Ultra-Short-Term Wind Power Prediction Based on eEEMD-LSTM
Abstract
:1. Introduction
2. Wind Power Time Series Decomposition Method Based on eEEMD
2.1. EMD Method and EEMD Method
Algorithm 1 EEMD algorithm |
Initialization: , , |
while to do |
Add Gaussian noise to the original . |
while to do |
Find the local maximum and minimum values of the input. |
Calculate the average of the two envelopes. |
Calculate the difference according to and . |
if the stop condition is met then |
end while |
Calculate the average of . |
end while |
return |
2.2. Wind Power Decomposition with eEEMD
Algorithm 2 eEEMD algorithm |
Initialization: , |
while to do |
if then |
Re-standardize to a non-negative interval. |
else |
Calculate the proportion of the sample of . |
if then |
else |
Calculate difference coefficient . |
end while |
Reconstruct to according to specific merger rules. |
return , , |
3. eEEMD-LSTM Method Based on Sub-Model Differential Training
3.1. LSTM Network
Algorithm 3 LSTM algorithm |
Initialization: Initialize the hyper parameters, including Uf, Ui, Uc, bf, bi, bo, and bc, set ho = 0, Co = 0. |
For to do |
Calculate forget gate ft, decide how much information to forget. |
Calculate input gate it, determine the information to be stored. |
Calculate output gate Ot, filter information. |
Calculate temporary state C′t. |
Calculate current state Ct, update cell state. |
Calculate output ht. |
return |
3.2. Differential Training Based on Sequence Mean
3.3. Wind Power Prediction Based on eEEMD-LSTM
4. Experiment and Result Analysis
4.1. Data Preprocessing
4.1.1. Data Selection
4.1.2. Evaluation Metrics
4.2. Predictive Performance Analysis
4.2.1. Efficiency Analysis
4.2.2. Accuracy Analysis
5. Conclusions
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
References
- International Energy Agency. World Energy Outlook 2023. Available online: https://www.iea.org/reports/world-energy-outlook-2023 (accessed on 30 October 2023).
- Nadeem, F.; Aftab, M.A.; Hussain, S.M.S.; Ali, I.; Tiwari, P.K.; Goswami, A.K.; Ustun, T.S. Virtual Power Plant Management in Smart Grids with XMPP Based IEC 61850 Communication. Energies 2019, 12, 2398. [Google Scholar] [CrossRef]
- Veers, P.; Dykes, K.; Lantz, E.; Barth, S.; Bottasso, C.L.; Carlson, O.; Clifton, A.; Green, J.; Green, P.; Holttinen, H.; et al. Grand Challenges in the Science of Wind Energy. Science 2019, 366, eaau2027. [Google Scholar] [CrossRef] [PubMed]
- Hoyos-Santillan, J.; Miranda, A.; Lara, A.; Rojas, M.; Sepulveda-Jauregui, A. Protecting Patagonian Peatlands in Chile. Science 2019, 366, 1207–1208. [Google Scholar] [CrossRef]
- Xing, Y.; Chen, Y.; Huang, S.; Wang, P.; Xiang, Y. Research on Dam Deformation Prediction Model Based on Optimized SVM. Processes 2022, 10, 1842. [Google Scholar] [CrossRef]
- Shan, Y.; Chen, S.; Zhong, Q. Rapid Prediction of Landslide Dam Stability Using the Logistic Regression Method. Landslides 2020, 17, 2931–2956. [Google Scholar] [CrossRef]
