Design And Implementation Of Passenger Flow Analysis System For Inteligent Station

Intelligent station is a cross-disciplinary field that has emerged in recent years and incorporates technologies such as big data,the Internet of Things,cloud computing,and artificial intelligence.After China proposed the " Excellent Project,Intelligent Jing Zhang ",it is imperative to use high technology to build the world’s leading intelligent station.China’s high-speed rail is constantly developing and catching up with other countries.The technical level is already in the forefront of the world,and most passengers are willing to travel by highspeed rail.At the same time,passengers are not only satisfied with their travel needs,but also pay more attention to the overall service experience.With the continuous expansion of the railway transportation network,the railway department has not only accumulated large-scale data on passenger flow data,but also found some problems that need to be resolved at this stage.On the one hand,the current way of selling tickets for high-speed rail still needs to be improved.It usually follows the principle of "first come,first served" to sell tickets during daily time and follows the principle of "inherent tickets" to sell tickets during the peak period.However,neither of these methods is well adapted to the complex and changing demand market.On the other hand,with the development of the railway transportation network and the increase of passenger number,manual means and video have been unable to evaluate the operating conditions of stations and trains in a timely and effective manner.The response of railway departments to large passenger flows and emergencies is often lagging,and they can only passively divert passenger flows.Passenger flow data can well reflect changes in passenger travel needs.Collecting,processing,storing,analyzing,and digging passenger flow data is also an important part of the railway’s intelligent process.Passenger flow data has immeasurable value,which can provide auxiliary decision support for railway departments and help them improve their operation management capabilities and service quality.In view of the above problems,this article has conducted a lot of research on big data and artificial intelligence technology,and designed and implemented a passenger flow analysis system from three aspects: real-time data analysis,historical data analysis,and trend prediction.The main research contents of this article are as follows:(1)Aiming at the continuous and high-frequency real-time passenger flow data,with considerations of performance requirements such as data processing speed,high-availability of clusters in real-time processing,a real-time calculation module is designed and implemented using technologies such as Spark,Kafka,Zookeeper and Redis in order to complete the calculation of passenger flow related indicators.(2)Aiming at the continuous growth and large-scale historical passenger flow data,taking into account the needs of data storage and data analysis,an offline computing module is designed and implemented.Using technologies such as Spark,Kafka,Zookeeper and Hive to design and implement the storage of massive historical passenger flow data into the database for subsequent analysis.Using technologies such as Azkaban,Sqoop,Hive and My SQL to design and implement the work of historical passenger flow data analysis.(3)Aiming at the demand for passenger traffic trend prediction,a passenger traffic prediction module is designed and implemented based on the LSTM neural network,and the passenger traffic data counted in the offline calculation module is used as training data for the LSTM network to predict the passenger traffic in the future.It is helpful for the railway department to understand the changes in passenger flow in the future and prepare measures to divert large passenger flows.This system has passed rigorous tests,basically achieved the expected results,and can work stably and efficiently.This system can analyze passenger flow information for railway departments,which has far-reaching significance for the development of railway intelligence.