Accurate and real-time measurement of tractor drive wheel slip ratio under plowing conditions is essential for improving overall machine performance and tillage quality. To address the limitations of existing methods—namely low measurement accuracy, poor anti-interference capability, and low efficiency—this study proposes an online slip ratio measurement method based on multi-sensor fusion and adaptive filtering. A real-time measurement system was developed by integrating GNSS, IMU, and wheel encoders. Furthermore, a lens-based quasi-oppositional learning strategy and a good-point-set initialization mechanism were introduced to enhance the mayfly algorithm, which was then used to optimize a parallel Kalman filter, forming the improved mayfly algorithm–parallel Kalman filter (IMA-PKF). This approach enables adaptive real-time adjustment to random noise disturbances encountered during plowing operations, thereby enhancing robustness. Simulation results show that under non-interference conditions, the IMA-PKF algorithm achieves a root mean squared error (RMSE) of 0.0214, representing a 74.8% reduction compared with the conventional KF algorithm. In addition, compared with PSO-PKF and MA-PKF, the RMSE accuracy is improved by approximately 62.23% and 49.41%, respectively. When disturbance points are introduced, IMA-PKF still maintains the lowest estimation error, with an RMSE of 0.0359, demonstrating excellent stability and anti-interference capability. Field experiments under different plowing depths further validate the robustness of the method: the maximum slip ratio measurement error is only 1.94%, with bias controlled within 2%. Compared with KF, the proposed method reduces mean absolute error (MAE) and RMSE by up to 36.29% and 37.06%, respectively. Overall, the IMA-PKF algorithm enables accurate and stable online measurement of tractor drive wheel slip ratio under diverse plowing conditions, providing a solid theoretical and technical foundation for improving tractor performance and operational efficiency.

To achieve high-precision non-destructive monitoring of SPAD values in winter wheat, this study proposes an estimation method integrating multi-feature optimization with machine learning. Based on UAV multispectral imagery and synchronous ground measurement data from 33 plots, the research was conducted during three critical growth stages: jointing, heading, and grain filling. The PCC-RF-CV method was employed for feature fusion and optimization, identifying optimal feature combinations for each stage from multiple vegetation indices and texture features. Six machine learning models were constructed for comparison. Results indicate: the multi-source feature fusion strategy demonstrated superior performance throughout all growth stages; PCC-RF-CV effectively optimized feature inputs, establishing optimal feature sets for each stage; The XGBoost model developed for the grain filling stage achieved the best estimation performance (validation set R² = 0.92, RMSE = 0.36, MAE = 0.30). This study provides a reliable method for accurately estimating SPAD values in winter wheat and analyzing canopy spectral dynamics, offering robust technical support for crop growth monitoring and precision agriculture.

Frequent greenhouse collapses under strong wind conditions highlight the limitations of existing structural design methods. Although most previous studies have focused on the stability and load-bearing capacity of greenhouse frames, the dynamic interaction among the greenhouse film, tensioning straps, and structural frame remains insufficiently investigated. This study proposes a novel flat-elliptical pipe plastic greenhouse and derives its mechanical equilibrium equations to analyze the stress behavior of the greenhouse film under wind loading. Using ABAQUS finite element software, an advanced contact model was developed to examine the wind-induced response of the greenhouse system. The results indicate that the structural frame plays a critical role in governing film deformation. In particular, the contact pressure and shear stress between the frame and the film are significantly higher than those between the film and the tensioning straps, underscoring the necessity of explicitly considering contact effects in greenhouse structural design.

Population growth and technological progress significantly contribute to the climate change intensification, the increasing global energy demand, and the depletion of natural resources, while also generating massive amounts of waste. Agriculture is one of the main sectors with the highest biomass production, representing an essential factor for the bioeconomy. Agricultural wastes represent a sustainable and abundant source of organic matter, used in the production of biofuels, which are considered a promising alternative to meeting the global energy demand. Thus, considering the challenges associated with agricultural waste generation and their environmental impacts, this review aims to provide a comprehensive analysis of recent advances in the valorization of agricultural waste for sustainable biofuel production. The paper focuses on the main categories of agricultural waste, the associated conversion technologies, and the integration of these processes within the circular economy framework. In addition, to achieve the objectives of the study, a bibliometric analysis was conducted on publications from 2012–2024 indexed in the Web of Science Core Collection (WoS), highlighting research trends, leading journals, and thematic clusters in the field of agricultural waste valorization.

Teff (Eragrostis tef), a staple crop in Ethiopia and an emerging global superfood, is increasingly being cultivated through transplanting to improve yield and resource efficiency. The development of efficient teff transplanting machinery requires a thorough understanding of the physical and mechanical properties of teff seedlings. This study investigated the physical and mechanical characteristics of four teff varieties (Quncho, Dukem, Tseday, and the local variety Gemechis) at 20, 25, and 30 days after sowing. The analyzed parameters included seedling weight, height, stem diameter, moisture content, tensile strength, shear stress, and coefficient of friction. Descriptive statistical analyses were conducted using Statistix 10 and MATLAB 2024b, and results were expressed as mean values for each parameter. The findings revealed that as the seedlings matured, their weight, height, stem diameter, tensile strength, and shear stress increased, whereas moisture content decreased. The measured tensile strength values were as follows: Quncho (12.17±0.053 N, 14.46±0.964 N, and 18.297±0.577 N), Dukem (12.36±0.058 N, 14.597±1.11 N, and 18.763±0.57 N), Tseday (12.66±0.057 N, 14.097±0.81 N, and 17.9±1.155 N), and Local (Gemechis) (12.18±0.057 N, 14.32±0.764 N, and 19.12±0.012 N) for 20-, 25-, and 30-day-old seedlings, respectively. All tensile strength values exceeded the forces applied by the transplanter, indicating that seedlings of these four teff varieties possess adequate strength and resilience, making them suitable for the design and development of mechanical teff transplanters.

