Parameters of a discrete element ballasted bed model based on a response surface method

Numerical and experimental investigations of the effect of temperature and moisture on the repose angle of railway ballast

“…), and the motor slowly lifted the cylinder at a speed of 10 mm/s until the ballast rolled out from the bottom of the cylinder and stabilized. 13 Thus, the ballast, in a natural stacking state, could be obtained. Then the ballast particles scattered outside the accumulation body were removed, and all of the ballast particles participating in forming the accumulation body were retained.…”

“…, n). According to Xiao et al, 13 n ¼ 4 was used for the tests in this work, as shown in Figure 1(c).…”

“…The particles' density was 2600 kg/m 3 , the Young's modulus was 5.45 Â 1010 Pa, and Poisson's ratio was 0.24. 13 The cylinder in the simulation was lifted at a uniform speed of 10 mm/s, which is the same as the laboratory tests, and the process is shown in Figure 4(c). The AOR of the ballast in the DEM was measured by the method shown in Figure 1(c).…”

“…When modeling ballast beds with the DEM, determining how to reasonably select model parameters becomes a key and basic task. 13 It is an intuitive and widely used technique to calibrate the DEM parameters of a ballast bed based on physical test results (e.g. direct shear testing, triaxial and biaxial testing).…”

“…The theoretical predicted and experimentally measured values fitted well. Xiao et al 13 introduced the response surface method into the calibration test of ballast AOR and found that the average AOR of China's special-grade gravel ballast under normal conditions was 36.10 AE3.62 . They also found out the optimal combination of contact parameters; that is, the restitution coefficient ¼ 0.72, the static friction coefficient ¼ 0.56, and the rolling friction coefficient ¼ 0.27.…”