Clay cutting is of great significance in the optimal structure and motion parameters of cutting tools in agricultural machinery. The better performance of soil tillage can be achieved to reduce the work power consumption, and cutting resistance. However, clay shear failure has posed a serious challenge to agricultural cultivation. Simulation and prediction of shear resistance are also difficult so far, due to the mechanical properties and the complex deformation of clay during shear failure. In this study, a clay shear failure model was established to accurately describe the shear failure process of clay, and then estimate the shear force using a smoothed-particle hydrodynamics (SPH) framework. A Grady-Kipp tensile damage failure model was constructed using the elastic-plastic constitutive model. A mixed failure model was obtained to couple the two models. The kernel function was chosen as a cubic spline function under the SPH framework, with a smooth length 1.3 times the particle spacing. The tensile instability was also reduced using numerical techniques, such as artificial viscosity, artificial stress, and XSPH. After that, the contact algorithm was used to calculate the cutting resistance between the tool and the clay during cutting using the Fortran language on the Visual Studio 2019 platform. The velocity, displacement, and stress distribution maps were then captured using Tecplot software. The parameters of the model were obtained via a series of geotechnical tests. Specifically, the direct shear tests were to obtain the cohesion and internal friction angle, while the ring knife method was to measure the density of clay. The cutting experiments were conducted to validate the model using boxes with acrylic panels. The results show that there was better consistency between the experiment and simulation, in terms of cutting failure deformation. The development direction of cracks was also aligned better with the shape of the cracks. The evolution of tool-cutting resistance was predicted during clay-cutting failure, according to the previous clay-tool interaction model. The cutting resistance fluctuated in the early stage, and then stabilized near a certain value in the later stage, as the displacement increased. The reason was that there was a gap between clay particles at the beginning of cutting, leading to initial instability. As the cutting process proceeded, the particles were rearranged into a more compact state, indicating a gradually stable curve. A systematic analysis was also implemented to determine the influence of cutting depth on the failure morphology and cutting resistance during clay cutting. It was found that the larger the cutting depth was, the larger the blocks formed during clay crushing, and the greater the cutting resistance was. At the same time, there was also some influence of the tool friction coefficient on the cutting resistance. The smallest cutting resistance was then achieved when the friction coefficient was 0.1. Therefore, the mixed failure model can be expected to better accurately describe the clay cutting failure, the propagation of cracks, as well as the formation of clay fragments and granular particles during clay cutting, compared with the shear. This finding can provide a strong reference for the numerical simulation of clay cutting.

Agricultural operations (such as tillage, soil turning, and deep loosening) significantly affect the soil structure, aeration, and nutrient distribution in crop production. The performance of soil-engaging components during operations can directly impact the agricultural machinery and the quality of soil preparation. Traditional mechanical analysis has been limited to capture the micro-movements of soil particles and the dynamic interactions in three-dimensional space. It is very necessary to accurately assess the tillage in the optimization of soil-engaging component. In particular, the SPH method can also offer a mesh-free approach. The complex behavior of soil can be expected to effectively simulate, including large deformations and interactions with mechanical components. In this study, the more accurate and comprehensive model was developed for the soil-engaging component interactions using SPH method. A three-dimensional contact algorithm was also incorporated using soil elastoplastic constitutive theory and penalty functions. The elastoplastic constitutive model was accurately represented the behavior of soil under various stress conditions, including elastic deformation and plastic yielding. The contact algorithm included the virtual planes and penalty functions. There was the complex contact behavior between soil particles and soil-engaging components. The algorithm was implemented in C++ on the CUDA (Compute Unified Device Architecture) parallel computing platform, thus leveraging the power of GPU (Graphics Processing Unit) acceleration. Numerical simulation was conducted using Visual Studio 2019. A series of experiments were also conducted to validate the accuracy of the improved model. The sand collapse experiment was firstly conducted to create a controlled environment, where the sand was allowed to collapse under its own weight. The diameter of collapse extension was theoretically calculated using empirical formulas. A comparison was also made on the simulation and the observed one. Theoretical diameter of collapse extension was 0.648 m with a relative error of only 2.16%. There was the close match between the simulation and experiment. The improved model was accurately predicted the soil behavior under large deformations. After that, the small-scale soil cutting apparatus was designed and then fabricated in the sand collapse experiment. The cutting forces were precisely measured under various conditions, including different tillage depths, speeds, and soil-engaging component angles. The predictive accuracy of the improved model was assessed to compare the numerical simulations with the experimental data. The model was further validated to predict the cutting forces for the soil cutting morphology. There was the high degree of correlation between the simulation and experiment. The patterns of sand cutting resistance were observed in numerical simulations and physical experiments, when the soil-engaging component. The combination of initial parameters was set as the depth of 0.06 m, a cutting speed of 0.01 m/s, and a penetration angle of 60°. The soil cutting forces were accurately predicted to represent the behavior of soil during cutting. The relative error was found to be only 7.02% after prediction, which was within an acceptable range for practical applications. Lastly, the systematic investigation was also implemented to explore the impact of various tillage depths (0.04, 0.06, and 0.08 m), tillage speeds (0.01, 0.02, and 0.03 m/s), and soil contact angles (45° and 60°) on the cutting force exerted by soil-engaging components. The tillage depth was the predominant influencing factor on the cutting force of soil-engaging components. There was the significant increase in the cutting force, as the tillage depth increased. There was the relatively minor influence of tillage speed and soil contact angle on the cutting force. This research can offer a solid scientific and technical support to optimize the agricultural machinery and tillage efficiency.