交通运输系统工程与信息 ›› 2021, Vol. 21 ›› Issue (2): 189-195.

• 系统工程理论与方法 • 上一篇    下一篇

基于设备画像的机车标签体系构建方法研究

李鑫1,史天运*2,常宝3,马小宁3,刘军3   

  1. 1. 中国铁道科学研究院,研究生部,北京 100081;2. 中国铁道科学研究院集团有限公司,北京 100081; 3. 中国铁道科学研究院集团有限公司,电子计算技术研究所,北京 100081
  • 收稿日期:2020-12-10 修回日期:2021-01-13 出版日期:2021-04-25 发布日期:2021-04-25
  • 作者简介:李鑫(1990- ),男,山西运城人,博士生。
  • 基金资助:

    中国国家铁路集团有限公司重大课题/Major Program of the China State Railway Group Co. Ltd.(K2019Z006)。

Locomotive Label System Construction Method Based on Equipment Portrait

LI Xin1 , SHI Tian-yun*2 , CHANG Bao3 , MA Xiao-ning3 , LIU Jun3   

  1. 1. Postgraduate Department, China Academy of Railway Sciences, Beijing 100081, China; 2. China Academy of Railway Sciences Corporation Limited, Beijing 100081, China; 3. Institute of Computing Technology, China Academy of Railway Sciences Corporation Limited, Beijing 100081, China
  • Received:2020-12-10 Revised:2021-01-13 Online:2021-04-25 Published:2021-04-25

摘要:

机车是铁路运输生产的重要牵引设备,通过整合利用机车各类数据,构建基于设备画像的机车标签体系,有利于客观、全面地掌握机车质量状态,实现机车的精准画像、差异化运维和精细化管理。本文通过总结机车设备画像的概念和标签技术的相关理论知识,提出满足机车质量分析、运维优化及安全决策等多个应用场景的机车设备画像3级标签体系技术架构,全面分析所包含的数据采集层、标签库层和标签应用层,详细阐释机车各级标签的内容及其生成、管理、优化和分析流程,形成机车设备画像研究方法。针对聚类这一标签的产生方式,通过改进K-means聚类算法的初始质心选取方法,提高标签获取的精度和稳定性。并在某铁路局开展机车设备画像实地应用研究,形成完整可行的机车标签体系。

关键词: 信息技术, 设备画像, 标签技术, 机车, 聚类, 大数据

Abstract:

Locomotives are important traction equipment for railway transportation. Integrating and utilizing the locomotives data can create a locomotive label system based on equipment portrait, which is helpful for grasping the quality status of locomotives and realizing accurate portrait, differentiated operation and efficient management of locomotives. This paper presents the concept of locomotive equipment portrait and the related theoretical knowledge of label technology. The technical framework of the three levels label system for locomotive equipment portrait is proposed to meet the requirements of multiple application scenarios including locomotive quality analysis, maintenance optimization and safety decision- making. The technical framework is analyzed in data acquisition layer, label library layer and label application layer. The three levels label system of locomotive and the generation, management, optimization, and analysis process of labels are explained in detail. The research method of locomotive equipment portrait is thus formed. In the clustering acquisition mode of labels, the selection method of initial centroid of Kmeans clustering algorithm is improved to enhances the accuracy and stability of clustering. The proposed method has been applied in a railway bureau to generate a complete and feasible label system for locomotive equipment portraits.

Key words: information technology, equipment portrait, label technology, locomotive, clustering algorithm, big data

中图分类号: