基于轮轨响应的车轮多边形阶数识别研究

Investigation on Order Identification of Wheel Polygonization Based on Wheel-Rail Response

摘要: 针对轨道车辆车轮多边形故障动态识别存在单一传感器适应性不足、易受车速变化干扰的难题，研究首先通过对比分析轨腰与轴箱传感器的安装便利性与检测信号的准确性，采用轨腰与轴箱协同互补的检测方案。其次，提出基于沿钢轨纵向布置的多组等距传感器响应冲击特征的有效信号分割和实时车速获取方案，给出车轮多边形阶数识别算法设计。最后，基于学校试验线开展低速工况车轮多边形阶数识别测试验证。实测数据表明：车轮多边形在轴箱处的冲击响应特征相较于轨腰处更显著；轮轨响应的时域冲击、频域特征频率与理论计算吻合；基于轮轨响应数据，可有效解决车速动态获取与车轮多边形阶数识别问题。

Abstract: This study targets the challenges in the dynamic identification of wheel polygonal faults in rail vehicles, particularly due to the inadequate adaptability of single sensors and their susceptibility to variations in vehicle speed. Initially, a comparative analysis of the installation convenience and detection signal accuracy between rail web and axle-box sensors is conducted, leading to the adoption of a complementary detection scheme that utilizes both sensor types. Furthermore, a method for effective signal segmentation and real-time train speed acquisition using multiple sets of equidistant sensors arranged longitudinally along the rail is proposed, along with the design of a wheel polygonal order identification algorithm. Finally, tests for identifying the polygonal order of wheels under low-speed conditions are conducted in experimental trackline of SIT. Experimental data indicate that the impact response characteristics of polygonal wheels at the axle box are more pronounced than those at the rail web; the time-domain impact responses and frequency-domain characteristic frequencies of the wheel-rail interactions are consistent with theoretical calculations. By the wheel-rail response data, effective solutions to the challenges of dynamic vehicle speed acquisition and wheel polygonal order identification can be achieved.