To address the challenge of accurately calibrating discrete element method (DEM) parameters for superabsorbent polymer (SAP) particles with irregular morphology and micron-scale size, a combined experimental–numerical calibration strategy was developed to enhance simulation reliability. The angle of repose was measured as 39.17° using digital image processing and adopted as the response variable. A DEM model was established in EDEM based on the actual particle size distribution and morphology, and the JKR contact model was employed for calibration. Three significant parameters governing particle flow behavior were identified via Plackett–Burman design: particle–particle rolling friction coefficient, particle–stainless steel restitution coefficient, and particle–stainless steel static friction coefficient. The optimal region was determined using a two-stage steepest ascent method. A refined second-order regression model was subsequently constructed using Box–Behnken design, yielding optimal values of 0.188, 0.469, and 0.735. Validation results show that the mean simulated angle of repose is 39.81° with the relative error is 1.63%. These results demonstrate that the calibrated parameters can reliably reproduce the macroscopic behavior of SAP particles. This study provides a reproducible framework for DEM calibration of irregular micron-scale particles.