PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 1, 2026Applied Sciences0 citationsOpen Access

A Hybrid LSTM Framework for Short-Term Regional Wind Speed Forecasting Based on PCA and SSA-Optimized VMD

View Full Paper
HLH. D. LiZMZhengzheng MaLCLiang Chen

Key Points

  • The aim is to develop a hybrid framework for accurate short-term regional wind speed forecasting by addressing data non-stationarity.
  • Utilized Principal Component Analysis (PCA) to extract spatial features from a 9 × 9 wind field grid.
  • Employed Variational Mode Decomposition (VMD) optimized by the Sparrow Search Algorithm (SSA) to decompose wind data components.
  • Integrated dominant components into a Long Short-Term Memory (LSTM) network for prediction.
  • Achieved an average coefficient of determination (R2) of 0.41 in the first forecasting step.
  • Maintained R2 of 0.39 in the third forecasting step, significantly mitigating error accumulation.
  • Compared to ConvLSTM, which dropped from R2 0.70 to 0.24, this method demonstrated greater stability and lower computational overhead.

Abstract

Accurate regional wind speed forecasting is critical yet challenging due to inherent spatiotemporal correlations and data non-stationarity. This paper proposes a hybrid framework combining Principal Component Analysis (PCA), Variational Mode Decomposition (VMD), and Long Short-Term Memory (LSTM) networks. First, PCA extracts dominant spatial features from a regional wind field (9 × 9 grid), retaining 99.5% of the information to reduce redundancy. Next, an adaptive VMD strategy, optimized by the Sparrow Search Algorithm (SSA), decomposes these components to mitigate temporal non-stationarity. High-correlation sub-signals are then fed into the LSTM predictor. Experimental results demonstrate that the framework achieves an average coefficient of determination (R2) of approximately 0.41 in the first forecasting step. Crucially, it significantly mitigates error accumulation in multi-step forecasting, maintaining a stable R2 of 0.39 in the third step. Conversely, complex spatiotemporal models like ConvLSTM achieve high initial accuracy but suffer severe degradation (R2 dropping from 0.70 to 0.24) alongside significantly higher computational overhead. The proposed strategy effectively prevents overfitting to high-frequency noise, ensuring a computationally efficient and robust solution for multi-step regional wind forecasting.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Li et al. (2026) studied this question.

synapsesocial.com/papers/69f443e8967e944ac55670b2https://doi.org/10.3390/app16094225
Ask AI
Helpful
Bookmark
Share
View Full Paper