- Lin, Z.; Liu, X.; Collu, M. Wind Power Prediction Based on High-Frequency SCADA Data along with Isolation Forest and Deep Learning Neural Networks. Int. J. Electr. Power Energy Syst. 2020, 118, 105835. [Google Scholar] [CrossRef]
- Chen, X.; Zhang, X.; Dong, M.; Huang, L.; Guo, Y.; He, S. Deep Learning-Based Prediction of Wind Power for Multi-Turbines in a Wind Farm. Front. Energy Res. 2021, 9, 723775. [Google Scholar] [CrossRef]
- Gu, B.; Zhang, T.; Meng, H.; Zhang, J. Short-Term Forecasting and Uncertainty Analysis of Wind Power Based on Long Short-Term Memory, Cloud Model and Non-Parametric Kernel Density Estimation. Renew. Energy 2021, 164, 687–708. [Google Scholar] [CrossRef]
- Han, L.; Jing, H.; Zhang, R.; Gao, Z. Wind Power Forecast Based on Improved Long Short Term Memory Network. Energy 2019, 189, 116300. [Google Scholar] [CrossRef]
- Huang, J.; Li, C.; Huang, Z.; Liu, P.X. A Decomposition-Based Approximate Entropy Cooperation Long Short-Term Memory Ensemble Model for Short-Term Load Forecasting. Electr. Eng. 2022, 104, 1515–1525. [Google Scholar] [CrossRef]
- Shahid, F.; Zameer, A.; Muneeb, M. A Novel Genetic LSTM Model for Wind Power Forecast. Energy 2021, 223, 120069. [Google Scholar] [CrossRef]
- Yang, H.; Jiang, Z.; Lu, H. A Hybrid Wind Speed Forecasting System Based on a ‘Decomposition and Ensemble’ Strategy and Fuzzy Time Series. Energies 2017, 10, 1422. [Google Scholar] [CrossRef]
- Zhao, N.; Li, R. EMD Method Applied to Identification of Logging Sequence Strata. Acta Geophys. 2015, 63, 1256–1275. [Google Scholar] [CrossRef]
- Ambark, A.S.A.; Ismail, M.T.; Al-Jawarneh, A.S.; Karim, S.A.A. Elastic Net Penalized Quantile Regression Model and Empirical Mode Decomposition for Improving the Accuracy of the Model Selection. IEEE Access 2023, 11, 26152–26162. [Google Scholar] [CrossRef]
- Michelson, E.A.; Cohen, L.; Dankner, R.E.; Kulczycki, A.; Kulczycki, A. Eosinophilia and Pulmonary Dysfunction during Cuprophan Hemodialysis. Kidney Int. 1983, 24, 246–249. [Google Scholar] [CrossRef] [PubMed]
- Wang, J.; Du, G.; Zhu, Z.; Shen, C.; He, Q. Fault Diagnosis of Rotating Machines Based on the EMD Manifold. Mech. Syst. Signal Process. 2020, 135, 106443. [Google Scholar] [CrossRef]
- Yang, Y.; Yang, Y. Hybrid Method for Short-Term Time Series Forecasting Based on EEMD. IEEE Access 2020, 8, 61915–61928. [Google Scholar] [CrossRef]
- Zhang, G.; Zhou, H.; Wang, C.; Xue, H.; Wang, J.; Wan, H. Forecasting Time Series Albedo Using NARnet Based on EEMD Decomposition. IEEE Trans. Geosci. Remote Sens. 2020, 58, 3544–3557. [Google Scholar] [CrossRef]
- Meng, A.; Chen, S.; Ou, Z.; Ding, W.; Zhou, H.; Fan, J.; Yin, H. A Hybrid Deep Learning Architecture for Wind Power Prediction Based on Bi-Attention Mechanism and Crisscross Optimization. Energy 2022, 238, 121795. [Google Scholar] [CrossRef]
- Huang, J.; Niu, G.; Guan, H.; Song, S. Ultra-Short-Term Wind Power Prediction Based on LSTM with Loss Shrinkage Adam. Energies 2023, 16, 3789. [Google Scholar] [CrossRef]
