Parameter identification of DEM-FEM coupling model to simulate traction behavior of tire-soil interaction

Abstract

This paper presents a novel parameter identification method for DEM-FEM coupling model to investigate the trafficability of off-road tires on granular soils. Initially, an experimental device is developed to measure the bulk responses of granular materials i.e., angles of repose and shear. A series of numerical tests, including the Plackett-Burman tests, steepest-climbing tests and three-factor orthogonal tests, are then performed to formulate the mathematical regression and constraint equations. These equations establish the correlation between the three key model input parameters (namely, coefficients of static friction of acrylic wall-particle and particles, and coefficient of restitution of acrylic wall-particle) and the aforementioned bulk responses. After that, the non-dominated sorting genetic algorithm II (NSGA-II) is implemented to iteratively calculate the equations based on the multi-objective optimization method to obtain the optimal solution set. Finally, the effectiveness and feasibility of the parameter identification method are confirmed by comparing the results of indoor soil-bin tests and the corresponding numerical simulations in terms of tire sinkage, ruts and soil deformation and flow.