Development Of Gasoline Mass Spectrum Analysis System Based On Cloud Architecture

The combination of gasoline detection and mass spectrometry technology not only enables users to know the specific information of gasoline,but also promotes the development of quality supervision technology in the refined oil industry.At present,the test methods widely used in the field of gasoline identification are near-infrared spectroscopy,nuclear magnetic resonance and gas chromatography.The above method can be used to test and identify the fuel properties,although the gasoline quality information can be obtained accurately.However,due to the complex experimental process and many factors affecting the results of interference identification,mass spectrometry is used.The gas samples were identified in the laboratory by means of atmospheric pressure ionization source mass spectrometry,and the original mass spectrometry data were obtained.For the development of gasoline mass spectrum analysis system with cloud architecture,the system architecture is designed according to the characteristics of B/S architecture system,combined with the functional requirements and design objectives of the system.The front end of the system uses Vue framework,and the back end uses Java development.The service registration and discovery center Eureka,the service Gate Way gateway and the circuit breaker Hystrix are configured.Besides,the whole system is divided into function modules by using the idea of micro-service,and the core business functions in the system are divided into two micro-service modules,user management and mass spectrum detection.Secondly,the data characteristics of the original mass spectrum data are analyzed,the original data is preprocessed,and uploaded to the system in the form of Excel files.Then the database is designed to store the standard mass spectrum data and the information data of the system operation.By using the data in the original mass spectrum database and using XGBoost algorithm to train the detection model of the system,the obtained identification model is used to identify the data uploaded by users to be detected,and the final identification result is returned to the browser page for users to check.Moreover,each divided micro-service module is packaged and loaded with Docker container,and then deployed on the cloud server.Users only need to log in the browser to access the system,without the need to install a separate system client,so that the detection function is provided to users in the form of cloud services.Thus,the system and database can be deployed in the "cloud",so that the system can get rid of the hardware dependence on CPU,memory and storage,and greatly reduce the burden of system developers and ordinary users.Based on the experimental results of mass spectrometry detection technology,this paper uses Java language to divide the system functions into different micro-service modules,and then deploits the system and database in the cloud.It has the characteristics of low development cost and easy operation,and can be widely used in the field of gasoline detection.