In order to solve the problem of insufficient depth of trenching, low depth of deep-buried, high resistance to trenching in straw returning work, the design of trenching device for pneumatic straw returning to field was optimized. A parameter optimization design of the deep-buried straw was proposed to simulate the front-foot curve of the rake, and the cutting ability of the side cutting edge designed by Archimedes' helical design was used to cut the soil to complete the trenching operation. Matlab was used to reduce the resistance and consumption during the trenching operation of the bionic part obtained by binarization, expansion and edge coordinate processing of the front foot, and then the discrete element software EDEM was used to design the bionic trencher and the ordinary trencher. The simulations were compared and the simulation results were verified by the soil trough test. The analysis of the test results showed that the simulation trencher can reduce the resistance by 9.48% compared with the ordinary trencher during the trenching process. The quadratic orthogonal rotation combination design of Design-Expert 8.0 software was used in combination with response surface methodology to establish a regression mathematical model for deep burying rate and operating efficiency of the straw, speed of implement advancement, deep burial depth of straw and straw mulching amount. The analysis result showed that the order of the factors affecting the deep burial rate of straw in a descending trend was straw coverage, deep burial depth of straw and tool advancement speed;the effect of each factor on work efficiency was the speed of implement advancement, deep burial depth of straw and straw cover;interaction in depth of straw depth and straw coverage, machine tool speed and straw coverage had a significant impact on work efficiency. After optimization, the optimal working parameters of the trenching device were achieved with deep buried rate weight of 0.7 and work efficiency weight of 0.3. The forward speed of the implement was 1.63m/s, the deep buried depth was 27.97cm, and the straw coverage was 340.54kg/hm2, the depth buried rate of straw was 90.491%, and the working efficiency was 5.4hm2/h. The analysis result showed that the optimized work effect was in line with the straw agronomy requirements. The research results can provide a theoretical reference and basis for the optimal design of trenching device for deep returning of corn straw.