Introduction The formation of ruts induced by vehicle traffic poses a significant challenge for agricultural soils due to soil compaction both at the surface and deeper layers. This phenomenon compromises vehicle performance increases energy consumption, and leads to long-term environmental degradation, such as soil erosion and fertility reduction. To enhance vehicle performance and reduce soil damage, it is crucial to accurately predict how factors such as vehicle speed, vertical load, and the number of passes impact rut depth. The findings of this study hold significant practical implications, facilitating the development for the creation of more efficient agricultural practices, while simultaneously minimizing environmental impact. The complexity of these interactions necessitates using machine learning models, especially artificial neural networks (ANNs), to predict rut depth based on input parameters. In this study, two machine learning models, namely the multilayer perceptron (MLP) and the radial basis function (RBF) networks, were employed to predict rut depth. Materials and Methods Experiments were conducted using a soil bin that allows for precise control of independent parameters, measuring 24 meters in length, 2 meters in width, and 0.8 meters in depth. The soil used was agricultural soil, comprising 35% sand, 22% silt, and 43% clay, with a moisture content of 8%. The tests included three independent parameters: vertical load (2, 3, and 4 kN), forward speed (1, 2, and 3 km h(-1)), and number of wheel passes (up to 15). Two types of traction devices, including a rubber wheel and a track wheel, were tested. A caliper was used to measure the rut depth after each pass with an accuracy of 0.02 mm. The data collected from soil bin tests were used to train neural network models in MATLAB 2021-b software. The MLP model had a topology with two hidden layers and included three inputs and one output. In the RBF model, the network topology had a single hidden layer. The trial-and-error method was used to adjust the hyperparameters of the neural networks, including the number of neurons in the hidden layers, the learning rate, and momentum for the MLP network, as well as the spread rate and regularization rate for the RBF network. Results and Discussion Experimental data confirmed that increasing the vertical load and the number of passes resulted in deeper ruts. Conversely, an increase in speed led to a reduction in rut depth, particularly during the initial pass. Both artificial neural network (ANN) models accurately predicted rut depth, with the multilayer perceptron (MLP) neural network outperforming the radial basis function (RBF) neural network. Specifically, the root mean square error (RMSE) for the optimal MLP model, which utilized a learning rate of 0.001 and a momentum of 0.67, was 0.10. In contrast, the optimal RBF model, with an expansion rate of 0.23456, yielded an RMSE of 0.12. The findings indicate that the MLP artificial neural network model surpasses the RBF neural network model in terms of accuracy and overall performance. However, the RBF neural network exhibits a faster response time, making it particularly suitable for real-time applications. Conclusion This study demonstrates the efficacy of machine learning techniques, particularly artificial neural networks (ANNs), in predicting rut depth caused by off-road vehicle traffic. Both multilayer perceptron (MLP) and radial basis function (RBF) neural networks exhibited robust predictive capabilities, with the MLP model providing slightly superior accuracy and the RBF model offering better computational efficiency. These findings highlight the potential of machine learning in modeling complex interactions between soil and vehicles, which can enhance vehicle performance, mitigate soil erosion, and guide the design of off-road vehicles. Future research directions could include investigating additional soil parameters, various vehicle configurations, and the real-world implementation of autonomous off-road vehicles to promote more environmentally sustainable operations.

One of the critical steps in the root crop harvesting process is screening tubers from soil. However, low screening efficiency seriously hinders the rapid development of the root crop industry. Clarifying the tuber-soil mixture separation behaviour and establishing the connection between vibration, airflow parameters, and separation index (SI) is critical to increasing screening efficiency. Corydalis Yanhusuo is employed as the research object, and the three-dimensional scale distribution and mechanical properties of tubers and soil particles are first counted. Then, a vibration and airflow coupling separation model of the tuber-soil mixture was constructed using the computational fluid dynamics and discrete element method (CFD-DEM) coupling method, and the physical parameters in the model were calibrated. A new method for calculating the SI is proposed. The relationship between vibration amplitude, frequency, airflow velocity, SI, and separation velocity was analysed. Simultaneously, the porosity change in the particle group during the separation process was investigated, and the relationship between vibration, frequency, and airflow velocity on the separation dynamics of binary mixtures was revealed by utilising data visualisation and frequency domain analysis. The platform for the vibration and airflow separation physical test was built. The separation behaviour of mixed particles in various parameters was discussed, as was the feasibility and accuracy of the numerical simulation results. The results of this study can provide theoretical support for the efficient screening of tuber-soil mixtures and further promote the rapid development of the root industry.

Machinery traffic is associated with the application of stress onto the soil surface and is the main reason for agricultural soil compaction. Currently, probes are used for studying the stress propagation in soil and measuring soil stress. However, because of the physical presence of a probe, the measured stress may differ from the actual stress, i.e. the stress induced in the soil under machinery traffic in the absence of a probe. Hence, we need to model the soil -stress probe interaction to study the difference in stress caused by the probe under varying loading geometries, loading time, depth, and soil properties to find correction factors for probe -measured stress. This study aims to simulate the soil -stress probe interaction under a moving rigid wheel using finite element method (FEM) to investigate the agreement between the simulated with -probe stress and the experimental measurements and to compare the resulting ratio of with/without probe stress with previous studies. The soil was modeled as an elastic -perfectly plastic material whose properties were calibrated with the simulation of cone penetration and wheel sinkage into the soil. The results showed an average 28% overestimation of FEM-simulated probe stress as compared to the experimental stress measured under the wheel loadings of 600 and 1,200 N. The average simulated ratio of with/without probe stress was found to be 1.22 for the two tests which is significantly smaller than that of plate sinkage loading (1.9). The simulation of wheel speed on soil stress showed a minor increase in stress. The stress over -estimation ratio (i.e. the ratio of with/without probe stress) noticeably increased with depth but increased slightly with speed for depths below 0.2 m.

Introduction. The discrete element method is the most promising method for modeling soil tillage. With the use of DEM modeling it is possible to create a digital twin for technological process of interaction of tools with soil, analyze the operation of tillage and seeding machines having various design and technological parameters, and predict energy and agrotechnical indicators of etheir work. To improve the prediction accuracy, it is necessary to compare the obtained data with the results of field experiments. Aim of the Study. The study is aimed at developing a digital twin of the tillage bin through using the discrete element method and optimizing the main design and technological parameters of the dual -level opener. Materials and Methods. To simulate the process of the soil -opener interaction, there was used the discrete element method; the advanced Hertz - Mindlin model was used as a contact model. For DEM modeling there is created a virtual tillage bin, which is filled with spherical particles of 10 mm diameter with the specified rheological parameters of the selected contact model. The main design factors for carrying out computer experiments in order to optimize them were the opener blade rake angle alpha and the furrow rake angle beta . The opener traction resistance R was chosen as the output optimization parameter. Results. Implementation of multifactor experiments on the digital twin of the soil bin in the Rocky DEM program allowed to optimize the design and technological parameters of the dual -level opener: bit inclination angle alpha = 75(o ), furrow former inclination angle beta = 21(o) , vertical distance between the bit and furrow former Delta a = 11 - 14 mm. Discussion and Conclusion. As a result of the modeling, it was found that the opener traction resistance increases in quadratic dependence on the opener operating speed and surface energy of the contact model.