To accurately grasp the evolution characteristics and wear patterns of the wheel tread in the whole life cycle of high-speed train wheels, it is necessary to study an effective wheel profile prediction model. Aiming at the problem of large computation amount and long time of the traditional vehicle-track dynamics model in calculating wheel wear, this paper provides a simplified vehicle-track coupled dynamics model. Based on this, this paper studies the wheel-rail contact calculation model, and combines the dynamics model, the contact calculation model and the Archard wear calculation model to calculate the wear amount of the cell. Then, this paper uses the cubic spline interpolation calculation and the five-point cubic smoothing method to superimpose the wear amount of the cell on the wheel profile. In this process, this paper innovatively applies the method of mathematical statistics to the establishment of the relationship between the wheel rolling contact position and the lateral displacement, and discusses the distribution rules of the lateral displacement and the creepage rate. After establishing a relatively complete wheel wear prediction simulation model, this paper analyzes the effects of different working conditions such as vehicle running speed, line condition (straight line length and curve radius and length) and wheel material (hardness) on wheel wear.