Simulations of micro-dynamics of soil, tool, and crop residue using the discrete element method (DEM)

土壤-工具-残留物相互作用不仅是许多农业田间作业的中心，也是农业工程中土壤动力学的基本方面之一。在现代保护性农业系统中，至少30%的农田被以前作物的作物残留物所覆盖。因此，在田间作业（如耕作和播种）中，与土壤接触的工具（如犁、圆盘、扫荡器等）与土壤和作物残留物相互作用。土壤-工具-残渣是一个复杂的系统，因为土壤是由离散颗粒的矩阵组成的，这些颗粒具有随机、异质的性质，并表现出复杂的非线性行为。这样的系统不能用传统的解析或数值方法来模拟。在提出的研究计划中，将采用一种创新的数值方法-离散元法（DEM）来建立通用的土壤-工具-残留物相互作用模型。在模型中，多个不同大小和形状的球体将代表一个土壤颗粒。粒子本构定律将以这样一种方式定义，即基本粒子可以形成聚集体，并具有摩擦、内聚、粘性和可变形的行为，就像真正的农业土壤一样（新颖性1）。粘性行为将用接触的土壤颗粒之间的键来描述。在土壤吸附工具的作用下，化学键容易断裂。模型微观特性（如颗粒弹性模量、粘结强度和颗粒摩擦系数）将被定义为与颗粒大小无关，因此它们是通用的（新颖性2）。由于DEM颗粒的微特性是不可测量的，将开发方法来校准微特性。校准将使用基本的土壤力学测试进行，因此校准的微特性是通用的，独立于土壤啮合工具（新颖性3）。在模型中，残茬粒子将由一组基本球体来表示，这些球体的排列方式反映了真实作物残茬的形状。残渣团簇将被整合到土壤颗粒中，形成一个由离散土壤和残渣颗粒混合而成的异质模型域（新颖性4）。通用模型将能够预测土壤微动力学：包括每个单独的土壤和残留物颗粒的力和位移，颗粒组合中的应力分布以及土壤接合工具表面的压力分布（新颖性5）。模型将使用实验室和现场数据进行验证。农业土壤动态行为的DEM建模是一个新兴的领域。了解土壤、残留物和工具的微动力属性将标志着土壤动力学领域的潜在突破。这将导致新的高性能土壤接合工具的出现，这些工具消耗的拖拉机功率最小，并为可持续农业中的粮食生产创造最佳的土壤和残留物条件。

Soil-tool-residue interaction is not only at the centre of many agricultural field operations, but also one of the fundamental aspects of soil dynamics in agricultural engineering. In modern conservation agriculture systems, at least 30% of an agricultural field is covered by crop residue from the previous crops. Thus, soil engaging tools (e.g. plows, discs, sweeps, etc.) interact with both soil and crop residue during a field operation, such as tillage and seeding. Soil-tool-residue is a complex system, since soil consists of a matrix of discrete particles that are of random, heterogeneous nature, and exhibit sophisticated non-linear behaviours. Such a system cannot be modelled using traditional analytical or numerical methods. In the proposed research program, an innovative numerical method, the discrete element method (DEM), will be used to develop universal soil-tool-residue interaction models. In the model, multiple spheres of various sizes and shapes will represent a soil particle. The particle constitutive laws will be defined in such a way that basic particles can form aggregates and have frictional, cohesive, viscous, and deformable behaviours like real agricultural soil (NOVELTY 1). The cohesive behaviour will be described using bonds between soil particles which are in contact. Bonds are breakable under the impact of soil engaging tools. Model microproperties (such as particle modulus of elasticity, strength of bond, and particle friction coefficient) will be defined to be independent of particle sizes, so that they are universal (NOVELTY 2). As the DEM particle microproperties are not measurable, methods will be developed to calibrate the microproperties. Calibrations will be performed using fundamental soil mechanical tests, so that the calibrated microproperties are universal and independent of the soil engaging tool (NOVELTY 3). In the model, a residue particle will be represented by a cluster of basic spheres arranged in a way to reflect the shapes of real crop residues. Residue clusters will be integrated into soil particles, forming a heterogeneous model domain with a mixture of discrete soil and residue particles (NOVELTY 4). The universal models will be able to predict soil micro dynamics: including forces and displacements of each of individual soil and residue particles, stress distribution in particle assembly, and pressure distribution on the surfaces of soil engaging tools (NOVELTY 5). Models will be validated using laboratory and field data. DEM modeling of dynamic behaviours of agricultural soil is an emerging field. Understanding of micro dynamic attributes of soil, residue, and tools will mark a potential breakthrough in the area of soil dynamics. This will lead to the emergence of new high performance soil engaging tools which consume minimum tractor power and create optimal soil and residue conditions for food production in sustainable agriculture.