Simulation test of end-effector operation

The interaction between shovel-tooth and connecting rod involves multi-body kinematics, and the interaction between shovel-tooth and soil involves multi-body kinematics and discrete element theory. Therefore, MBD-DEM coupling method is used for the analyses. This paper used the Adams-EDEM coupling simulation method to simulate the operational state of the shovel-tooth end-effector. On this basis, a simulation model of the end-effector and soil was established to simulate the interaction between end-effector and soil during the operation of the test bench.

Adams software is capable of simulating the motion of end-effector, and the motion characteristics can be analyzed through its post-processing interface.

The end-effector model was established and assembled in Creo 3D software, exported as step format, and then imported into Adams View environment for dynamic simulation. The component material was set to be steel, forming the center of mass. The model was simplified and the fasteners were removed to reduce the number of unnecessary constraints in the simulation process to improve the simulation efficiency. Constraints needed to be added to each mechanism, and the constraint Settings for each component were shown in Table 1. A moving drive was added between the support plate and the earth to simulate the process of end-effector insertion and excavation. A moving drive was added between the end-effector servo electric cylinder and the push rod on the end actuator servo electric cylinder. By moving the push rod of the electric cylinder up and down, the folding and stretching motion of the shovel-tooth can be driven by the connecting rod. A moving drive was added between the end-effector servo electric cylinder and the electric push rod, and the motion of the shovel-tooth could be driven by the up and down movement of the push rod of the electric cylinder.

The 3D model of the shovel-tooth end-effector was imported into EDEM. The soil particle bed, measuring 800mm×600mm×300mm, was established 100mm below the end-effector equipped with a shovel-tooth. All particles were single ball particles with a size of 8mm. Because the adhesion force and friction force between rape roots and soil were much smaller than the force of soil extrusion, the discrete element model was simplified and the rape plant model is not established.

In order to better simulate the interaction between end-effector and soil, it is necessary to select an appropriate discrete element contact model and determine appropriate model parameters [22–24]. Researchers at the University of Edinburgh proposed an Edinburgh Elasto-Plastic Adhesion (EEPA) model. This model included the plasticity and viscosity of soil particles, making it suitable for simulating soils with strong plasticity. Hertz-Mindlin with JKR contact model was a cohesive contact model suitable for simulating wet viscous soil. This model was consistent with the effect of end-effector on soil moisture and viscosity during operation. In this paper, the EEPA model was selected as the soil-soil contact model, and the Hertz-Mindlin with JKR model was selected as the soil-end-effector contact model. By consulting the literature [25], we determined the contact parameters and fundamental physical parameters, as shown in Table 2.

During the MBD-DEM coupling simulation test, the randomness of the simulation test was reflected by regenerating the soil particle bed before each test. Configure the *.acf file and *.cosim file while ensuring that the coupling interface between Adams and EDEM was open. Output the *.adm file through Adams and modify the environment variables in the *.adm file. Load *.cosim file in Adams Co-simulation to realize two-way real-time data transfer of Adams-EDEM. The time step of EDEM was 0.0004 s, and the target save time was 0.01 s. The DEM-MBD coupling simulation model was established, as shown in Fig 4.