Accurate wind speed prediction is significant for wind farm development and utilization of wind energy. In order to improve the prediction accuracy of short-term wind speed, a combined prediction model with variable weight of short-term wind speed based on Ensemble Empirical Mode Decomposition (EEMD), Least Absolute Shrinkage and Selection Operator (LASSO), Genetic Algorithm (GA), General Regression Neural Network (GRNN) and long-term and short-term memory is proposed. First, the ensemble empirical mode decomposition technique is used to decompose the original wind speed time series into multiple sub-sequences. Then, using the Least Absolute Shrinkage and Selection Operator (LASSO) algorithm, the historical data of each sub-sequence are filtered, and representative variables are extracted as prediction inputs. Finally, using the global optimization ability of a GA, the weight coefficients of the combined prediction model composed of GRNN and LSTM are adaptively solved by moving samples, and the final prediction results are obtained by weighting. The simulation results show that the proposed variable weight combination model has higher prediction accuracy than a single model and a traditional combination model, and has superiority in wind speed prediction. This work is supported by National Natural Science Foundation of China (No. 61973117).