The impact of radar data assimilation on the numerical simulation of a rare squall line in Zhejiang Province on 4 March 2018 was studied in this paper. The weather research and forecasting (WRF) model and the gridpoint statistical interpolation three-dimensional variational (GSI-3DVar) data assimilation system were used. In this study, multiple-Doppler radar data assimilation was applied to initialize a squall-line convective system, and its impact on the simulation was evaluated and discussed. The results showed as follows. Radar data (especially the radar reflectivity) effectively improve the simulation of the 〖JP2〗boundary layer〖JP〗 characteristics of the squall line, which further improves the simulation effect of the composite reflectivity, precipitation and wind process of the squall line. Radar reflectivity assimilation, which directly adjusts the hydrometeors, improves the precipitation simulation in the storm and its induced evaporative cooling and forms a more intense cold pool close to reality, thus improving the simulation of the squall line. The radar reflectivity assimilation〖JP2〗 has a greater contribution than the radial velocity assimilation. Relative to assimilating radar data in non-critical areas, assimilating the small and medium-scale information carried by radar observations in key areas is more important for improving the simulation of the squall line.