考虑磨耗演化的高速列车车轮型面多目标优化  

Multi-objective optimization of high-speed train wheel profile considering wear evolution

作　　者：周骏[1,2] 刘林芽 李纪阳[1] ZHOU Jun;LIU Linya;LI Jiyang(State Key Laboratory of Performance MonitoringProtecting of Rail Transit Infrastructure,East China Jiaotong University,Nanchang 330013,China;Jiangxi Industrial Polytechnic College,Nanchang 330096,China;Pingxiang University,Pingxiang 337055,China)

机构地区：[1]华东交通大学轨道交通基础设施性能监测与保障国家重点实验室,江西南昌330013 [2]江西工业职业技术学院,江西南昌330096 [3]萍乡学院,江西萍乡337055

出　　处：《铁道科学与工程学报》2025年第1期51-65,共15页Journal of Railway Science and Engineering

基　　金：国家重点研发计划资助项目(2022YFB2062200);国家自然科学基金资助项目(52272348,51968025);江西省自然科学基金重点资助项目(20192ACBL20009);江西省交通运输厅重点项目(2022Z0003)。

摘　　要：高速铁路车轮的磨耗性能对于车轮型面的设计和性能评估至关重要。微小的型面变化可能会导致轮轨接触几何特性发生重大改变,合理的轮轨型面匹配能够改善轮轨磨耗,提高车辆性能和脱轨安全性。提出一种考虑磨耗演化的车轮型面优化方法。首先,通过追踪测试不同运营时期的磨耗型面,确定车轮磨耗的主要范围作为优化区域。然后,结合非均匀有理B样条(NURBS)拟合算法对车轮型面进行曲线插值拟合,构建了车轮型面的参数化模型。接着,将考虑车轮磨耗演化的预测模型引入到多目标函数中,同时引入了目标函数自适应权重调整因子,建立了车轮型面的多目标优化模型。采用拉丁超立方采样方法建立RBF代理模型进行模拟仿真计算,并利用TS-NSGA-II多目标优化算法进行车轮型面的优化设计。最后,对优化型面LMA-opt与LMA型面进行对比验证分析,包括车轮磨耗演化分析、轮轨静接触分析、轮轨动态接触分析和车辆动力学性能分析4个方面。结果表明,优化后的车轮型面磨耗得到显著改善,具备更好的静态几何接触性能和动态特性。这有助于提升车辆曲线通过性能,并在镟修周期内有效降低车轮的磨耗量和疲劳损伤,从而延长车轮使用寿命,提高车辆的可靠性和安全性。这项研究为车轮型面的优化提供了一种智能化的设计方法和参考。The wear performance of high-speed railroad wheels is crucial for the design and performance evaluation of wheel profiles.Small profile changes can lead to significant changes in wheel-rail contact geometry,and reasonable wheel-rail profile matching can improve wheel-rail wear,vehicle performance,and derailment safety.In this paper,a wheel profile optimization method considering the evolution of wear was proposed.First,the main range of wheel wear was determined as the optimization region by tracking and testing the wear profiles in different operation periods.Then,a parametric model of the wheel profile was constructed by fitting the curve interpolation to the wheel profile with the non-uniform rational B-spline(NURBS)fitting algorithm.Then,the prediction model considering the evolution of wheel wear was introduced into the multi-objective function.The objective function adaptive weight adjustment factor coefficients were also introduced to establish the multi-objective optimization model of the wheel profile.The Latin hypercubic sampling method was used to establish the RBF agent model for simulation calculation,and the optimization design of the wheel profile was carried out by using the TS-NSGA-II multi-objective optimization algorithm.Finally,the optimized profile LMA-opt was compared with the LMA profile for validation analysis,including four aspects:wheel wear evolution analysis,wheel-rail static contact analysis,wheel-rail dynamic contact analysis,and vehicle dynamic performance analysis.The results show that the wear of the optimized wheel profile is significantly improved,with better static geometric contact performance and dynamic characteristics.The results can enhance the vehicle’s curve passing performance and effectively reduces wheel wear and fatigue damage during the turning cycle,thereby prolonging the wheel’s service life and improving the vehicle’s reliability and safety.This study can provide an intelligent design method and serves as a reference for wheel profile optimization.

关 键 词：轮轨磨耗 滚动接触疲劳损伤 多目标优化 车辆动力学 代理模型 

分 类 号：U270.33[机械工程—车辆工程]