Abstract: In order to improve the burying uniformity in soil during rice straw returning, based on the straw crushing-rotary tillage and burying returning machine, an internal and external opposed double-screw burying roller was designed. A single-factor discrete element simulation test (two-factors and ten-levels) was conducted to analyze the influence of straw length (6-15 cm) and the screw angle (30°-75°) of the external screw belt in the burying roller on the coefficient of variation of straw mass distribution. After getting patterns and optimization conclusions from the simulation tests, the parameters of the burying roller were optimized and field experiments were conducted to verify the performance of the device. The simulation results showed that, when the straw length was constant, as the screw angle increased, the coefficient of variation of straw mass distribution initially decreased and then increased. The quadratic function was used to fit the relationship between the coefficient of variation of straw mass distribution and the screw angle under different straw lengths. The curves of 10 different straw lengths were analyzed. It was found that the coefficient of variation of straw mass distribution in the range of 55° -62° had the minimum value. When the screw angle was constant, the coefficient of variation of straw mass distribution was linearly related to straw length. Specifically, for every 1 cm reduction in straw length, the coefficient of variation of straw mass distribution decreased by 3.255 percentage points. The reason for this phenomenon was that the straw length was positively correlated with the resistance encountered in the soil. Shorter straw encounters less resistance in the soil and it was more easily buried into the deeper soil layer, thereby reducing the coefficient of variation of straw mass distribution. When the screw angle of the screw blade was 55° and the straw length was 6 cm, the coefficient of variation of straw mass distribution reached its minimum value of 27.59 percentage points, and the straw burying uniformity was the highest, which was the optimal working parameter on the simulation. And the spatial distribution of straw in soil showed a decreasing distribution along the depth of soil which the proportion of straw mass in the upper soil layer (>0-50 mm) was greater than that in the middle soil layer (50-100 mm), which in turn was greater than that in the lower soil layer (100-150 mm). Because the straw was initially distributed on the soil surface and moved to the depth of the soil under the action of the burying roller. The greater the downward movement, the greater the displacement and the stronger the resistance encountered, and the less the straw distribution. With the changes in the two factors of the screw angle and straw length, the percentage of straw mass in the upper soil layer (>0-50 mm) and the lower soil layer (100-150 mm) had opposite trends, while the percentage of straw mass in the middle soil layer (50-100 mm) fluctuated slightly. This indicated that the increase in straw mass in the lower soil layer mainly comes from the upper soil layer, and reducing the straw mass in the upper soil layer was the key to improving the uniformity of straw mass distribution. Field experiments showed that, when the screw angle of the external screw belt was 55° and the straw length was 6 cm, the percentage of straw mass in the soil layers at the depth of >0-50 mm, 50-100 mm and 100-150 mm were respectively 40.91%、34.58% and 24.51%, the coefficient of variation of straw mass distribution was 25.35 percentage points. Compared to the results of the simulation, the deviations the percentage of straw mass in the different soil layers and the coefficient of variation of straw mass distribution were 0.04-2.24 percentage points, with relative errors of 0.2%-8.8%. The simulation results were basically consistent with the experimental results, confirming the effectiveness of the parameter optimization of the burying roller and indicating good uniformity of rice straw returning and high seedbed quality after the machine operation. The research results provide support for the design of straw crushing-rotary tillage and burying returning machine and the improvement of straw returning quality.