Seed corn has much seed crushing and shedding losses during the peeling process, which seriously affects seed corn's yield and economic efficiency. Therefore, a combination of theoretical analysis, discrete element simulation, and orthogonal tests was used to investigate the interaction between the seed corn ears and the peeling mechanism. The optimal combination of peeling parameters was determined to optimize the peeling process of seed corn. Firstly, a theoretical analysis of the force and motion of the seed corn ear in the peeling mechanism was conducted to investigate the peeling mechanism-seed corn interaction during the peeling process and identify the main factors affecting the peeling performance. Secondly, a simulation model of seed corn ear-peeling mechanism interaction was established based on DEM. The better working range of peeling roller speed, peeling roller inclination angle, and pendulum swing were determined by analyzing the seed damage and shedding. Finally, according to the Box-Behnken design method, a three-factor, three-level orthogonal test was designed. The best combination of operating parameters for the seed corn peeling mechanism was screened by ANOVA and response surface analysis as follows: peeling roller speed of 300r/min, peeling roller inclination angle of 10°, and pendulum swing amplitude of 5°, at which the bract stripping rate was 94.13%, the grain falling rate was 1.564%, and the grain crushing rate was 1.292%. The optimal combination of operating parameters of the peeling mechanism obtained from the experiment significantly improved the peeling effect of seed corn. The research results can provide a reference for the optimal design of the seed corn peeling mechanism to enhance the peeling efficiency of seed corn and reduce the grain crushing rate and the grain falling rate during the peeling operation.