为研究重载铁路小半径曲线钢轨型面优化问题，以外、内轨钢轨磨耗面积及滚动圆半径差为优化目标，在考虑减缓磨耗的同时对车辆通曲能力进行优化，运用克里格理论拟合目标函数得到相应的数学模型；通过NURBS曲线对钢轨廓形进行参数化描述，结合第二代非支配遗传算法(NSGA-II)对目标函数进行求解得到优化目标对应的优化廓形，并从优化型面的磨耗性能、动力学性能、钢轨滚动接触疲劳及车轮磨耗性能等方面进行评价。结果表明：设计变量的范围对目标函数影响显著，优化设计变量范围后求解得到的最优廓形在磨耗面积和滚动圆半径差上的表现均优于初始设计变量范围对应的最优廓形；优化型面在磨耗指数、磨耗功及钢轨磨耗量方面均优于初始CN75型面，在轮对横移量等动力学性能方面也均优于初始CN75型面，可以更好保证列车运行的安全性和稳定性；在钢轨滚动接触疲劳方面，优化型面均优于初始CN75型面，其中右轨的优化效果最为显著；在车轮磨耗方面，优化型面与初始CN75型面对车轮的磨耗影响差异较小，外、内轨优化型面整体磨耗量较初始CN75型面有微幅减少，优化型面车轮磨耗性能优于初始CN75型面。

To investigate the optimization problem of steel rail profile on small radius curves in heavy haul railways, the wear area and rolling circle radius difference of the outer and inner rails are taken as the optimization objectives. While considering reducing wear, the ability of vehicles to negotiate curves is also optimized. The Kriging theory is used to fit the objective function and obtain the corresponding mathematical model. The NURBS theory is used to parameterize the rail profile, and the second-generation Non-dominated Sorting Genetic Algorithm (NSGA-II) is employed to solve the objective function and obtain the optimized rail profile. The wear performance, dynamic performance, rail rolling contact fatigue, and wheel wear performance of the optimized profile are analyzed to evaluate the optimization profile from multiple perspectives. The results show that the range of design variables significantly affects the objective function. After optimizing the range of design variables, the performance of the obtained optimal profile in terms of wear area and rolling circle radius difference is superior to that of the optimal profile corresponding to the initial range of design variables. The optimized profile outperforms the CN75 profile in terms of wear index, wear power, and rail wear volume. It also exhibits better dynamic performance, such as wheel lateral shift, compared to the CN75 profile, ensuring better safety and stability of train operation. In terms of rail rolling contact fatigue, the optimized profile is superior to the CN75 profile, with the right rail showing the most significant improvement. In terms of wheel wear, the difference in the wear impact on the wheel between the optimized profile and the CN75 profile is small. The overall wear volume of the optimized profile for the outer and inner rails is slightly reduced compared to the CN75 profile, and the wheel wear performance of the optimized profile is better than that of the initial CN75 profile during the early stage of wear.