Field investigation on effects of railway track geometric parameters on rail wear

Rail wear has dramatic impact on track performance, ride quality and maintenance costs. The amount of rail wear is influenced by various elements among which geometric parameters play an important role. The amount of wear in Iran’s railway lines and its imposed maintenance costs oblige us to make modifications on the various geometrical parameters. In order to ensure the effectiveness of these changes, it is necessary to investigate these parameters and their effects on the wear. This research is aimed at studying the effects of different track geometrical parameters on the vertical and lateral wear by conducting a three phase field investigation. The first phase was carried out at the switches of a station, the second phase at a straight line, and the third at a curved line out of the station. The results obtained are analyzed and the role of each track geometrical parameter in the rail wear is discussed. Recommendations for prevention or reduction of rail wear are presented.

Research on Wheel Rail Wear for a 140 t Open Type Hot Metal Car

To maintain the safety of an open-type hot-metal car and to reduce wheel-rail wear during transportation, simulation models of the main components of such car were built using Pro/E software and then tested. In particular, the Pro/E models were imported into ADAMS/Rail for assembly and then used to construct a complete hot-metal car dynamic model. Locomotive wheel-rail attack angle, wheel-rail lateral force, and wear index were used as evaluation parameters during the simulation to analyze the effects of bogie parameter, rail parameter, and speed of the hot-metal car on wheel-rail wear. An improvement scheme for reducing wheel-rail wear was proposed based on the result of the dynamic simulation, wherein wheel-rail wear and curving performance were analyzed and compared. The simulation provided an important reference for evaluating and improving the dynamic performance of the hot-metal car. The applied effect showed that the improvement scheme is effective.

Research on rail wear of small radius curve in EMU depot

Purpose–With the help of multi-body dynamics software UM,the paper uses Kik–Piotrowski model to simulate wheel-rail contact and Archard wear model for rail wear.Design/methodology/approach–The CRH5 vehicle-track coupling dynamics model is constructed for the wear study of rails of small radius curves,namely 200 and 350 m in Guangzhou East EMU Depot and those 250 and 300 m radius in Taiyuan South EMU Depot.Findings–Results show that the rail wear at the straight-circle point,the curve center point and the circlestraight point follows the order of center point>the circle-straight point>the straight-circle point.The wear on rail of small radius curves intensifies with the rise of running speed,and the wearing trend tends to fasten as the curve radius declines.The maximum rail wear of the inner rail can reach 2.29 mm,while that of the outer rail,10.11 mm.Originality/value–With the increase of the train passing number,the wear range tends to expand.The rail wear decreases with the increase of the curve radius.The dynamic response of vehicle increases with the increase of rail wear,among which the derailment coefficient is affected the most.When the number of passing vehicles reaches 1 million,the derailment coefficient exceeds the limit value,which poses a risk of derailment.

Parameters studies for rail wear in high-speed railway turnouts by unreplicated saturated factorial design

Rail wear is one of the main reasons for reducing the service life of high-speed railway turnouts in China. The rail wear characteristics of high-speed railway turnouts are influenced by a large number of input parameters of the complex train-turnout system. To reproduce the actual operation conditions of railway turnouts, random distributions of these inputs need to be considered in rail wear simulation. For a given nominal layout of the high-speed railway turnout, 19 input parameters for rail wear simulation in high-speed railway turnouts are investigated based on orthogonal design of experiment. Three dynamic responses(wheel-rail friction work, normal contact force and size of contact patch) are defined as observed values and the significant factors(direction of passage, axle load, running speed, friction coefficient, and wheel and rail profiles) are determined by two unreplicated saturated factorial design methods, including the half-normal probability plot method and Dong 93 method. As part of the associated rail wear simulation, the influence of the wear models and the local elastic deformation on the rail wear was separately investigated. The calculation results for the wear models are quite different, especially for large creep mode. The local elastic deformation has a large effect on the sliding speed and rail wear and needs to be considered in the rail wear simulation.