- Cho, K.; van Merrienboer, B.; Bahdanau, D.; Bengio, Y. On the Properties of Neural Machine Translation: Encoder–Decoder Approaches. In Proceedings of the 8th Workshop on Syntax, Semantics and Structure in Statistical Translation, SSST 2014, Doha, Qatar, 25 October 2014; Association for Computational Linguistics (ACL): Doha, Qatar, 2014; pp. 103–111. [Google Scholar]
Data | The Number of IMFs with EEMD | The Number of IMFs with eEEMD |
---|---|---|
Data 1 | 12 | 6 |
Data 2 | 12 | 5 |
Data 3 | 12 | 5 |
Parameter | Value |
---|---|
Epochs | 600 |
Learning rate | 0.001 |
Training error targets | 0.0001 |
Hidden layers | 2 |
Number of neurons per layer | 4 |
Optimization functions | Adam |
Batch_size | 1024 |
Kmax | 10 |
Data | Model | RMSE | MAE | MAPE |
---|---|---|---|---|
Data 1 | BR | 0.1658 | 0.1177 | 43.71% |
EEMD-BR | 0.2041 | 0.1433 | 15.27% | |
eEEMD-BR | 0.09290 | 0.09125 | 10.33% | |
SVR | 0.06804 | 0.05575 | 32.71% | |
EEMD-SVR | 0.05709 | 0.04668 | 5.425% | |
eEEMD-SVR | 0.05708 | 0.04670 | 5.427% | |
GRU | 0.04316 | 0.02961 | 13.78% | |
EEMD-GRU | 0.04711 | 0.03540 | 15.71% | |
eEEMD-GRU | 0.03491 | 0.02106 | 15.98% | |
LSTM | 0.04378 | 0.03042 | 22.62% | |
EEMD-LSTM | 0.03715 | 0.02175 | 17.42% | |
eEEMD-LSTM | 0.02444 | 0.01797 | 8.343% | |
Data 2 | BR | 0.1390 | 0.1045 | 45.64% |
EEMD-BR | 0.1736 | 0.1323 | 14.86% | |
eEEMD-BR | 0.04737 | 0.03223 | 3.506% | |
SVR | 0.06408 | 0.04815 | 26.67% | |
EEMD-SVR | 0.05473 | 0.04259 | 4.859% | |
eEEMD-SVR | 0.05769 | 0.04409 | 4.964% | |
GRU | 0.05191 | 0.03551 | 8.313% | |
EEMD-GRU | 0.04775 | 0.03422 | 3.764% | |
eEEMD-GRU | 0.04161 | 0.03151 | 3.563% | |
LSTM | 0.05540 | 0.03667 | 11.54% | |
EEMD-LSTM | 0.03661 | 0.02870 | 3.217% | |
eEEMD-LSTM | 0.04078 | 0.02856 | 3.200% | |
Data 3 | BR | 0.1452 | 0.1098 | 8.850% |
EEMD-BR | 0.06111 | 0.04591 | 4.240% | |
eEEMD-BR | 0.04977 | 0.03767 | 2.560% | |
SVR | 0.06570 | 0.05555 | 36.34% | |
EEMD-SVR | 0.05504 | 0.04594 | 5.452% | |
eEEMD-SVR | 0.05283 | 0.04366 | 5.139% | |
GRU | 0.1416 | 0.2515 | 13.46% | |
EEMD-GRU | 0.07479 | 0.05786 | 6.287% | |
eEEMD-GRU | 0.04959 | 0.03255 | 5.655% | |
LSTM | 0.09904 | 0.07579 | 11.50% | |
EEMD-LSTM | 0.05270 | 0.04022 | 3.953% | |
eEEMD-LSTM | 0.04303 | 0.03112 | 3.345% |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Huang, J.; Zhang, W.; Qin, J.; Song, S. Ultra-Short-Term Wind Power Prediction Based on eEEMD-LSTM. Energies 2024, 17, 251. https://doi.org/10.3390/en17010251
Huang J, Zhang W, Qin J, Song S. Ultra-Short-Term Wind Power Prediction Based on eEEMD-LSTM. Energies. 2024; 17(1):251. https://doi.org/10.3390/en17010251
Chicago/Turabian StyleHuang, Jingtao, Weina Zhang, Jin Qin, and Shuzhong Song. 2024. "Ultra-Short-Term Wind Power Prediction Based on eEEMD-LSTM" Energies 17, no. 1: 251. https://doi.org/10.3390/en17010251
APA StyleHuang, J., Zhang, W., Qin, J., & Song, S. (2024). Ultra-Short-Term Wind Power Prediction Based on eEEMD-LSTM. Energies, 17(1), 251. https://doi.org/10.3390/en17010251