The seeds movement under the highspeed operation condition of corn air suction metering device is very complicate, and it is difficult to carry out accurate analysis and calculation. The discrete element (DEM) and computational fluid dynamics (CFD) coupling method was used to accurately simulate the working process of air suction metering device, the bonding model of Zhengdan 958 corn seed cohesive particles was established, the airway structured grid was divided, and the related parameters were set up. The coupling simulation of EDEM and CFD was realized. Through simulation, the working process of the air suction metering device was accurately simulated, and the drag force and movement speed of the seeds were accurately analyzed. It was concluded that the seed filling ability of different seeds from big to small was as follows: small flat shape, round shape and big flat shape. In order to improve the operation quality of seed metering device and ensure the adaptability of seed metering device to seeds, the optimum design of seed plate structure for big flat shape was studied. The mathematical model of filling process was established to optimize the height of the plate type hole embossment and the angle parameter of the type hole embossment. Because of the complex movement caused by the interaction between seed populations and the interaction between seed populations and metering devices when the metering plate was rotated, further analysis and determination of the optimum angle were needed. In order to obtain the best performance parameters of the seed metering device, the angle of the type hole embossment, the height of the type hole embossment and the height of the seed layer were taken as the test factors, and the three factors of the seeding qualified index, multiple index and missing index were used as test indicators. Performing a threefactor quadratic rotation orthogonal combination test, the DesignExpert 8.0.6 software was used to analyze the test data by multiple regression analysis and response surface analysis. The primary and secondary factors affecting the qualified index and multiple index were obtained as follows: the height of type hole embossment, the height of seed layer and the angle of type hole embossment. The primary and secondary factors that affecting the missing index were the height of type hole embossment, the angle of type hole embossment and the height of seed layer. Using multiobjective optimization method, the optimum combination of parameters was determined as follows: the angle of type hole embossment was 35.76°, the height of type hole embossment was 3.11mm, the height of seed layer was 55.61mm, and the qualified index of seed metering was the highest. Under this condition, the qualified index of seed metering device was 9160%, the missing index was 390%, and the multiple index was 450%. The validation test of the optimization results was carried out. When the operating speed was 14km/h and the wind pressure was -3kPa, the qualified index was 91.90%, the missing index was 3.89%, and the multiple index was 4.21%. The validation results were basically consistent with the optimization results, compared with the firstgeneration seed metering device, the qualified index and missing index were better. It can ensure the precision seeding of corn.