Abstract: A three-dimensional computational fluid dynamics (CFD) model coupled with the discrete phase model (DPM) is employed. Based on the L25(53) orthogonal experimental design method, the effects of three parameters—nozzle orifice spacing (10–50 mm), gas impingement angle (30°–50°), and inlet gas pressure (1–5 MPa)—on powder particle size D50 are systematically investigated. Range analysis and analysis of variance (ANOVA) are utilized to conduct statistical processing of the simulation data, revealing the hierarchical order, significance levels, and contribution rates of each factor’s effect on powder particle size. The results indicate that the hierarchical order of significance for the three factors affecting the median particle diameter D50 is as follows: inlet gas pressure (range 70.214 μm) > nozzle orifice spacing (range 49.518 μm) > gas impingement angle (range 25.226 μm). The influence degree of inlet gas pressure is significantly greater than that of the other factors, demonstrating that gas kinetic energy input serves as the decisive variable controlling powder particle size D50 within the investigated parameter range. Analysis of variance further validates the statistical reliability of these conclusions: inlet gas pressure (P = 0.000257 < 0.01) and nozzle orifice spacing (P = 0.007999 < 0.01) exhibit extremely significant effects on D50, whereas the gas impingement angle fails to reach the significance level (P = 0.136843 > 0.05). Contribution rate analysis reveals that inlet gas pressure contributes 54.35%, nozzle orifice spacing contributes 24.09%, and their cumulative contribution exceeds 78%, while the gas impingement angle contributes only 9.00%, further confirming the dominant roles of pressure and geometric structural parameters in the atomization process. Response surface analysis further reveals the nonlinear coupling effect between inlet gas pressure and nozzle orifice spacing. The optimal parameter window is identified as an inlet gas pressure of 2 MPa, a nozzle orifice spacing of 40 mm, and a gas impingement angle of 35°. Under this parameter combination, the numerically predicted D50 is 64.14 μm, deviating by less than 5% from the experimentally measured D50 (61.13 μm) of actually atomized powder, with good powder sphericity.