铁道建筑

为进一步推行重载铁路线路设备单元维修，提高线路设备养护维修效率，在重载铁路线路设备综合状态评价基础上，结合线路设备养护维修能力、大机维修的干扰，提出了线路设备单元维修决策方法；对比分析了不同聚类分析方法的特点，提出适用于线路设备单元维修的聚类分析方法。根据某重载铁路月度单元状态评价结果，提出了下月计划维修单元，并与人工制定的月度维修计划进行对比，验证该方法的合理性和可行性。结果表明：K-Means可以根据天窗数量确定簇的个数，制订维修计划时宜采用K-Means聚类分析方法；采用本文方法进行单元维修后，上行线、下行线的轨道质量指数（Track Quality Index,TQI）较上年同期分别降低了0.66、0.38 mm，明显提升了线路设备质量。

In order to further promote the unit maintenance of heavy haul railway line equipment and improve the efficiency of line equipment maintenance and repair, a decision making method for line equipment unit maintenance was proposed based on the comprehensive status evaluation of heavy haul railway line equipment, by combining with the maintenance and repair capabilities of line equipment and the interference of large machinery maintenance. A comparative analysis was conducted on the characteristics of different clustering analysis methods, and a clustering analysis method suitable for the maintenance of line equipment units was proposed. Based on the monthly unit status evaluation results of a certain heavy haul railway, a planned maintenance unit for the next month was proposed, and compared with the manually formulated monthly maintenance plan to verify the rationality and feasibility of this method. The results indicate that K-Means can determine the number of clusters and develop maintenance plans based on the number of skylights, and it is recommended to use K-Means clustering analysis method. After using the method described in this article for unit maintenance, the track quality index(TQI) of the up and down lines decreased by 0.66 mm and 0.38 mm respectively compared to the same period of last year, significantly improving the quality of the line equipment.