A gridded observation data set interpolated with daily observation data by the National Climate Center is selected to evaluate the capability of a regional climate model COSMO- CLM(CCLM)for simulating extreme precipitation in China. The paper mainly applies the Intensity-Area-Duration (IAD) method to identify the extreme precipitation events with certain intensity in certain area at the given time scale, and project the trend of extreme precipitation events in the future (2016-2050) based on the CCLM output. The research results show that:1) CCLM can well capture the distribution and trend of the extreme precipitation in China. 2) The extreme precipitation events in China will increase during 2016- 2050. The increasing trend is more significant and the events are more intensive in RCP 8.5 scenario than in RCP 2.6 and RCP 4.5.3) Extreme precipitation events beyond the maximum intensity/ coverage of baseline period are likely to occur in all emission scenarios during 2016-2050. The events with larger area will mainly distribute in North China and Northeast China. The events with higher intensity will mainly distribute in Southwest China and South China.