Biochar is a multifunctional material efficiently developed from biomass. It is combined with fertilizers to prepare biochar-based fertilizers， which has excellent slow-release performance and minimal soil burden. The carbonization temperature， carbonization time， and heating rate during biomass carbonization process affect the physical and chemical properties of biochar. The biochar under different carbonization temperatures， carbonization times and heating rates have a significant impact on the slow-release performance of biochar-based fertilizers. In this study， BP neural network coupled with genetic algorithm was used to predict and optimize key process parameters during the carbonization process of coffee shell biochar in order to improve the slow-release performance of biochar-based fertilizers. Results had shown that based on BP neural network-genetic algorithm， rapid prediction and optimization of the slow-release performance of coffee shell biochar-based fertilizer had been achieved through experiments. The optimal process parameters were： carbonization time of 2.8 h， carbonization temperature of 780.7 ℃， and the heating rate of 15.1 ℃/min. The seven-day cumulative nutrient release rate of the biochar-based fertilizer prepared under this process parameter was 45.9%， and the slow-release performance was improved. The study proposed a new method for optimizing the parameters of biochar carbonization process， which provided new ideas for the development of high-performance biochar preparation processes and had certain reference significance for improving the performance of biochar-based fertilizers.

Panax Notoginseng is one of the most extensively cultivated and utilized bulk medicinal materials in China， and Yunnan Province is the main producing area for Panax Notoginseng in the country. Due to the influence of terrain and agronomic requirements， traditional agricultural machinery faces difficulties in entering the planting areas for operation， resulting in the current manual operation of Panax Notoginseng transplantation. Therefore， the development of a Panax Notoginseng transplanting machine is crucial for promoting the industrialization of Panax Notoginseng. As the crucial load-bearing structure of the transplanting machine， the frame significantly affects the overall performance of the vehicle. This paper focuses on the structural analysis of the transplanting machine frame， aiming to improve the performance of the frame and the entire vehicle， and provide theoretical basis for frame structural design. A three-dimensional model of the frame is established using SolidWorks software and imported into ANSYS software for static finite element analysis. After determining the relative error range between the stress values obtained from static electrical tests and experimental stress values， the dynamic loading performance of the frame under different working conditions and the modal vibration deformation analysis of the first eight modes are conducted. The analysis results indicate that the frame exhibits good strength performance， but significant deformation occurs at the seat position， indicating insufficient stiffness of the structure. Topology optimization is employed to optimize the frame design， aiming to reduce frame deformation while ensuring reasonable stress distribution. By altering the arrangement of diagonal brace brackets， the goal of reducing deformation is achieved. According to the optimization design results， the total mass of the frame increases by 8.739%， while the deformation is reduced by 88.268%， and the maximum stress is decreased by 11.693%. The improved frame exhibits reduced maximum stress and significantly improved deformation， demonstrating the applicability of finite element method and topology optimization technology in guiding the structural design of transplanting machine frames.

Cotton is one of the most important economic crops in the world, and verticillium wilt is the number one disease in major cotton production areas in the world. Verticillium wilt causes the leaves to wilt, fade and even fall off by infecting the roots of cotton, resulting in severe decline both cotton quality and its yield. The national standard divides the leaves suffering from verticillium wilt into five grades. The traditional detection method mainly relies on manual labor, which has problems such as subjectivity, inefficiency, and poor repeatability. The cotton verticillium wilt disease classification method, which is based on VFNet-Improved, Deep Sort, and collision line matching mechanisms as the main algorithm framework, realizes the statistics of the number of diseased leaves and the classification of the disease level under the condition of rotating video input. Based on the VFNet target detection network, the research first combined multi-scale training, dynamic convolution and other optimization methods to achieve accurate positioning of diseased leaves in rotating videos. Then the Deep Sort tracker was used to realize the correlation of the same leaf in the front and back frames, and a mask collision line matching mechanism was designed for the ID jump problem in the tracking process; finally, OpenCV was used to perform feature extraction and disease classification of leaves passing the mask line. The results showed that the VFNet-Improved model could achieve the best effect in the object detection algorithms with 0.906 mAP75, which was 0.012 higher than the VFNet model, and the FPS reached 12.9 frames/s. The tracking result MOTA of the Deep Sort was 0.835. R2, RMSE, MAE and MAPE of VFNet-Improved were 0.890, 5.138, 4.300 and 14.967% respectively, which performed high consistency with manual measurements. In conclusion, this study has demonstrated a novel tool for the accurate and efficient evaluation of cotton verticillium wilt, which would be of great significance for the cotton breeding and genetic analysis research.