Design Of Multi-Source Heterogeneous Data Integration And Analysis System For Structure Test

Structure test is of great significance to the safety test of building structures and the development of structure science.In the process of structure test,various kind of data need to be collected,including environmental parameters,such as temperature and humidity;measurement parameters of building structure,such as strain,pressure,acceleration;measurement parameters of loading device,such as frequency and displacement;control signal of controller;audio and image,etc.These collected data are multi-source heterogeneous.At present,the data management method of structure test is to collect data through various kind of software and store these data in different files.This data management method results in the isolation of various data of structural test and the inefficiency of data management and analysis.This paper designs and implements a multi-source heterogeneous data integration and analysis system for structure test.The system integrates structure test data and stores them in local host and cloud databases.The system has the functions of data analysis,data complete backup and data sharing,so as to realize efficient management and analysis of structure test data.First the system eliminates the heterogeneity of various kind of data by data integration algorithm and unifies the expression of data.Then the unified data is stored in local host and cloud databases.Data encryption system is used to ensure the security of data in the process of data storage,and message queue storage model is used to complete the high concurrent storage of data.Finally,the experimental data is analysed and visualized through cloud platform.The main work of this paper is as follows:1.Design and implement a multi-source heterogeneous data integration algorithm for structure test.The algorithm achieves unified expression by classifying,parsing and encapsulating all kinds of data.Based on this data integration algorithm,a local data management platform is developed,which can integrate real-time data and historical data files of structure tests.2.Storage of structure test data.In this paper,a data storage method combining local storage and cloud storage is adopted.There are two main problems to be solved in the process of cloud storage:data security and high concurrency storage.For data security,this paper designs and implements a data encryption system to ensure data security;for data storage with high concurrency,this paper uses message queue storage model to support concurrency.3.Analysis and visualization of structure test data.Through the development of structure test cloud platform,this paper realizes the functions of data analysis and visualization display.Cloud platforms support various commonly used data analysis methods for structure test,such as statistical analysis,data fitting,data filtering,principal component analysis,and data display.At the same time,cloud platform provides data sharing function,and multi-users can share experimental data with each other,so as to reduce research costs and improve research efficiency.