Aiming at the characteristics of complex terrain and the variety of radar types in Sichuan Province, this paper designs an algorithm for evaluating the quality of single or dual-polarization weather radar data over a long period of time. The algorithm qualitatively evaluates whether the weather radar data have the error of Empty Echo, electromagnetic interference or ground object clutters, and quantitatively evaluates the data quality of reflectivity factor (Z_h), differential reflectivity (Z_dr), coefficient of correlation (C_C), differential phase shift (φ_dp) and ground object clutter (G_C). By evaluating the data quality of 12 new-generation weather radars in Sichuan Province from May to September 2022, the following results are obtained: (1) The weather radar in Sichuan Province is affected by electromagnetic interference, which seriously affects the data quality. More than 74% of the Yibin and Guangyuan radar data have electromagnetic interference echoes. The data of Yibin and Leshan radar stations have many error data of “Empty Echo”. More than 77.9% of the Guangyuan radar data have ground object echo interference that is caused by wind turbines. (2) If the radar performance is normal, the mean value of the ground object clutter coefficient G_C does not change much. The standard deviation of G_C coefficient can effectively reflect the problem of radar data, especially the problem of “Empty Echo” data. (3) The standard deviation of Z_h in the 12 weather radars in Sichuan Province is generally small, and the standard deviation of Z_h in Ya’an, Kangding and Guangyuan stations is better than that of others. (4) The quality of Z_dr and C_C data of 3 dual-polarization weather radars in Sichuan Province is optimum, but the data quality of φ_dp is poor.