An Integrated CEEMDAN to Optimize Deep Long Short-Term Memory Model for Wind Speed Forecasting DOI Creative Commons

Yingying He,

Likai Zhang, Tengda Guan

и другие.

Energies, Год журнала: 2024, Номер 17(18), С. 4615 - 4615

Опубликована: Сен. 14, 2024

Accurate wind speed forecasting is crucial for the efficient operation of renewable energy platforms, such as turbines, it facilitates more effective management power output and maintains grid reliability stability. However, inherent variability intermittency present significant challenges achieving precise forecasts. To address these challenges, this study proposes a novel method based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) deep learning-based Long Short-Term Memory (LSTM) network forecasting. In proposed method, CEEMDAN utilized to decompose original signal into different modes capture multiscale temporal properties patterns speeds. Subsequently, LSTM employed predict each subseries derived from process. These individual predictions are then combined generate overall final forecast. The validated using real-world data Austria Almeria. Experimental results indicate that achieves minimal mean absolute percentage errors 0.3285 0.1455, outperforming other popular models across multiple performance criteria.

Язык: Английский

Informer learning framework based on secondary decomposition for multi-step forecast of ultra-short term wind speed DOI

Zihao Jin,

Xiaomengting Fu,

Ling Xiang

и другие.

Engineering Applications of Artificial Intelligence, Год журнала: 2024, Номер 139, С. 109702 - 109702

Опубликована: Ноя. 22, 2024

Язык: Английский

Процитировано

4
Multi-step ahead wind power forecasting based on multi-feature wavelet decomposition and convolution-gated recurrent unit model DOI

S.N. Shringi,

Lalit Mohan Saini, S. K. Aggarwal

и другие.

Electrical Engineering, Год журнала: 2025, Номер unknown

Опубликована: Март 12, 2025

Язык: Английский

Процитировано

0
A Multi-Channel Spatiotemporal Segnet Model for Short Term Wind Power Prediction with Sequence Decomposition and Reconstruction DOI

Xingdou Liu,

Liang Zou, Li Zhang

и другие.

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

0
Prediction method for instrument transformer measurement error: Adaptive decomposition and hybrid deep learning models DOI

Zhenhua Li,

Jiuxi Cui,

Heping Lu

и другие.

Measurement, Год журнала: 2025, Номер unknown, С. 117592 - 117592

Опубликована: Май 1, 2025

Язык: Английский

Процитировано

0
Noise Reduction Method for Wind Turbine Gearbox Vibration Signals Based on CVMD-DRDSAE DOI
Yao Jinbao, B. Yue, Yizhu Wang

и другие.

Measurement Science and Technology, Год журнала: 2024, Номер 35(11), С. 116146 - 116146

Опубликована: Авг. 22, 2024

Abstract Wind turbine gearbox fault feature extraction is difficult due to strong background noise. To address this issue, a noise reduction method combining comprehensive learning particle swarm optimization-variational mode decomposition (CLPSO-VMD) and deep residual denoising self-attention autoencoder (DRDSAE) proposed. Firstly, the proposed CLPSO-VMD algorithm used decompose noisy wind vibration signals. Subsequently, intrinsic functions highly correlated with original signals are selected through Spearman correlation coefficient utilized for signal reconstruction, thereby filtering out high-frequency outside frequency band in domain characterization. Secondly, improved DRDSAE learn latent representations of data first-level denoised signal, further reducing within while retaining important features. Finally, envelope spectrum highlights weak signal. Experimental results demonstrate effectiveness under

Язык: Английский

Процитировано

2
An Integrated CEEMDAN to Optimize Deep Long Short-Term Memory Model for Wind Speed Forecasting DOI Creative Commons

Yingying He,

Likai Zhang, Tengda Guan

и другие.

Energies, Год журнала: 2024, Номер 17(18), С. 4615 - 4615

Опубликована: Сен. 14, 2024

Accurate wind speed forecasting is crucial for the efficient operation of renewable energy platforms, such as turbines, it facilitates more effective management power output and maintains grid reliability stability. However, inherent variability intermittency present significant challenges achieving precise forecasts. To address these challenges, this study proposes a novel method based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) deep learning-based Long Short-Term Memory (LSTM) network forecasting. In proposed method, CEEMDAN utilized to decompose original signal into different modes capture multiscale temporal properties patterns speeds. Subsequently, LSTM employed predict each subseries derived from process. These individual predictions are then combined generate overall final forecast. The validated using real-world data Austria Almeria. Experimental results indicate that achieves minimal mean absolute percentage errors 0.3285 0.1455, outperforming other popular models across multiple performance criteria.

Язык: Английский

Процитировано

0