Influence of Angle of Attack and Lateral Force between Wheel and Rail on Wear of Rail ——a Laboratory Study

Angle of attack and lateral force are two important parameters influencing wheel rail wear. This paper deals with the question of influences of the angle of attack and the lateral force on the wear of rail. A series of experiments are conducted on 1/4 JD 1 Wheel/Rail Tribology Simulation Facility. The angles of attack selected in the tests are 0°16′30″, 0°37′40″ and 1°0′0″ respectively. The lateral forces selected in the tests are 0.694 kN, 1.250 kN and 2.083 kN, respectively corresponding to the lateral forces of 25 kN, 45 kN and 75 kN measured in the field, with the aim of keeping the same ratio of L/V between laboratory and field conditions. It is found that the larger the angle of attack is, the more serious the wear of rail is. The relation of rail wear rate versus angle of attack is non linear, and the relation of rail wear rate versus lateral force is approximately linear. The influence of angle of attack is more serious than that of lateral force. For the tractive wheelset, the wear index involving linear and quadratic function terms of angle of attack has good agreement with the limited experimental data. Some conclusions are given.

Typical wheel–rail profile change rules and matching characteristics of high speed railway in China

Purpose–In order to systematically grasp the changes and matching characteristics of wheel and rail profiles of high speed railway(HSR)in China,172 rail profile measurement points and 384 wheels of 6 high-speed electric motive unites(EMUs)were selected on 6 typical HSR lines,including Beijing–Shanghai,Wuhan–Guangzhou,Harbin–Dalian,Lanzhou–Xinjiang,Guiyang–Guangzhou and Dandong–Dalian for a two-year field test.Design/methodology/approach–Based on the measured data,the characteristics of rail and wheel wear were analyzed by mathematical statistics method.The equivalent conicity of wheel and rail matching in a wheel reprofiling cycle was analyzed by using the measured rail profile.Findings–Results showed that when the curve radius of HSR was larger than 2,495 m,the wear rate of straight line and curve rail was almost the same.For the line with annual traffic gross weight less than 11 Mt,the vertical wear of rail was less than 0.01 mm.The wear rate of the rail with the curve radius less than 800 m increased obviously.The wheel tread wear of EMUs on Harbin–Dalian line,Lanzhou–Xinjiang line and Dandong–Dalian line was relatively large,and the average wear rate of tread was about 0.05–0.06 mm$(10,000 km)
1,while that of Beijing–Shanghai line,Wuhan–Guangzhou line and Guiyang–Guangzhou line was about 0.03–0.035 mm$(10,000 km)
1.When the wear range was small,the equivalent conicity increased with the increase of wheel tread wear.When the wear range of wheel was wide,the wheel–rail contact points were evenly distributed,and the equivalent conicity did not increase obviously.Originality/value–This research proposes the distribution range of the equivalent conicity in one reprofiling cycle of various EMU trains,which provides guidance for the condition-based wheel reprofiling.

Investigation into rail corrugation in high-speed railway tracks from the viewpoint of the frictional self-excited vibration of a wheel–rail system

A finite element vibration model of a multiple wheel-rail system which consists of four wheels, one rail, and a series of sleepers is established to address the problem of rail corrugation in high-speed tracks. In the model, the creep forces between the wheels and rail are considered to be saturated and equal to the normal contact forces times the friction coefficient. The oscillation of the rail is coupled with that of wheels in the action of the saturated creep forces. When the coupling is strong, self- excited oscillation of the wheel-rail system occurs. The self-excited vibration propensity of the model is analyzed using the complex eigenvalue method. Results show that there are strong propensities of unstable self-excited vibrations whose frequencies are less than 1,200 Hz under some conditions. Preventing wheels from slipping on rails is an effective method for suppressing rail corrugation in high-speed tracks.

Study on Match Relationship between Rail and Wheel

The match relationship between rail and wheel was studied by investigating the behavior of the contact fatigue wear in rail/wheel systems.The hardnesses of samples were close or equal to that of the real rail and wheel.Meanwhile the probe of study went further into the condition match and the material match based on the hardness match.The experimental results show that the wear rate depends on the hardness ratio(H/H)between rail and wheel,and the safe value of H/Hequals 1.00-11.20.The fatigue life of materials relies on the operating conditions except hardness.The selected experimental conditions satisfy the condition match except Rheat-treated out-line.The factor H/Sis the main element effecting wears in rail steel and wheel steel.The nature of the hardness match is the microstructure match under specified operating conditions.

