Considering the problems of high tillage resistance, high energy consumption, and failure to break the plow pan in subsoiling caused by the high viscosity of tea garden soil, a new design method of a subsoiler underlying the subsoiling mechanism based on structural bionics and the discrete element method is proposed. Taking the largest toe of mole cricket's forefoot as the bionic object, the biological information of the contour of mole cricket's toe was extracted using the methods of image processing and reverse engineering, and the edges of its outer and inner contours were obtained. Combined with the structure of an existing subsoiler, initial point cloud data of a bionic subsoiling shovel were mapped by the methods of proportional amplification and rotation. Taking the minimum sum of outer and inner contour fitting errors as the optimal objective function and selecting high-order polynomial as the fitting function, the optimization model of the outer and inner contour point cloud fitting function was established. The optimal fitting function of the outer and inner contour point cloud was obtained, and the bionic subsoiler model was established based on these two optimal functions. In the study of the discrete element, four soil particle models were proposed based on the actual shape of the cohesive soil of a tea garden. Based on the principle of discrete elements and the agronomic requirements, a modeling method of the plow pan soil was proposed. First, the adjacent soil particles at the bottom were connected by bonding, and the soil with large porosity was squeezed by the preset force to form the plow pan soil. Then, six tillage layers were added above the plow to form a soil structure model of the tea garden. The bionic subsoiler was assembled into a four-bar subsoiling mechanism. The process of the subsoiling mechanism acting on the soil was analyzed by the co-simulation of ADAMS-EDEM and compared with the common subsoiler. The results showed that the bionic subsoiler, compared with the common subsoiler, reduced the tillage force in the horizontal direction by 16.34% and in the vertical direction by 24.53%, reduced energy consumption by 9.64%, increased the damage to the plow pan by 5.10%, and caused greater disturbance to the internal soil. The influence of different driving speeds on the tillage performance was studied, which provides a theoretical reference for the better selection of the working speed for subsoiling.

针对茶园土黏度高导致深松时耕作阻力大、能耗大、犁盘不破等问题，提出了一种基于结构仿生学和离散元的深松机制下深松机设计新方法。提出了方法。以蝼蛄前足的最大脚趾为仿生对象，采用图像处理和逆向工程的方法提取蝼蛄脚趾轮廓的生物信息，得到其外轮廓和内轮廓的边缘。结合现有深松机的结构，采用比例放大和旋转的方法绘制仿生深松铲的初始点云数据。以内外轮廓拟合误差之和最小为最优目标函数，选择高阶多项式作为拟合函数，建立了内外轮廓点云拟合函数的优化模型。得到外轮廓点云和内轮廓点云的最优拟合函数，并基于这两个最优函数建立仿生深松模型。在离散元的研究中，根据茶园粘性土的实际形状，提出了四种土壤颗粒模型。根据离散元原理和农艺要求，提出了一种犁盘土的建模方法。首先，底部相邻的土壤颗粒通过粘合连接，孔隙率大的土壤被预先设定的力挤压成犁盘土。然后，在犁上方增加六层耕作，形成茶园土壤结构模型。仿生深松机被组装成一个四杆深松机构。通过ADAMS-EDEM联合仿真分析了深松机制作用于土壤的过程，并与普通深松机进行了比较。结果表明，仿生深松机与普通深松机相比，水平方向耕作力降低16.34%，垂直方向降低24.53%，能耗降低9.64%，犁盘损伤增加5.10%。 %，并对内部土壤造成较大的干扰。研究了不同行驶速度对耕作性能的影响，