Wind power prediction of CNN-LSTM based on improved ALA

This research aims to address the challenges in enhancing the accuracy of wind energy prediction and the stability of the system by improving the Artificial Lemming Algorithm (ALA) to optimize the CNN-LSTM architecture. The study leverages the improved ALA optimization algorithm to conduct comprehensive hyperparameter tuning, which is designed to strike a balance between the exploration and exploitation phases, thereby enhancing the training outcomes. Through experimental comparisons, the model has significantly improved the Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) metrics on the test dataset, and also boosted the R 2 coefficient. These enhancements have augmented the model’s ability to capture complex patterns in wind energy datasets. The results confirm its superior predictive performance in the application of renewable energy prediction.