实测轮轨蠕滑曲线对钢轨磨耗影响分析

轮轨蠕滑曲线会影响轮轨动态相互作用,进而影响钢轨磨耗,为研究实测轮轨蠕滑曲线对钢轨磨耗的影响,首先,基于最小二乘法获得适用于Polach模型和修改FASTSIM算法的参数,模拟40~400 km/h行车速度范围内的实测蠕滑曲线;随后,在SIMPACK软件中建立车辆系统动力学模型,并通过Polach模型测得实测蠕滑曲线;最后,采用Kik-Piotrowski模型和修改的FASTSIM算法进行轮轨非赫兹滚动接触计算,并结合USFD磨耗模型预测钢轨磨耗,对比了理想与实测蠕滑曲线条件下钢轨磨耗的差异.研究表明:理想蠕滑曲线条件下钢轨磨耗深度明显大于实测蠕滑曲线下的结果,随着车辆通过次数的增加,理想条件下钢轨磨耗分布范围更大,内外轨磨耗分布范围分别为实测蠕滑曲线的1.5倍和1.3倍;摩擦系数和磨耗率显著影响钢轨磨耗大小及磨耗分布情况,故在车辆动力学仿真和钢轨磨耗计算中有必要考虑实测轮轨蠕滑曲线;形成了确定实测蠕滑曲线参数的前处理程序,可服务于车辆动力学仿真和钢轨磨耗计算,可以有效指导现场进行钢轨打磨等养护维修工作.

钢轨材料局部激光熔覆自熔性合金涂层的磨损与滚动接触疲劳行为

激光熔覆技术可用于钢轨局部损伤表面的局部修复,但局部修复钢轨材料的磨损与滚动接触疲劳损伤规律尚不清楚。通过在钢轨试样表面切除凹槽来模拟局部损伤,在凹槽处激光熔覆Ni基、Fe基和Co基自熔性合金粉末,分析修复钢轨微观组织与硬度,然后利用双轮对滚试验研究局部修复钢轨试样的磨损与滚动接触疲劳行为。结果表明,激光熔覆涂层形成了共晶与枝晶组织,Ni基涂层组织粗大、硬度较小,Fe基与Co基涂层组织尺寸较小,Fe基涂层硬度最大,Co基涂层硬度居中。相比未熔覆区域,激光熔覆区(涂层)塑性变形层厚度较小,且涂层原始硬度越高,硬化后硬度越大,但硬化率和硬化层厚度更小。未熔覆区滚动接触疲劳裂纹较长,但裂纹角度较小;熔覆区裂纹长度均有所降低,但裂纹扩展角度明显增大;熔覆区与未熔覆区结合处疲劳损伤最为严重,疲劳裂纹角度和深度均比熔覆区和未熔覆区更大。对比分析发现,Stellite21(Co基)熔覆试样摩擦因数较低,熔覆区与未熔覆区磨损深度差较小,抗滚动接触疲劳性能较好,较为适合钢轨局部损伤的激光修复。研究结果可为激光熔覆技术在钢轨局部修复上的应用与优化提供理论与技术指导。

基于GM(1,1)-IPSO-BP的重载铁路小半径曲线钢轨磨耗预测方法

为实现重载铁路小半径曲线段钢轨磨耗量的精准预测,提出一种非等间距灰色模型GM(1,1)与改进粒子群算法(IPSO)优化BP神经网络相结合的钢轨磨耗预测方法。首先,根据积分原理优化GM(1,1)非等间距模型的背景值计算方法,基于改进的模型得到实测磨耗序列的初步预测结果;然后,利用IPSO算法对BP神经网络的权值和阈值进行自动寻优,对GM(1,1)模型初步预测序列的残差进行校正;最后,将优化后的两种模型组合构建基于GM(1,1)-IPSO-BP的重载铁路小半径曲线地段钢轨磨耗量预测模型。以某重载铁路桥上半径400 m曲线为例,利用长期的磨耗监测数据进行方法的适用性分析,研究结果表明:GM(1,1)-IPSO-BP模型克服了磨耗数据的非线性、随机性特征对计算结果的影响,预测精度优于单独使用GM(1,1)、IPSO-BP模型;背景值优化后的GM(1,1)模型预测准确性更可靠;IPSO优化算法提高了BP神经网络计算的精度和速度;预测结果和实测数据之间的相对误差不大于4%;在预测区间上的绝对误差小于0.4 mm,运用该方法能够较准确地得到钢轨磨耗的发展规律。研究结果可为重载铁路小半径曲线钢轨的精准维修和科学使用提供参考。

