PSO-LSTM optimally weighted short-term wind speed prediction based on meteorological factor screening

Accurate wind speed forecasting is essential for weather prediction and wind energy planning. This paper employs fast correlation filtering and frequency domain coherence to select highly correlated meteorological factor for a multidimensional feature set. A PSO-based approach is introduced for parameter optimization of an LSTM-based wind speed prediction model, automating and iterating the search for the best parameters. The model integrates various temporal features of wind speed to enhance prediction accuracy. The methodology is validated using data from a Shanghai weather station, with results compared against measured data and other models to assess accuracy and generalization.