Design And Implementation Of Pavement Structure Data Management Platform Based On Big Data Technology

Under the background of the new infrastructure,our country has put forward the major strategy of becoming a traffic power,and has begun to lay out a large number of intelligent transportation infrastructure,which has given the connotation of accurate perception and accurate analysis of traditional transportation infrastructure."Smart road" is a strong support for the construction of traffic power.At the same time,the current code takes the internal mechanical response of pavement structure as an important index for road design,evaluation and maintenance,which requires staff to obtain the internal mechanical response data of pavement comprehensively and accurately,so as to use these data for scientific research.Buried sensors of pavement structure are the core basis for realizing active sensing of internal mechanical response of pavement structure.Sensors of different road layers will monitor the state of the road in real time,thus producing a large amount of real-time data,a huge amount of data and a fast data generation rate,which brings great challenges to the staff for data collection.The pavement structure data management platform designed and implemented in this paper provides a complete set of convenient solutions,which can also meet the data collection,mass data storage,intelligent data completion,real-time data processing,equipment management and other functions.The main work content of this paper is as follows:(1)In this paper,the requirements of users and the data characteristics of sensors are comprehensively analyzed,and the total requirements of data acquisition,data basic services and business applications are determined by considering the goals and functional requirements to be realized by the system.(2)According to the requirement analysis,the overall architecture of the system is designed in this paper,and the system is divided into five layers:user service layer,business application layer,basic platform layer,data access layer and monitoring equipment layer.(3)In this paper,the system is designed in detail under the overall logical framework.In the aspect of data processing,the network communication protocol of sensor data is designed,the data access and parsing are realized by using Netty communication framework,and the data is distributed by Kafka distributed message queue.For the"user-driven" business,a massive data storage platform based on Hadoop is designed,and the intelligent completion of missing road data is realized.For the "data-driven" high real-time business,a real-time stream processing platform based on Flink is designed.In terms of business application,the conceptual structure of different devices is designed,and the collaboration process of the main modules is designed.(4)The data processing system and the business application system are implemented in this paper.The data processing system is deployed under the distributed framework,and the work of data collection,data distribution,massive data storage,missing data completion,realtime stream processing,etc.The business application system uses the MVC architecture idea.Completed the development of equipment management,data service,early warning management,user management and other services.At present,the system has been put into use to serve the collection and management of pavement structure data.It has been proved by practice that the system can run smoothly and users have a good response.