Abstract: The accurate prediction of frictional temperature during wheel-rail sliding contact is of great significance to the studies for wear and fatigue performance of wheel-rail system. Current analytical or semi-analytical models of wheel-rail frictional temperature usually employ the elliptical distribution of contact pressure in the Hertz elastic contact theory and a single temperature-dependent material property, which differs from the actual heat-transfer state in many wheel-rail sliding contact conditions. Therefore, introducing the plastic correction of contact pressure and the temperature dependence of various properties in the calculation model of wheel-rail sliding temperature rise simultaneously could be great helpful to improve the accuracy of prediction result. Based on the elasto-plastic contact theory, considering the temperature dependence of thermal conductivity, specific heat capacity and friction coefficient simultaneously, the integration of thermal conductivity with respect to temperature is set as the quantity that needs to be solved by applying the Kirchhoff transformation method, and the nonlinear Fourier heat conduction equation is transformed into a corresponding simple partial differential equation with single variable coefficient, a unified implicit difference scheme with arbitrary form of temperature dependence is derived, and the influences of convection coefficient, vertical load, creepage and train speed on the temperature rise over rail surface are discussed, respectively. Results show that the convection coefficient has little effect on the temperature rise in the high-speed condition; the increase of creepage and train speed can enhance the friction power, and thus causes the increase of frictional temperature; and the maximum temperature rise also increases approximately linearly with the increasing vertical load. In addition, considering the temperature dependence of various thermophysical properties in the calculation model of temperature rise induced by the wheel-rail sliding contact can effectively avoid the overestimation of temperature rise, and the temperature-dependent friction coefficient has a more remarkable effect on the prediction result than thermal conductivity and specific heat capacity.