Abstract: Non-meteorological echoes, often manifested as clear-air echoes, can interfere with cloud observations made by millimeter-wave cloud radars (MMCR). Conditions for the formation of clear-air echoes are complex and variable. These echoes can often merge with meteorological echoes, posing a persistent challenge for MMCR quality control (QC). This study develops a clear-air echo QC model based on the U-net deep learning model, referred to as CR-Unet (Cloud Radar U-net), and utilizes minute-by-minute observational data from 49 MMCR stations across multiple provinces in China provided by the China Meteorological Administration from July 2023 to June 2024. Evaluation of the model performance reveals the following results: (1) The CR-Unet exhibits good generalization performance and is applicable to QC of reflectivity factors from most Chinese MMCRs, overcoming the limitation of threshold methods that require individual parameter tuning for each station. (2) The CR-Unet employs a four-level down-sampling and up- sampling architecture, utilizing three MMCR observables — radial velocity (V), linear depolarization ratio (LDR), and radar reflectivity factor (Z) — which are fused into a three-channel dataset. Data augmentation is applied to MMCR observation images of less frequent special weather phenomena such as snowfall and fog events. Using meteorological echo retention rate and clear-air echo identification rate as evaluation metrics, the model achieves a retention rate that exceeds 99% and an identification rate of 96%—97%. (3) Visual validation using MMCR observation data not included in the training demonstrates that the CR-Unet can perform classification and identification of MMCR data from various terrains across China, including plateaus, plains, basins, and coastal areas. It can also effectively identify cases with predominantly single-layer and multi-layer echoes, weak echoes, and missing LDR. The results also indicate that when LDR observations are faulty or significantly different from other MMCR data, they may have adverse effects on CR-Unet identification and classification. The deep learning-based CR-Unet shows a great application potential in the MMCR image quality control.