Abstract: In this study, a comprehensive effect assessment of a convective–stratiform mixed cloud precipitation enhancement operation in Anhui Province on June 17, 2023, was carried out using hourly rainfall data from national and regional automatic rainfall stations in Anhui Province, S-band radar data, and sounding data in the Anqing area. The results showed that shortly after cloud seeding, a narrow, strong echo region was observed within 1 km above the seeding height. This region expanded 24 min after the cloud seeding and formed a strong echo center at the seeding height. In addition, the echo shape changed from strip to block, accompanied by a concentrated area of severe convection. Proper echo units were identified and tracked using the centroid optimization-matching method and the Lagrangian method, and the best comparison unit was selected based on similarity measurement. Time-series variations in the five radar physical parameters for the seeded unit and comparison unit were further analyzed. The results showed that following the operation, the radar physical parameter values of the seeded unit increased significantly, reaching peaks within 42 min and then remaining stable. Conversely, the corresponding parameter values in the contrast unit exhibited a decreasing trend. The double-ratio values of the five radar physical parameters were greater than 1 within 1 h after the operation, indicating that the echo intensity of the contrast unit gradually decreased, while the seeded unit developed more vigorously and maintained a prolonged lifespan, demonstrating the obvious seeding effect. In addition, we optimize the estimation of natural rainfall in the affected area using a cluster-based historical regression method for the floating area. First, the K-Medoids clustering algorithm, combined with principal component analysis for dimensionality reduction, was used to accurately classify precipitation characteristics in southern Anhui Province. Then, the performance of six regression models was evaluated using cross-validation, and the results showed that the ElasticNet regression model performed best at predicting rainfall in the affected area. Finally, the regression model was applied in the individual cloud seeding case, providing the results of 2.92 mm rainfall increase in 3 h after operation and a 22.3% rainfall enhancement effect; the results of a one-sample t-test showed that the precipitation enhancement effect in the affected area was significant at the 95% confidence level.