRU has significant advantages in prediction accuracy and stability compared to other models. This model provides strong support for optimizing wind energy grid integration and ensuring a stable power supply.https://ieeexplore.ieee.org/document/10155125/Decomposition-optimization-reconstructiongated recurrent unitcomplete ensemble empirical mode decomposition with adaptive noisewhale optimization algorithmwind power prediction |
spellingShingle | Andi Sheng Lewei Xie Yixiang Zhou Zhen Wang Yuechao Liu A Hybrid Model Based on Complete Ensemble Empirical Mode Decomposition With Adaptive Noise, GRU Network and Whale Optimization Algorithm for Wind Power Prediction IEEE Access Decomposition-optimization-reconstruction gated recurrent unit complete ensemble empirical mode decomposition with adaptive noise whale optimization algorithm wind power prediction |
title | A Hybrid Model Based on Complete Ensemble Empirical Mode Decomposition With Adaptive Noise, GRU Network and Whale Optimization Algorithm for Wind Power Prediction |
title_full | A Hybrid Model Based on Complete Ensemble Empirical Mode Decomposition With Adaptive Noise, GRU Network and Whale Optimization Algorithm for Wind Power Prediction |
title_fullStr | A Hybrid Model Based on Complete Ensemble Empirical Mode Decomposition With Adaptive Noise, GRU Network and Whale Optimization Algorithm for Wind Power Prediction |
title_full_unstemmed | A Hybrid Model Based on Complete Ensemble Empirical Mode Decomposition With Adaptive Noise, GRU Network and Whale Optimization Algorithm for Wind Power Prediction |
title_short | A Hybrid Model Based on Complete Ensemble Empirical Mode Decomposition With Adaptive Noise, GRU Network and Whale Optimization Algorithm for Wind Power Prediction |
title_sort | hybrid model based on complete ensemble empirical mode decomposition with adaptive noise gru network and whale optimization algorithm for wind power prediction |
topic | Decomposition-optimization-reconstruction gated recurrent unit complete ensemble empirical mode decomposition with adaptive noise whale optimization algorithm wind power prediction |
url | https://ieeexplore.ieee.org/document/10155125/ |
work_keys_str_mv | AT andisheng ahybridmodelbasedoncompleteensembleempiricalmodedecompositionwithadaptivenoisegrunetworkandwhaleoptimizationalgorithmforwindpowerprediction AT leweixie ahybridmodelbasedoncompleteensembleempiricalmodedecompositionwithadaptivenoisegrunetworkandwhaleoptimizationalgorithmforwindpowerprediction AT yixiangzhou ahybridmodelbasedoncompleteensembleempiricalmodedecompositionwithadaptivenoisegrunetworkandwhaleoptimizationalgorithmforwindpowerprediction AT zhenwang ahybridmodelbasedoncompleteensembleempiricalmodedecompositionwithadaptivenoisegrunetworkandwhaleoptimizationalgorithmforwindpowerprediction AT yuechaoliu ahybridmodelbasedoncompleteensembleempiricalmodedecompositionwithadaptivenoisegrunetworkandwhaleoptimizationalgorithmforwindpowerprediction AT andisheng hybridmodelbasedoncompleteensembleempiricalmodedecompositionwithadaptivenoisegrunetworkandwhaleoptimizationalgorithmforwindpowerprediction AT leweixie hybridmodelbasedoncompleteensembleempiricalmodedecompositionwithadaptivenoisegrunetworkandwhaleoptimizationalgorithmforwindpowerprediction AT yixiangzhou hybridmodelbasedoncompleteensembleempiricalmodedecompositionwithadaptivenoisegrunetworkandwhaleoptimizationalgorithmforwindpowerprediction AT zhenwang hybridmodelbasedoncompleteensembleempiricalmodedecompositionwithadaptivenoisegrunetworkandwhaleoptimizationalgorithmforwindpowerprediction AT yuechaoliu hybridmodelbasedoncompleteensembleempiricalmodedecompositionwithadaptivenoisegrunetworkandwhaleoptimizationalgorithmforwindpowerprediction |