Abstract: To address the issues of poor seeding quality in traditional grooved wheel seed metering devices for alfalfa row seeding, specifically inaccurate seeding amount and uneven seed distribution, this study designed a novel screw-internal grooved wheel combination precision seed-metering device for alfalfa. The device employs a seed-pushing screw to convey seeds, an internal grooved wheel for quantitative seed filling, and staggered grooved teeth to ensure uniform seed discharge, thereby achieving a uniform and continuous seed flow. Based on the material properties of alfalfa seeds and the agronomic requirements for planting, the structural parameters of the key components of the seeding wheel were designed. The kinetic characteristics of seeds during the seed-filling and seed-discharging stages were analyzed to identify the key parameters influencing seed-metering performance. Using extended discrete element simulation software (EDEM), the movement patterns and distribution of seeds during the filling and discharging processes were simulated under different grooved tooth types and grooved tooth numbers, thereby defining the fundamental structure of the seeding wheel. A two-factor optimization experiment and performance comparison validation experiments for the screw-internal grooved wheel combination seed-metering device were conducted using a seed-metering performance testing platform, thereby enabling the optimization and validation of the optimal working performance. Single-factor simulation experiments with different types of grooved tooth indicated that the linear type grooved tooth can discharge all seeds from the perforations more effectively than the arc type grooved tooth or the vertical type grooved tooth. Under this configuration, the seeding wheel achieved a single-revolution seeding amount of 2.26 g, which was consistent with actual measurement results. Single-factor simulation experiments with different numbers of grooved tooth indicated that when the number of grooved tooth was 9, the seed flow was more continuous and uniform compared to when the number of grooved tooth was six or twelve. During seed discharging over a single rotation of the seeding wheel, the seeding rate exhibited minimal fluctuation, with a standard deviation of 0.024 g/s. Bench experiments indicated that through solving the two-factor experimental regression equation, optimal seed-metering performance was achieved when the inclination angle of the tooth surface was 52.5° and the working speed was 9 km/h. Validation experiments demonstrated that under these optimal parameter conditions, the seed-metering device exhibited a coefficient of variation of displacement of 1.8%, a coefficient of variation of uniformity of 15.0%, a seed damage rate of 0.03%, and a blocking rate of 0.07%. These results were found to be largely consistent with the optimization results from the two-factor experiments. Furthermore, in comparison with the toothed type external grooved wheel seed-metering device operating at the same working speed, the screw-internal grooved wheel combination precision alfalfa seed-metering device demonstrated coefficients of variation of displacement, coefficients of variation of uniformity, seed damage rate, and blocking rate that were 2.8, 2.6, 0.03, and 0.02 percentage points lower, respectively. These improvements indicate that the proposed device fully meets the requirements for precision strip seeding of alfalfa. The findings obtained in this study provide a valuable reference basis for enhancing the seeding quality of precision strip seeding equipment designed for small-particle seeds.