Abstract: The strongest precipitation days of three north-moving typhoons affecting Northeast China in 2020 were selected as examples, the best method was used to determine the threshold and identify the precipitation subjects and individuals, and the SAL (structure, amplitude and location) test was carried out to analyze and evaluate the effects of 12-36 h forecast of the China Meteorological Administration's Global Generalized Forecast System (CMA-GFS), European Center for Medium-Range Weather Prediction (EC) model and China Meteorological Administration Mesoscale Numerical Weather Prediction System (CMA-MESO) model on typhoon precipitation in Northeast China.The results show that the forecasting effect of CMA-GFS for the precipitation process of typhoon Bavi is the best.For Typhoon Mavsak, the precipitation process intensity forecasted by the CMA-MESO model is stronger than those of the other two models.The structure forecast by the CMA-GFS model is better, and the precipitation position forecasted by the EC model is the best.The CMA-GFS and EC models are better at predicting the intensity of the precipitation process of Typhoon Haishen, and the EC model is the worst at predicting the structure but the best at predicting the precipitation location.We test and analyze three typhoon models affecting Northeast China in 2020 to explore preliminarily the forecast tendency of each model in typhoon forecast.In the forecast of typhoon precipitation intensity in Northeast China in the future, especially when similar typhoons occur, selective reference and purposefully subjective correction can be operated according to the model forecast characteristics.