In order to accurately establish the discrete element model of moist sugarcane compound fertilizer, reasonably set the simulation parameters, and improve the accuracy of the simulation results of fertilizer particle dynamics, the angle of repose was used as the response value, and the sugarcane compound fertilizer with different moisture content was tested. The ternary regression equation of moisture content and angle of repose was established by the unloading bin method, and the coefficient of determination was 0.99. HertzMindlinwith JKR was used as the bonding model, and JKR surface energy, the shear modulus of sugarcane compound fertilizer, the static friction coefficient between sugarcane compound fertilizer and stainless steel, the static friction coefficient between sugarcane compound fertilizers, and the coefficient of restitution between sugarcane compound fertilizers were selected from nine initial parameters. Then the angle of reposediscrete element parameters model was established, the Pvalue of the model was less than 0.01, the coefficient of variation was 6.35%. Finally, through the simulation angle of repose test of the optimal parameter combination under different moisture content, the relative error between the simulation results and the physical angle of repose test was less than 7.66%, which verified that the calibration results and research methods of sugarcane compound fertilizer parameters with different moisture contents were reasonable and reliable. The research results could provide reference for the simulation research of sugarcane fertilization mechanical dynamics based on discrete element method.

In order to improve the structural stiffness of the small sugar cane harvester and avoid the occurrence of resonance while satisfying the demand for light quantization, the topology optimization method is used to optimize the frame. Firstly, the threedimensional model and finite element model of the small sugar cane harvesting carrier were established. The accuracy of the finite element mode is verified by modal test. Secondly, the static analysis of the frame is carried out and perform topology optimization with the objectives of minimum compliance and maximum dynamic stiffness, respectively. After the screening of optimized models, the safety of the optimization structure is verified by static analysis. Finally, physical properties are processed according to the optimized model and dynamic performance testing. The results of the transfer function test and the vibration test were obtained. The firstorder frequency is increased by 0.1 Hz, the peak frequency is slightly improved, and the quality is reduced by 10.6%. The amplitude of each frequency segment is larger than that of the frame before optimization. The firstorder frequency optimized by the side support beam is reduced by 0.3 Hz, the quality is reduced by 21.77%, the peak frequency is slightly improved, and the amplitude under 1-3 Hz is slightly smaller than the original frame. The firstorder frequency of the plateshaped structure optimized by dynamics is 0.9 Hz. The indicators of the transfer function are reduced. The amplitudes of each frequency section are less than the frame before optimization, and the frame quality is reduced by 12%, which is fully reached. The lightweight and dynamic characteristics of the frame beam structure are realized. It is known from the modal verification test: The frame of the small sugarcane harvester is optimized by three objectives, and the frequencies of each frame avoid the frequency range of engine excitation and other main excitations.

Due to the change of the ridge spacing of sugarcane plants, it is difficult for the conventional cultivator to control the quality of soil cultivation on both sides of the ridge, which causes inadequate soil cultivation, i.e., the “crater”phenomenon. We thus developed a multi-functional sugarcane intertillage cultivator and proposed a machine learning-based method to locate and identify sugarcane seedlings and calculate the coordinate classification during the intertillage period. Based on the YOLOv4 network, the method established a recognition model, which identifies and locates the area between the roots and the soil and accesses the coordinates. Then, the Support Vector Machine divided the coordinate data into two groups and processed them separately in real-time to obtain the tilt value. Finally, the data were adjusted according to the tilt value to modify the quantity and direction of soil supply. The results of the present study indicate that in use of the YOLOv4 recognition model, the recognition accuracy can reach 95.50%, and the classification accuracy using the SVM is 92.60%, which realizes the real-time dynamic recognition and classification calculation of sugarcane seedlings in the intertillage period, and provided data foundation for the development of intelligent sugarcane protection and joint operation machinery.