基于多时域特征量的轨道不平顺状态综合评估

准确评估轨道不平顺状态对保障列车安全运营具有重要意义。目前,针对线路状态评估的指标主要采用轨道质量指数(TQI),但在实际管理中发现,该方法可能会造成轨道的欠维修或过维修。为了弥补现有评估方法的不足,充分利用采集的大量轨检数据,提出15个时域特征量对TQI进行补充,并利用主成分分析法(PCA)对数据进行降维处理,大大提升了此方法的时效性。以某高速铁路实测数据为应用实例,给选定的特征量99%的置信概率,结合动力学仿真和时频分析方法,综合评估该线路的轨道状态。结果表明,同一里程位置处的不同指标分布情况存在明显差异,TQI满足规范要求的轨道区段其动力学指标仍存在超限情况。本文方法可以实现轨道区段的潜在病害识别,有利于工务部门完成对轨道状态更为科学严谨的监测与管理。

Comprehensive Evaluation of Track Irregularity Based on Multiple Time-domain Feature Quantities

Accurately assessing the track irregularity is of great significance for ensuring the safe operation of trains. At present, the track quality index(TQI) is mainly used for the evaluation of the line state, but it is found in actual management that the method may cause under-maintenance or over-maintenance of the track. In order to make up for the shortcomings of existing evaluation methods, this paper makes full use of the massive data collected by rail inspection vehicle, proposes 15 time-domain feature quantities to supplement TQI, and greatly improves the timeliness of this method by using Principal Component Analysis(PCA) to reduce the dimensionality of data. Taking the measured data of a high-speed railway as an example of application, the paper selects a 99% confidence probability for the feature quantities and combines with the dynamics simulation and time-frequency analysis methods to comprehensively evaluate the track conditions of the line. The results show that there are significant differences in the distribution of different indicators at the same mileage position, and the dynamic responses of some track sections that meet the requirements of the TQI are still over limit. The method in this paper can identify potential defects in the track section, which is beneficial for the work departments to fulfill more scientific and rigorous monitoring and management of track conditions.