From March to May every year, China usually suffers frequent cold fronts. Becaus e of active cold fronts, during these 3 months, weather in China is characterize d by sharp temperatures fluctuations and frequent dusty weather conditions. In o rder to examine the forecast effect of numerical model and accumulate a wealth o f forecasting experience, the 96hour forecasting products of T213 model were e xamined and compared with those of ECMWF model and Japan model. The result is th at T213, ECMWF and Japan models all work relatively well in midrange forecast. The ECMWF model shows a good ability in adjusting macroscale circulation and forecasting the main body of subtropical highpressure system, while the Japan model is more effective in forecasting temperatures at 850hpa than other 2 model s. In addition, we chose a dusty weather process which ever occurred in 810 Ma y , 2007 as a case study. Through analyzing this case, we found that T213 and E CMWF models are more effective than the Japan model in midrange forecast of th e strong surface winds, which caused this dusty weather process.