Abstract: By studying the distribution characteristics of precipitation raindrop spectra (DSD) in different cloud systems, it is possible to deepen the understanding of the microphysical process of precipitation, which is of great significance for improving the accuracy of radar quantitative precipitation estimation. This paper analyzed the observation data of the Parsivel precipitation particle spectrometer during the rainfall process in Huainan, Anhui Province from June to August 2016. The relationship between particle distribution characteristics, precipitation intensity and particle size, particle number density distribution and rain intensity contribution rate distribution, Z-I relationship and the M-P and Gamma distribution fitting are studied, and the conclusions are as follows: 1) The number density of particles smaller than 1mm in the convective cloud precipitation accounts for 73.22%, and its contribution to rain intensity is 34.07%. Large particles with a small number density provide most of the precipitation contribution rate; The number density of particles smaller than 1 mm in stratiform cloud precipitation accounts for 84.09 and its contribution rate to rain intensity is 58.82% 2) The Z-I relationship of summer precipitation in Huainan, Anhui in 2016 can be estimated by the convective cloud fitting curve Z=53.17×I^1.65. The Z-I relationship empirical formula (Z=300×I^1.40) underestimates precipitation under the same radar reflectivity factor. 3) In the process of convective cloud precipitation, Gamma distribution fitting is more accurate than M-P distribution fitting. In the process of stratiform cloud precipitation, the two fitting methods have little difference. Gamma distribution can more accurately represent the relationship between particle number density and diameter in Huainan, Anhui.