Abstract: The rice-wheat cropping system is the main grain production model in many areas of China, especially in Jiangsu with the area of about 1.8×106 hm2. In other provinces of China along the Yangtze River valley, such as Sichuan, Anhui and Hubei, there are large areas of such cropping system. However, the rice straw processing before wheat planting has caused some problems in these places. As straw burning is forbidden by the government, straw incorporated to field is an available alternative method to process the rice straw. Traditionally, rice straw should be smashed and buried before wheat planting, which requires time and energy. Also, burying rice straw in soil will influence the growing of wheat if the seed stay on the straw. Furthermore, the soil nutrient will be consumed during straw decomposing. The clean area planting technology under full straw cropland, invented by Nanjing Research Institute for Agricultural Mechanization, Ministry of Agriculture and Rural Affairs, China, can complete seeding and straw mulching in the same time, which is a good method to deal with the straw processing problem. However, as the quantity of rice straw is relatively larger than any other straw, the excess rice straw mulching will impact the emergence of wheat, which is likely to cause low yield. To solve the problem, in this paper, we provided the rice straw processing model of part burying and part mulching based on the clean area planting technology under full straw cropland. Then, different straw distributed retention devices with different structures, including flat board with horizontal upward opening, oblique board with horizontal downward opening, flat board with several longitudinal openings and curved board with several longitudinal openings, were designed and introduced respectively. After the tests of these distributed retention devices under different parameters, the structure of curved board with several longitudinal openings was selected as the best one. On the basis of structural development, the central composite test method was used to optimize the key parameters of straw distributed retention device. The gross width of longitudinal opening, quantity of longitudinal opening and revolving speed of picking and smashing device were taken as the influencing factors. The variable coefficient of uniformity and deviation rate of distribution were taken as response values in the experimental study. Orthogonal rotational quadratic combination test with three factors and three levels was made to evaluate the combined influence of the factors on the test index value. Furthermore, regression equations to describe the relationships between the factors and each assessment index were established by using the regression analysis and response surface analysis with the software Design-Expert 8.0.6. The optimum combination of the selected parameters was obtained and verified, and the experimental verification of the mathematical model was also conducted. The results showed that trial factors had great effects on the performance of straw distributed retention device. The significant effects of quantity of longitudinal opening, gross width of longitudinal opening and revolving speed of picking and smashing device on reducing the variable coefficient of uniformity were in a decreasing order. The significant effects of gross width of longitudinal opening, quantity of longitudinal opening and revolving speed of picking and smashing device on reducing the deviation rate of distribution were in a decreasing order. The best parameters of the integrated straw distributed retention were as follows: when the gross width of longitudinal opening was 600mm, the quantity of longitudinal opening was 7 and the revolving speed of picking and smashing device was 1900r/min, the variable coefficient of uniformity and deviation rate of distribution were 18.75% and 0 respectively. The model validation tests were repeated for 3 times on the simulation field by using the optimization results, the values of practical variable coefficient of uniformity and deviation rate of distribution were 19.68% and 0 respectively. The percentage points of absolute error between the experimental and predicted values of variable coefficient of uniformity and deviation rate of distribution were 0.93 percentage point and 0.33 percentage point respectively, which indicated a reasonable choice of optimization conditions. The research results can provide references of technology and equipment for wheat planting after the harvest of rice in rice-wheat cropping system area in China.