基于大数据和动力学模型的车轮磨耗预测研究
Research on Wheel Wear Prediction Based on Big Data and Dynamic Model

为了研究车轮磨耗的特点和演变过程，对机车车轮的磨耗进行了大数据分析研究发现：轮径及轮缘厚度的磨耗率随轮径值及轮缘厚度的降低呈先减小后增大的趋势、且新轮状态下车轮踏面的磨耗率约为轮缘磨耗率的三倍左右；新轮状态下车轮踏面磨耗较为明显，一定运行里程后轮缘磨耗更为突出。建立了机车动力学模型、轮轨滚动接触模型、材料磨损模型一体的车轮磨耗计算模型，并使用实测车轮数据与优化算法相结合的方式来对车轮磨耗计算模型中磨耗系数k进行优化，计算发现：磨耗系数取平均值的车轮磨耗计算结果与实测值误差较大，而取优化值的计算结果与实测值的误差较小在3%~13%之间(车轮磨耗集中在？45~40 mm，磨耗最大位置在？10~？5 mm之间)。
In order to study the characteristics and evolution process of wheel wear, the big data analysis of locomotive wheel wear was carried out, and it was found that: The abrasion rate of wheel diameter and rim thickness decreases first and then increases with the decrease of wheel diameter and rim thickness, and the wear rate of wheel tread is about three times that of the rim wear rate under the new wheel state; The wear of the wheel tread is more obvious in the new wheel state, and the wear of the wheel rim is more prominent after a certain mileage of operation. A wheel wear calculation model integrating the locomotive dynamics model, wheel-rail rolling contact model and material wear model was established. The wear coefficient k in the wheel wear calculation model is optimized by combining the measured wheel data and the optimization algorithm. The results show that the error between the calculated results of wheel wear and the measured value with the average value of the wear coefficient is larger, while the error between the calculated result and the measured value with the optimized value is smaller between 3%~13% (the wheel wear is concentrated in ？45~40 mm, and the maximum wear position is between ？10~？5 mm).