摘要: |
Simulation results of truck dynamic loading based on measured road profiles are presented. Two truck configurations--two-axle straight trucks and five-axle tractor semitrailers--are used. The sprung mass (tractor and semitrailer) is modeled with bounce, roll, and pitch motions, and the unsprong mass (axle) has bounce and roll degrees of freedom. The suspension element is treated as a spring rate, a Coulomb friction, and a viscous damping coefficient. Tires are considered as linear springs with damping coefficients. Equations of motion were formulated from the separated-form virtual work principle, which automatically eliminates redundant coordinates. Computer programs developed on a personal computer include three portions: nonlinear integration, statistical analysis, and graphics. Runge-Kutta-Fehlberg's fourth-fifth-order algorithm with self-adjustable step sizes was adopted to solve the nonlinear truck equations. Numerical and graphical outputs can be provided in both time domain and statistical forms. Pavement profiles of both paths were measured by noncontact profilometers of the Ohio Department of Transportation and used as inputs to the program. Three levels of road roughness for each of the concrete and asphalt pavements, as well as three vehicle forward speeds, were used in the simulation. The results show that the dynamic force can deviate significantly from the static loading. A statistical analysis indicates that the axle loading may be described by a Gaussian (normal) distribution. The standard deviation (or dynamic load coefficient) is affected by the vehicle forward speed and the pavement roughness. |