Aiming at the lack of a precise discrete element model in the simulation analysis of corn straw crushing, this paper established bond models with different structure and particle models with different properties of corn straw based on Hertz-Mindlin with bonding, and verified the discrete element model by combining physical test and simulation optimization design methods. Firstly, this study generates the bonded particle models based on Hertz-Mindlin with bonding for epidermis-epidermis, inner flesh-internal flesh, and epidermis-internal flesh, respectively. The bonding parameters of the model are calibrated with the help of a bending damage test, bending damage simulation test, and actual test with 2 mm/min speed. It shows that the maximum destructive force errors of the epidermis and inner flesh are 2.4% and 1.6%, respectively. On this basis, a discrete element model of corn straw is established by combining the calibrated parameters of the bonding parameters of epidermis—epidermis, inner flesh—inner flesh, and epidermis—inner flesh bond. The bending failure test shows that the mechanical properties of corn straw are similar between the simulated test and the actual test, and the maximum destructive power is 288 N and 292 N, respectively. The relative error is 1.36%. The feasibility of the discrete element model for simulation analysis is verified, which shows that the established discrete element model can be applied to the simulation analysis of corn straw cutting and crushing process.

Moisture content is an important parameter for estimating the quality of pellet feed, which is vital in nutrition, storage, and taste. The ranges of moisture content serve as an index for factors such as safe storage and nutrition stability. A rapid and non-destructive model for the measurement of moisture content in pellet feed was developed. To achieve this, 144 samples of Caragana korshinskii pellet feed from various regions in Inner Mongolia Autonomous Region underwent separate moisture content control, measurement using standard methods, and captured their images using a hyperspectral imaging (HSI) system in the spectral range of 935.5–2539 nm. The Monte Carlo cross validation (MCCV) was used to eliminate abnormal sample data from the spectral data for better model accuracy, and a global model of moisture content was built by using partial least squares regression (PLSR) with seven preprocessing techniques and two spectral feature extraction techniques. The results showed that the regression model developed by PLSR based on second derivative (SD) and competitive adaptive reweighted sampling (CARS) resulted in better performance for moisture content. The model showed predictive abilities for moisture content with a coefficient of determination of 0.9075 and a root mean square error (RMSE) of 0.4828 for the training set; and a coefficient of determination of 0.907 and a root mean square error (RMSE) of 0.5267 for the test set; and a relative prediction error of 3.3 and the standard error of 0.307.

There are many small objects in UAV images, and the object scale varies greatly. When the SSD algorithm detects them, the backbone network’s feature extraction capabilities are poor; it does not fully utilize the semantic information in the deeper feature layer, and it does not give enough consideration to the little items in the loss function, which result in serious missing object detection and low object detection accuracy. To tackle these issues, a new algorithm called RSAD (Resnet Self-Attention Detector) that takes advantage of the self-attention mechanism has been proposed. The proposed RSAD algorithm utilises the residual structure of the ResNet-50 backbone network, which is more capable of feature extraction, in order to extract deeper features from UAV image information. It then utilises the SAFM (Self-Attention Fusion Module) to reshape and concatenate the shallow and deep features of the backbone network, selectively weighted by attention units, ensuring the efficient fusion of features to provide rich semantic features for small object detection. Lastly, it introduces the Focal Loss loss function, which adjusts the corresponding parameters to enhance the contribution of small objects to the detection model. The ablation experiments show that the mAP of RSAD is 10.6% higher than that of the SSD model, with SAFM providing the highest mAP enhancement of 7.4% and ResNet-50 and Focal Loss providing 1.3% and 1.9% enhancements, respectively. The detection speed is only reduced by 3FPS, but it meets the real-time requirement. Comparison experiments show that in terms of mAP, it is far ahead of Faster R-CNN, Cascade R-CNN, RetinaNet, CenterNet, YOLOv5s, and YOLOv8n, which are the mainstream object detection models; In terms of FPS, it slightly inferior to YOLOv5s and YOLOv8n. Thus, RSAD has a good balance between detection speed and accuracy, and it can facilitate the advancement of the UAV to complete object detection tasks in different scenarios.

This study focuses on developing a finite element model for Caragana korshinskii Kom. branches and calibrating the necessary parameters. Using Caragana korshinskii Kom. branches in full bloom as the subject, we established a finite element model with ABAQUS and analyzed the relationship between the shear force and displacement. The Johnson–Cook model was employed to create a finite element shear model for Caragana korshinskii Kom. branches and to calibrate its parameters. Plackett–Burman and climbing tests were conducted to identify key parameters influencing the shear mechanical model of Caragana korshinskii Kom. branches, specifically the yield stress and strain. This included the strengthening index and fracture strain. Through central composite design experiments, the optimal parameters for the shear model of Caragana korshinskii Kom. branches were identified: a yield stress of 18.65 MPa, a strain strengthening index of 0.59, and a fracture strain of 0.54. The results indicated that the maximum error in the simulated shear force values for Caragana korshinskii Kom. branches of varying diameters was 9.68%. Additionally, the difference between the simulated and actual tensile force values was 1.7%. These findings indicate that the calibrated finite element model is accurate and effective in simulating the failure behavior of Caragana korshinskii Kom. branches, providing a valuable tool for optimizing cutting and harvesting processes.

Caragana korshinskii Kom. (CKB), widely cultivated in Inner Mongolia, China, has potential for silage feed development due to its favorable nutritional characteristics, including a crude protein content of 14.2% and a neutral detergent fiber content below 55%. However, its vascular bundle fiber structure limits the efficiency of lactic acid conversion and negatively impacts silage quality, which can be improved through mechanical crushing. Currently, conventional crushing equipment generally suffers from uneven particle size distribution, high energy consumption, and low processing efficiency. In this study, a layered aggregate model was constructed using the discrete element method (DEM), and the Hertz–Mindlin with Bonding contact model was employed to characterize the heterogeneous mechanical properties between the epidermis and the core. Model accuracy was enhanced through reverse engineering and a multi-particle-size filling strategy. Key parameters were optimized via a Box–Behnken experimental design, with a core normal stiffness of 7.37 × 10¹¹ N·m⁻¹, a core shear stiffness of 9.46 × 10¹⁰ N·m⁻¹, a core shear stress of 2.52 × 10⁸ Pa, and a skin normal stiffness of 4.01 × 10⁹ N·m⁻¹. The simulated values for bending, tensile, and compressive failure forces had relative errors of less than 10% compared to experimental results. The results showed that rectangular hammers, due to their larger contact area and more uniform stress distribution, reduced the number of residual bonded contacts by 28.9% and 26.5% compared to stepped and blade-type hammers, respectively. Optimized rotational speed improved dynamic crushing efficiency by 41.3%. The material exhibited spatial heterogeneity, with the mass proportion in the tooth plate impact area reaching 43.91%, which was 23.01% higher than that in the primary hammer crushing area. The relative error between the simulation and bench test results for the crushing rate was 6.18%, and the spatial distribution consistency reached 93.6%, verifying the reliability of the DEM parameter calibration method. This study provides a theoretical basis for the structural optimization of crushing equipment, suppression of circulation layer effects, and the realization of low-energy, high-efficiency processing.