The project on study for formation mechanism and theories and methods of detection and prediction of severe weather disasters in China from National Key Basic Research Development Program has been studied by all the scientists from this project for five years. It has been made important progresses including seven aspects related to heavy rainfall: (1) Muti-scales physical models of heavy rainfall within the Meiyu front based on the real-time observed data have been proposed; (2) It has established a synoptic model for heavy rainfall within the Meiyu front; (3) Meiyu front is Meiyu frontal system which consists of multi-synoptic systems with different space-time scales. It has structure characteristics of sub-tropical frontal system which is between structure characteristics of extra-tropical frontal system and ITCZ. However, sometime it has dual-peaks structure in the middle-downstream of Yangtze River. And its maintenance and development mechanism has also been discussed; (4) Several kinds of quantitative remote sensing retrieval theories and methods have been proposed for meso-scale heavy rainfall; (5) Dual-Dopplor detecting methods and retrieval theories for meso scale heavy rainfall have successfully been studied; (6) The meso scale heavy rainfall numerical prediction model system with 3Dvar has been developed; (7) The field experiments of heavy rainfall within Meiyu front in the middle and down stream of the Yangtze River have successfully been organized in 2001 and 2002. And in the experiments we collaborated with Japans scientists. Base on these experiments, the database of heavy rainfall has been established and all the scientists can share all these data. The above progresses showed that the project on heavy rainfall study has been reached to the goals and finished the tasks in the last five years. It is substantial scientific base for advancing deeply studying of heavy rainfall in China. In this paper, we also discussed some problems existing in implementation of this project and further study in the future.