小半径曲线上P75钢轨打磨廓形设计及应用研究

以国内某重载铁路R400 m曲线P75钢轨为研究对象,建立了轮轨接触模型和轮径差曲线描述模型,选取代表性车轮踏面和钢轨廓形,采用轮径差曲线逆向求解法优化了轮轨接触区域钢轨廓形,并依据打磨量最小化原则设计得到了曲线钢轨打磨廓形。建立了实测参数下的车辆-轨道多体动力学仿真模型,对比分析了实测廓形和设计廓形匹配下的动力学指标。按钢轨廓形设计要求,在该R400 m曲线上实施了钢轨打磨试验和轮轨动力学测试。研究结果表明,优化后轮径差曲线趋于平滑,车辆曲线通过能力得到提高,抗脱轨能力得到保持;轮轨关系得到显著改善、轮轨接触点交叉跳跃和集中问题得到缓解;轮轨横向力减小20%以上,脱轨系数减小16%以上,钢轨法向接触应力减小32%以上,轮轨蠕滑力减小26%以上;实施打磨后,轮轨横向力减小34%以上,脱轨系数减小35%以上,钢轨振动加速度减小35%以上;打磨后3个月轨面疲劳伤损未见明显发展,钢轨使用寿命由前一换轨周期的8个月延长至13.5个月。仿真分析和打磨试验验证了钢轨廓形设计方法在重载铁路小半径曲线的应用效果。

不同牵引特性城轨列车车轮磨耗与疲劳研究

针对列车频繁牵引起动对车轮磨耗与疲劳损伤产生的影响问题,在恒定牵引力的基础上,探究了不同牵引特性曲线恒功率区转折点速度对车轮磨耗与疲劳的影响规律。采用SIMPACK和MATLAB软件进行联合仿真,建立了不同牵引特性下城轨列车动力学模型和基于Archard的车轮磨耗模型,分析在牵引电机不同起动转矩下牵引特性曲线恒功率区转折点速度对城轨列车轮轨蠕滑率、车轮磨耗深度、磨耗面积以及疲劳指数的影响规律。结果表明:列车牵引起动过程中,电机起动转矩越大,轮轨蠕滑率、车轮磨耗深度和磨耗面积也越大,而疲劳指数则越小;电机起动转矩一定时,随着列车牵引特性曲线恒功率区转折点速度的增大,轮轨蠕滑率、车轮磨耗深度和磨耗面积也增大,疲劳指数则有所减小;当电机起动转矩从800 N·m增大到1400 N·m,车轮磨耗深度和磨耗面积增长率最大为6.9%和12.6%,而疲劳指数下降率最大为7.2%;当恒功率区转折点速度从50 km/h增加到80 km/h,车轮磨耗深度和磨耗面积增长率最大为4.4%和4.3%,疲劳指数下降率最大为1.9%。本文研究了车辆起动转矩与恒功率转折点速度变化对车轮磨耗与疲劳的影响规律,可为车辆牵引特性设计和电机控制以及车轮运维提供参考。

轮轨摩擦系数对低地板轻轨动力响应及车轮磨耗的影响

为了确保低地板轻轨车辆安全且平稳地运行,以某轻轨线路为依托,利用多体动力学软件UM(Universal Mechanism)建立了低地板轻轨车辆-轨道耦合动力学模型。选用LM磨耗型车轮踏面以及R50标准钢轨,以美国六级不平顺轨道谱作为线路激扰。基于Hertz及Kalker简化理论、Archard模型,进行不同摩擦系数共5个工况下车辆动力响应以及车轮磨耗变化规律的仿真分析。在此基础上进一步分析了4个不同运行里程阶段对应共96组车轮磨耗型面下轻轨车辆安全性指标的变化规律。研究了4个不同运行里程阶段对应的不同车轮磨耗型面下车辆过曲线时安全性指标随摩擦系数的变化情况。结果表明:脱轨系数、轮轴横向力、轮轨横向力、车体横向加速度受摩擦系数的影响较为显著,而轮重减载率、车体垂向加速度对摩擦系数的改变并不敏感。车轮磨耗深度随里程和摩擦系数的增加而增大,且相同工况下独立旋转车轮的磨耗情况更加严重。在车辆运行40000 km后,其轮轨横向力、轮轴横向力、脱轨系数整体呈现随里程增加而增大的规律,而轮重减载率基本不受运行里程的影响。在不同摩擦系数及运行里程的叠加影响下,轮轨横向力、轮轴横向力、脱轨系数的峰值出现的位置不同,而轮重减载率却始终处于较为稳定的状态。

