<p>For improving the forecast precision of short-term wind speed, this paper proposes a short-term wind speed prediction model based on long short-term memory network with feature extraction. Firstly, for reducing the uncertainty and complexity of short-term wind speed, data cluster is carried out by K-means clustering algorithm. Then, for each clustered sample set, a convolutional neural network is used to further characterize the short-term wind speed, making the features of the wind speed data more regular. Finally, the short-term wind speed data after feature extraction is predicted using long short-term memory network. Meanwhile, 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. The accuracy of the prediction model is tested by using the actual short-term wind speed data with sampling time of 5 min and 30 min. The comparative analysis of relevant performance indicators shows that compared with other models, the established prediction model has good prediction performance and can well reflect the characteristics of short-term wind speed changes. The designed model is effective in improving the accuracy of short-term wind speed prediction.</p>