In this study, to obtain the contact parameters of particulate materials accurately and quickly in residual film mixture after sieving, the contact parameters of particulate materials were calibrated via a physical test and simulation test. By using the self-made dynamic angle of a repose measurement test bench, the dynamic angle of repose of the particulate materials was measured at 41.32°, and the standard deviation was 1.33°. A discrete element simulation of the dynamic angle of the repose test was performed via an EDEM screening experiment design through a simulation of a combination of different parameters, with the dynamic angle of repose as the response value. Through simulation experiments, three significant influencing factors, as well as the level range of each factor, were confirmed. By using the response surface experiment, a mathematical model of the dynamic angle of repose and the three most influential parameters was created. The analysis of variance showed that the determination coefficient R² and the correction determination coefficient R²_adj were 0.9824 and 0.9598, respectively. The model had a good fit. The variable coefficient was 2.06% and the lack of fit was non-significant, which showed that the regression model was very significant, and the dynamic angle of repose could be predicted according to the model. By solving the optimization for the mathematical model, the optimal combination of parameters with three important influencing factors were obtained. The results showed that the coefficient of the static friction between soil and soil was 0.38, the coefficient of the rolling friction between soil and soil was 0.08, and the coefficient of the static friction between cotton residue and cotton residue was 0.33. The relative error of the dynamic angle of repose between the simulation with the optimal parameter combination and the physical test value was 2.64%. The results could provide a reference for the calibration of the discrete element model parameters of other agricultural particulate material, as well as provide a theoretical basis for the design of related collecting and conveying machinery.

The friction characteristic parameters of cotton stalks are important basic physical parameters required to establish the crushing mechanics model and study the separation machinery of film residue mixtures. In this paper, the friction characteristics of cotton stalks were studied using response surface methodology, and the influence of the variation in contact materials, sampling location, and moisture content on the static sliding friction coefficient (μ_s) and stable static rolling angle (φ_s) were analyzed. The results show that the contact materials, sampling location, and moisture content significantly influence the μ_s and φ_s. When the contact material is cotton stalk bark, the μ_s and φ_s values of cotton stalks are maximum. When the sampling location of the cotton stalk changes from top to bottom, the μ_s increases and φ_s decreases, respectively, while the moisture content has an opposite influence on the μ_s and the φ_s. The model coefficient of the μ_s and the φ_s indicates that the influencing factors have a high degree of explanation for the influence on the μ_s and φ_s. This study on the friction characteristics of cotton stalks can provide theoretical references and basic parameters for the separation technology research and equipment development of film residue mixtures.

Addressing the issues of the sand cat swarm optimization algorithm (SCSO), such as its weak global search ability and tendency to fall into local optima, this paper proposes an improved strategy called the multi-strategy integrated sand cat swarm optimization algorithm (MSCSO). The MSCSO algorithm improves upon the SCSO in several ways. Firstly, it employs the good point set strategy instead of a random strategy for population initialization, effectively enhancing the uniformity and diversity of the population distribution. Secondly, a nonlinear adjustment strategy is introduced to dynamically adjust the search range of the sand cats during the exploration and exploitation phases, significantly increasing the likelihood of finding more high-quality solutions. Lastly, the algorithm integrates the early warning mechanism of the sparrow search algorithm, enabling the sand cats to escape from their original positions and rapidly move towards the optimal solution, thus avoiding local optima. Using 29 benchmark functions of 30, 50, and 100 dimensions from CEC 2017 as experimental subjects, this paper further evaluates the MSCSO algorithm through Wilcoxon rank-sum tests and Friedman’s test, verifying its global solid search ability and convergence performance. In practical engineering problems such as reducer and welded beam design, MSCSO also demonstrates superior performance compared to five other intelligent algorithms, showing a remarkable ability to approach the optimal solutions for these engineering problems.

To address the low efficiency and stress-inducing nature of traditional manual weighing for dairy cows, this study proposes a semantically guided 3D reconstruction and body weight estimation method for dairy cows. First, a dual-viewpoint Kinect V2 camera synchronous acquisition system captures top-view and side-view point cloud data from 150 calves and 150 lactating cows. Subsequently, the CSS-PointNet++ network model was designed. Building upon PointNet++, it incorporates Convolutional Block Attention Module (CBAM) and Attention-Weighted Hybrid Pooling Module (AHPM) to achieve precise semantic segmentation of the torso and limbs in the side-view point cloud. Based on this, point cloud registration algorithms were applied to align the dual-view point clouds. Missing parts were mirrored and completed using semantic information to achieve 3D reconstruction. Finally, a body weight estimation model was established based on volume and surface area through surface reconstruction. Experiments demonstrate that CSS-PointNet++ achieves an Overall Accuracy (OA) of 98.35% and a mean Intersection over Union (mIoU) of 95.61% in semantic segmentation tasks, representing improvements of 2.2% and 4.65% over PointNet++, respectively. In the weight estimation phase, the BP neural network (BPNN) delivers optimal performance: For the calf group, the Mean Absolute Error (MAE) was 1.8409 kg, Root Mean Square Error (RMSE) was 2.4895 kg, Mean Relative Error (MRE) was 1.49%, and Coefficient of Determination (R²) was 0.9204; for the lactating cows group, MAE was 12.5784 kg, RMSE was 14.4537 kg, MRE was 1.75%, and R² was 0.8628. This method enables 3D reconstruction and body weight estimation of cows during walking, providing an efficient and precise body weight monitoring solution for precision farming.