<p>Accurate short-term wind speed prediction is of great significance for improving wind energy utilization, optimizing wind farm operation, and ensuring the stable operation of the power system. A short-term wind speed prediction model based on dimension reduction and deep learning models is proposed. Firstly, kernel principal component analysis is used to determine the principal components of short-term wind speed data, reducing data dimensions and complexity, and improving data processing efficiency and accuracy. The squeeze-and-excitation attention mechanism is introduced into convolutional neural network to improve the effectiveness of feature extraction. Then, based on the input data of the long short-term memory network after feature fusion, an improved Whale optimization algorithm is proposed to optimize the learning rate, regularization coefficient, and number of hidden layer nodes of the long short-term memory network. Finally, the short-term wind speed is predicted through the optimized long short-term memory network. The comparison results show that the proposed model has high prediction accuracy, and the training time can meet the actual needs, which has a certain application prospect.</p>