不同曲线半径下轨底坡与轮径差最佳匹配研究

为了研究不同曲线半径下地铁车辆动力学性能最优的轨底坡与轮径差最佳匹配,运用SIMPACK软件建立了地铁车辆动力学仿真模型,设置9种轨底坡工况和5种轮径差工况,以轮轨横向力、轮对横移量、脱轨系数、轮重减载率和轮轨磨耗指数作为车辆动力学性能的评价指标,以4种不同C型曲线半径作为研究对象,分析45种轨底坡和轮径差耦合工况下各指标的变化规律。采用熵权法对各指标进行加权求和处理,得到不同工况下的车辆动力学性能综合得分,最终分析得出车辆在不同半径曲线下轨底坡和轮径差的最佳匹配。研究结果表明:在300,400,500,600 m这4种曲线半径下,外轨轨底坡、内轨轨底坡和轮径差最优匹配方案分别为(1/20,1/40,+1),(1/20,1/40,+1),(1/30,1/30,-1),(1/40,1/30,-1)。

列车轮缘润滑材料性能分析与应用研究

通过基础摩擦磨损试验研究了干态摩擦和3种轮缘润滑棒条件下的轮箍材料试样的摩擦磨损行为,结合扫描电镜和表面测量仪观察磨损表面形貌、损伤特征和磨痕几何形状,对3种轮缘润滑棒进行能谱分析,探究了润滑成分对轮轨材料减摩的影响,并根据试验结果进行了2个月的现场机车轮缘磨耗跟踪测量.研究结果表明施加轮缘润滑棒作为润滑介质后可降低界面摩擦系数至0.15~0.50,试样表面摩擦损伤均有不同程度降低,3种轮缘润滑棒工况相比干态摩擦工况的试样磨损率分别降低17.21%、28.58%和20.50%;润滑剂成分研究表明,石墨-二硫化钼的轮缘润滑棒能在摩擦接触表面形成延展性强、覆盖率高的致密固体润滑膜,抗磨减摩性能最优;轮缘润滑棒B在DF12机车1轴轮对安装使用后轮缘磨耗有明显减缓,第1、4轴轮对轮缘磨耗之比下降50%.

基于多目标优化策略的曲线钢轨非对称打磨设计

铁路曲线钢轨打磨一直是养护维修工作的重点之一,虽然钢轨的磨损不可避免,但适当打磨型面能有效减缓磨损。为了降低打磨去除量,改善轮轨接触性能,提高铁路系统的安全性和经济性,提出一种基于多目标协方差矩阵自适应优化策略(MO-CMA-ES)的曲线段打磨型面设计方法。首先,通过主成分分析方法(PCA),从钢轨磨耗型面中选取代表性的磨耗型面作为初始种群。接着,采用NURBS曲线拟合算法对这些型面进行曲线拟合,以建立参数化模型。随后,使用MOCMA-ES算法进行多目标优化,其中目标函数包括降低打磨去除量、轮重减载率和轮轴横向力。根据每个型面在不同目标函数下的适应度值,进行种群的进化与连续迭代,直到满足终止条件,即适应度函数值不再显著变化。最后,从最终的种群中选取目标适应度值最小的子代个体作为最优解。优化结果显示,优化型面降低了45.3%的打磨去除量,同时内外轨轨顶的打磨下降高度分别降低了0.84 mm和0.23 mm。静态轮轨接触方面,优化后的轮轨接触点分布更加均匀,且轨距角过渡区域的跳跃现象得到改善。动力学性能方面,优化型面降低了列车车体的横向加速度和轮轨横向力。此外,有效减小了轮对的横向位移,并微幅降低了轮轨最大接触应力。研究结果为曲线钢轨打磨提供了一种智能化的设计方法与参考。