To improve the accuracy of contact parameters in discrete element simulations of seed metering devices and avoid errors arising during parameter measurement, this paper does not determine the values of each parameter in advance during parameter calibration. Instead, it directly uses the sliding friction angle, angle of repose, and rebound height as indicators, and employs regression analysis to calibrate the static friction coefficient, rolling friction coefficient, and restitution coefficient. The accuracy of the simulation parameters is verified through comparative experiments with the seed metering device. The results show that: the static friction coefficient is proportional to the sliding friction angle, and the correlation between the two is independent of material properties. The calibrated static friction coefficients between seeds, PMMA, and PLA are 0.39, 0.43, and 0.54, respectively; the rolling friction coefficient of seeds is exponentially related to the angle of repose, and the rolling friction coefficient between seeds is 0.029; the restitution coefficient is quadratically related to the rebound height in drop tests, and the calibrated restitution coefficients of seeds with PMMA and PLA are 0.33 and 0.32, respectively; the verification experiments demonstrate that the motion patterns of seed populations in simulation and physical experiments are similar, and the differences in seed filling rates between the two are not significant, proving the reliability of the calibrated simulation parameters.

This study investigates the wind resistance performance of a novel flat-elliptical pipe greenhouse by developing a finite element model of the greenhouse frame using ABAQUS software. Based on the Davenport fluctuating wind speed spectrum and the Shiotani correlation coefficient, a MATLAB program was developed using an autoregressive (AR) model within a linear filtering method to simulate fluctuating wind speed time-history curves. Both static and full dynamic wind loads were applied to the model to evaluate the wind resistance performance of greenhouse frames with 5 arches and 50 arches. The results indicate that under static wind loading, the stresses in the frames did not exceed the elastic limit, and the deformations remained within the allowable range. However, under full dynamic wind loading, the stresses in the frames exceeded the elastic limit and entered the plastic stage. Although the lateral displacement in the plastic stage remained within permissible limits, the overall structure failed to meet the safety requirements of the greenhouse.

With the advancement of intelligent farming technology, computer vision-based animal behavior recognition has become an important tool for improving production efficiency and animal welfare in modern farming management. To overcome the challenge of balancing computational efficiency and accuracy in existing behavior recognition systems, this study proposes an optimized model based on YOLOv8-EAPNet for accurately recognizing four main pig behaviors: standing, sitting, lateral lying, and prone lying. The framework adopts a multi-level lightweight design, incorporating three advanced technologies—C2f-ECA, SPPELAN, and Detect_AFPN—to enhance joint feature response, resolve spatial differences between sitting and lateral lying, and reconstruct semantics in occluded areas. This strengthens the model's robustness in complex farming environments and significantly improves the accuracy of pig behavior recognition. Validated on farm data, the model achieved an average precision improvement of 1.5% on a self-built dataset, with specific accuracy increases of 0.9% for standing, 1.7% for sitting, 3.0% for prone lying, and 0.3% for lateral lying. This technology provides an automated tool for early warning of limb injuries and respiratory diseases in pigs, promoting the upgrade of intelligent health management in the livestock industry and supporting the modernization of large-scale pig farming.

The study investigates the influence of different substrates on seed germination and early plant development under controlled environmental conditions, using a specially designed intelligent plant growth simulation system. Experiments were conducted with lettuce and radish seeds sown in peat, garden compost, and podzol. The system allowed precise control of temperature, humidity, soil moisture, light, and carbon dioxide, enabling a reliable comparison of treatments. The initial physical and chemical properties of the substrates were characterized, showing the lowest moisture and density in peat, while compost and podzol displayed higher values and more balanced nutrient content. The results demonstrated that lettuce germinated successfully across all substrates, although peat delayed germination by 5–7 hours and true leaves appearance by up to 12 hours. In radish, the effect was more pronounced: compost and podzol ensured rapid germination (94–95 hours) with high success rates (~98%), whereas peat resulted in delays exceeding 100 hours and germination rates below 25%. Correlation analyses highlighted strong links between delayed germination and reduced performance, especially in peat. Overall, the study confirmed that substrate choice is critical for achieving uniform and efficient germination, with compost and podzol providing optimal conditions, and emphasized the practical value of simulation systems for improving crop establishment.

To ensure food security and reduce harvest losses, improving the monitoring accuracy of grain combine harvester operation loss is of great importance. This paper systematically analyzes the technical progress of piezoelectric sensing applications in this field. In terms of materials, piezoelectric thin films (PVDF) exhibit faster response speeds (signal attenuation shortened by 30%), but are prone to short circuits in high-humidity environments. Piezoelectric ceramics (PZT), when combined with a double-layer vibration isolation structure, can effectively reduce vibration interference errors to below 5%, providing better stability. Regarding sensor structure, the array layout enhances multi-target recognition, while the innovative double-layer cross structure enables analytical positioning of the spatial distribution of grain collisions, offering a new approach for accurately calculating loss rates. In signal processing algorithms, support vector machines (SVM) and decision trees perform well with small sample sizes; however, combining them with discrete element simulation (EDEM) is necessary to optimize feature extraction. Among these methods, the WOA-BP algorithm can control monitoring error within 6.23% through adaptive parameter adjustment. Nevertheless, current technologies still face challenges such as insufficient adaptability to varying material environments and limited algorithm generalization under complex working conditions. In the future, multidisciplinary collaborative innovation is required to develop hybrid algorithm models that integrate weather-resistant composite materials, intelligent adaptive sensor structures, and physical mechanisms, thereby establishing a high-precision, low-cost monitoring system and providing theoretical support for the research and development of grain loss detection equipment.