Research And Design Of Agile Subway Maintenance Support System Based On Big Data

As the number and size of subway lines continue to grow,the rail industry is facing more and more new challenges,such as the uneven distribution of staff,the rapid changes in related information technology,the varying needs of multiple lines,increasing passenger traffic,diverse equipment and equipment,and various other emergencies,etc.Improving the reliability,availability,Maintainability and Safety are urgent.The large scale of operation and complex equipment system,as well as the subsequent upcoming medium and major repairs and equipment renovation and renewal tasks,have increased the pressure and burden of metro O&M management.At present,most of China’s subway operation and maintenance mode is the traditional maintenance support mode,which has the following four characteristics: 1.more fault repair,regular repair and less foreseeable repair;2.a large number of human resources,organization and management efficiency is low;3.equipment and maintenance data is not detailed and comprehensive enough to meet the needs of efficient operation and maintenance;4.lack of system platform and intelligent application for processing and analyzing big data of equipment and facilities.This maintenance support model has affected the existing service quality requirements,and will bring security risks,has been difficult to meet the needs of the rapid development of the industry.It has become the focus of research for metro operating companies to improve the operational safety of metro through better maintenance support system and technical means.Based on the above-mentioned industry pain points,in order to reduce meaningless preventive maintenance or even excessive repair,this paper conducts a study in equipment fault prediction,based on the key technologies of fault prediction and health management,and a specific example study with an important single-point fault equipment turnout rutting machine.The accuracy rate of turnout fault prediction reaches more than80%,and in fault diagnosis,10 kinds of fault modes can be identified.The diagnostic study of equipment failure data facilitates rapid problem location and recovery,reducing the time to impact of failures.This achieves an agile and efficient maintenance support system at the source.Based on the health management of equipment,this paper also explores the research design of an agile maintenance support system based on big data and its implementation.For the design of the maintenance support system,a three-layer framework is used,including the collection layer,storage layer and application layer,and specific design ideas are proposed for several layers.Based on the actual needs of MTR turnout maintenance,the specific implementation explores the rapid implementation and continuous iteration of the business based on the business idea of mobile first,flexible Internet development framework and agile development idea,and achieves better results.The system implemented in this paper is now in practical use in the production environment of MTR and has been affirmed by the users.