车轮不圆工况下轨底坡对地铁车辆动力学性能影响及最优匹配研究

针对车轮不圆引起的车轮型面变化对轨底坡设置产生的影响,运用SIMPACK软件建立地铁车辆-轨道动力学模型,设置9种轨底坡和2种车轮不圆工况,选用直线和4种不同半径曲线,分析轨底坡耦合车轮不圆对车辆动力学性能的影响;以轮轨垂向力、轮轨横向力、脱轨系数、轮重减载率、轮对横移量和磨耗指数作为评价指标,基于熵权法建立多目标优化模型,对车辆直线运行平稳性和曲线通过能力进行评价,得到轨底坡和车轮不圆的最优匹配方案。研究表明,车轮存在不圆时,轨底坡变化对横向平稳性的影响大于垂向平稳性;车轮不圆会增加轨底坡对轮轨垂向力、轮轨横向力、轮重减载率的影响程度;存在车轮不圆时,直线线路以及R300、R400、R500和R600 m这4种曲线线路下,外/内轨最优轨底坡分别为(1/40,1/30)、(1/20,1/30)、(1/40,1/30)、(1/30,1/30)和(1/30,1/30);无车轮不圆时则为(1/40,1/30)、(1/30,1/40)、(1/30,1/30)、(1/30,1/30)和(1/40,1/30)。该研究可以为车轮维护及运营线路设计提供技术支持。

载荷对D2车轮钢滑动磨损表层微观组织和性能影响

目的研究滑动磨损条件下,载荷变化对D2车轮钢表层微观组织和磨损性能的影响。方法利用MRH-30(环块)滑动摩擦磨损试验机进行滑动磨损试验。使用金相显微镜(OM)、场发射扫描电镜(SEM)、X射线衍射仪(XRD)和场发射透射电镜(TEM)等设备,分析滑动磨损后表层微观组织、表面磨损形貌和硬度变化。结果不同载荷条件下,随着磨损时间的增加,轮轨试样磨损量逐渐增加。200 N载荷条件下,(环状)车轮试样的磨损量大于100 N车轮试样磨损量,但200 N钢轨(块状)试样的磨损量却低于100 N钢轨试样。滑动磨损后,在不同载荷条件下,车轮试样表面主要形成长条状白层和不连续的月牙状白层。在200 N载荷条件下,由于车轮试样的磨损量较高,其白层厚度和表面白层硬度都小于100 N载荷条件下的车轮试样。在200 N载荷条件下,车轮试样表面更易形成不连续的月牙状白层。由于轮轨试样表面存在微凸体,高载荷会加速车轮试样表面局部微凸体发生严重塑性变形,导致月牙状白层的形成。在白层内片状渗碳体发生明显溶解和铁素体晶粒显著细化。结论车轮表面白层对磨损性能有明显影响。滑动磨损过程中,高载荷条件下,月牙状白层和周边微观组织界面易形成裂纹,裂纹会逐渐沿着界面向内部扩展,导致不连续月牙状白层发生剥落,降低车轮试样表面的硬度,加速车轮试样磨损。另一方面,高载荷会加速无白层区域塑性变形,导致其形成棘轮失效,从而加速磨损。

基于线结构光的钢轨断面磨耗改进检测算法

经典的最近点迭代(IterativeClosest Point,ICP)配准算法常被应用于钢轨断面磨耗检测系统。当轨道检修车工作运行时,基于线结构光的钢轨断面磨耗检测系统采集的钢轨廓形数据通常会受到各种离群点噪声的干扰,导致计算得到的钢轨廓形几何形态变化较大。在实际应用中,轨道检修车钢轨断面磨耗动态检测精度要求侧磨为±0.5mm,垂磨为±1.0mm,故磨耗检测算法的设计不能因为廓形含离群点噪声而降低检测精度。因此,文章提出了一种改进的两阶段钢轨断面廓形磨耗检测算法。该算法在第一阶段首先利用钢轨廓形的特征点对进行快速初始配准,使得2个点云集具有较好的初始位姿,第二阶段采用改进的鲁棒ICP算法完成精确配准,最后计算得到钢轨断面磨耗几何参数。通过在实验室搭建钢轨断面磨耗检测系统试验平台,以试验模拟实测数据常见的轨腰段和轨底段离群点干扰,并以手工接触式磨耗仪的测量数据作为参考基准,对比经典ICP算法和文章所提出的改进算法,分析磨耗检测的精度误差和有效性,并在此基础上,对算法的检测速度进行对比分析,对改进算法的重复性测量精度进行验证,最后在地铁线路上开展轨道检修车实测验证。结果表明,文章所提的改进算法有效提高了在离群点干扰场景下钢轨断面磨耗检测的精度和速度,具有工程实用价值。