Abstract:Wind speed forecasting provides the upcoming wind information and is important to the safe operation of High-Speed Railway (HSR). However, it remains a challenge due to the stochastic and highly varying characteristics of wind. In this paper, we propose a novel Transformer-based method for short-term wind speed forecasting, named WindTrans. Two major cruxes are addressed. First, the task is performed on fine-grained wind speed gathered from multiple sensors. These data present dynamic intra-series and inter-series correlations, which are hard for previous methods to recover. We advance a Transformer-based deep learning model, which has two distinctive characteristics: (1) a graph encoder, which captures the dynamic spatial correlation among wind speeds at different locations, and (2) a temporal decoder to model long sequence wind speed time series, which is resistant to noise in time series. Second, wind speed patterns gradually evolve in long-term periods, thus deactivating prediction models trained on historical data. To tackle this bottleneck, we put forward an experience replay-based scheme to renew the model regularly. To ensure that the renewed model still dominates historical wind patterns, we store and replay only a small portion of historical data named episodic memory. A simple but efficient strategy is designed to constitute episodic memory and thus relieve the computation burden. Experiments conducted on two real-world datasets demonstrate the superiority of our method over existing approaches. Particularly, WindTrans surpasses state-of-the-art methods by up to 36.7%, 29.3% and 13.3% improvement in MAPE measure for 1 hour ahead prediction on 10-minute, 5-minute, and 1-minute-based tasks, respectively. Furthermore, via our continual learning scheme, the model retains competitive performance with only 6.9% datum stored and retrained on.

Rolling decomposition method in fusion with echo state network for wind speed forecasting

Wind speed forecasting based on variational mode decomposition and improved echo state network

Hybrid model based on VMD decomposition, clustering analysis, long short memory network, ensemble learning and error complementation for short-term wind speed forecasting assisted by Flink platform

Wind Speed Multi-Step Forecasting Using High-Resolution and Multi-Objective Ensemble Model with Error Correction

A combination forecasting model of wind speed based on decomposition

Wind speed prediction model using ensemble empirical mode decomposition, least squares support vector machine and long short-term memory

A Novel Hybrid Model for Short-Term Wind Speed Forecasting Based on Twice Decomposition, PSR, and IMVO-ELM

A Hybrid Wind Speed Prediction Approach Based on Ensemble Empirical Mode Decomposition and BO-LSTM Neural Networks for Digital Twin

An Ensemble Model of Wind Speed Forecasting Based on Variational Mode Decomposition and Bare-Bones Fireworks Algorithm

Hybrid wind speed forecasting model based on multivariate data secondary decomposition approach and deep learning algorithm with attention mechanism

WindTrans: Transformer-Based Wind Speed Forecasting Method for High-Speed Railway

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

Deep learning combined wind speed forecasting with hybrid time series decomposition and multi-objective parameter optimization

Short-Term Wind Speed Forecasting Based on Signal Decomposing Algorithm and Hybrid Linear/Nonlinear Models

Wind Speed Short-Term Prediction Based on Empirical Wavelet Transform, Recurrent Neural Network and Error Correction

A new two-stage decomposition and integrated hybrid model for short-term wind speed prediction

A new decomposition-ensemble strategy fusion with correntropy optimization learning algorithms for short-term wind speed prediction

Multi-step wind speed prediction based on LSSVM combined with ESMD and fractional-order beetle swarm optimization

A combined model for short-term wind speed forecasting based on empirical mode decomposition, feature selection, support vector regression and cross-validated lasso

A Novel Hybrid Forecasting System Based on Data Augmentation and Deep Learning Neural Network for Short-Term Wind Speed Forecasting

The univariate model for long-term wind speed forecasting based on wavelet soft threshold denoising and improved Autoformer