Abstract: Fine-scale quantitative precipitation forecast is a key issue and challenge in weather forecasting services. In this study, based on hourly precipitation from the 1 km× 1 km resolution Southwest China WRF-based Intelligent Numeric Grid forecast System (SWC-WINGS), a fusion-corrected experiment was conducted using time lag and probability matching methods. The fusion-corrected forecast of hourly pre⁃ cipitation was then verified utilizing the CMA Multi-source Precipitation Analysis System (CMAPS) three-source merged precipitation obser⁃ vation grid data from 1 July to 31 August 2022 in Sichuan. Finally, the fusion-corrected method was applied to a short-term heavy precipita⁃ tion process over the western Sichuan Basin. The results show that: (1) Compared with the model precipitation forecasts, the time-lagged en⁃ semble forecast was over-optimistic for small-scale precipitation and over-conservative for large-scale precipitation. (2) However, the fu⁃ sion-corrected method by time lag and probability matching methods overcame the above difficulties and showed significant improvement in the TS score, particularly in 1~2 h nowcast time. The TS score for hourly precipitation exceeding 0.1 mm, 5 mm, 10 mm, and 20 mm were in⁃ creased on average by 7.2%, 17.2%, 28.3%, and 36.3%, respectively. (3) A case studies also showed that the fusion-corrected method had good improvement and correction capabilities on the hourly precipitation forecast, especially for large-scale precipitation forecasts.