Design And Implement Of Globaal Crude Oil And Shipping Analysis System

Crude oil trade has a significant impact on global trade,with daily trading volume of global crude oil reaching more than 70 million barrels.It can be said that the crude oil economy is a barometer of the state of the global economy.Crude oil trade volume is usually an ex-post statistical value released by customs statistics.However,the advance forecast of crude oil trade volume is more important for traders,futures traders and related companies.This paper implements a crude oil trade and shipping analysis system through ship position data and Web technology,which can accurately predict the amount of crude oil arriving in the port in the future.Through detailed analysis of the system's data requirements,business function requirements,and system performance requirements,this paper clarifies the main points of designing and developing a trade and analysis system,and finally realizes the system development and deployment.The design and implementation of the system are mainly accomplished through the following processes.1.Based on HDDBSCAN(Hadoop based Distributed DBSCAN)algorithm,a program for clustering ship position data was developed,and static AIS data was clustered into berth information for crude oil.Dividing and forecasting procedures,dividing the ship's historical trajectory into voyages and predicting the ship's destination port.Provides the system with basic data that can be used for trade analysis.2.Based on the front-end and back-end technologies of Node.js,various functional modules of the system have been developed to realize the statistics and forecast of crude oil trade,and the real-time shipping dynamics of crude oil ships can be viewed in real time.In order to enhance the performance of the system server,load balancing of the system is achieved through Nginx.By deploying multiple Node.js services on one server,the server's hardware resources can be effectively used.3.In order to verify the performance of the system and its ability to handle high concurrency,this article stress-tests the trade and analysis system in a real environment and compares it with Web services implemented using traditional Java technology.The service is more stable when dealing with high concurrency,and at the same time,the performance test of Redis read and write capabilities is performed,which can meet the system requirements.Finally,the shortcomings in the design and development of the system are summarized,and the planning and outlook for the next work are given.