The Design And Implementation Of Big Data Management Platform For Rail Vehicle Battery

As an auxiliary power source for locomotives,batteries are widely used in the field of rail transit.In the rail transit safety assessment system,real-time monitoring of battery working status is important.Therefore,combining the Internet of Things and big data technology to construct a battery monitoring system has broad application prospects.Focusing on the rail vehicle battery,this dissertation investigates the battery data acquisition controller and big data service platform.The main work and research results are summarized in the following aspects:(1)The battery data acquisition controller is designed and implemented.The controller is based on STM32F4 microprocessor,equipped with Free RTOS operating system and Fat Fs file system,including touch screen,CAN communication interface,RS485 communication interface,data storage interface,and realizes battery test Data reception,data analysis and storage,test task start-stop control and other functions.In order to solve the problem of inconsistent data receiving and receiving rates,the scheme of local temporary storage is added in the data parsing task and the data remote transmission task.At the same time,the controller uses Lo Ra sensor network and 5G gateway to upload battery test data in real time.(2)The battery big data service cluster is designed and implemented.Cluster consists of a public network server that provides access to public resources and an intranet server for storage and computing.Among them,the system functions are developed based on the Spring Boot framework,using a My SQL database with a master-slave structure for read-write separation,a Redis cluster for caching and calculation,Minio as object storage,and middleware Rabbit MQ for traffic peak reduction,and Nginx server for load balanced.In the development process,the project is automatically pulled,packaged and released through the deployment and configuration of the Jenkins server.The Docker container technology is used for resource allocation.Finally,functions such as battery test data reception,locomotive multifunction vehicle bus(MVB)data reception,data export,and battery big data billboard are realized.(3)Functional and non-functional tests were carried out for the controller and server clusters to verify the availability,reliability,and scalability of the system.The experimental results show that the CAN bus data processing capacity of the controller is 200 frames/s,and the server cluster can support the data access of 200 sets of battery test experiments at the same time,and it can meet the MVB data access of 100 locomotives.