Surface meteorological elements and upper-air meteorological elements under typhoon weather conditions were predicted using Pangu-Weather, and the prediction of typhoon tracks was studied combined with the tropical cyclone tracking algorithm. The SWAN model was driven based on the typhoon track prediction results of Pangu-Weather and the predicted wind field data improved by a direct correction method. The calculated significant wave height data were compared with the measured data for verification, thereby evaluating the effectiveness of the model in predicting typhoon tracks and typhoon waves in Guangdong Province and Hainan Island. The results indicate that Pangu-Weather can achieve long-term typhoon track prediction, and the results are superior to those of the high-resolution numerical atmospheric model (ECMWF-HRES), but Pangu-Weather underestimates the typhoon wind field. Combining the typhoon track predicted by Pangu-Weather with the typhoon wind field improved by the direct correction method to drive the wave field can effectively improve the prediction accuracy of typhoon wave heights.