Research On Optical Fiber Sensing Massive Monitoring Data Storage Method Based On MongoDB

With the rapid development of modern industrial technology, safety monitoring of major machinery and equipment operating status has become the industry's research focus and even more necessary. Fiber grating sensors for its small size, light weight, good flexibility, high temperature and corrosion resistant, Strong ability of anti electromagnetic interference to replace the traditional measuring sensors, widely used in major machinery and equipment condition monitoring and Intelli Sense. How to build the reliable, High concurrency, scalability and availability storage system for these massive fiber optic sensing monitoring data,to provide reliable data support of real-time monitoring, information analysis and failure analysis systems, has become a key concern at present issue.This paper point at the storage requirements of major machinery and equipment operating status online monitoring system based on optical fiber sensing technology, combined with the different characteristics of the configuration management information and FBG sensor monitoring data, using traditional relational databases and No SQL databases separate storage data, design and implement multi-level buffering mechanisms to meet the efficient, real-time, accurate demand of data storage, ease the response delay between the memory and disk I/O. The main research work is as follows:(1) Combined the applications of different functional modules of the major machinery and equipment operating status online monitoring system with characteristics of various types of data, divided the storage into three parts:multi-level cache, traditional relational database systems and No SQL distributed database system.Use multi-level cache queue to achieve timely, accurate and efficient real-time data analysis and display. Use Traditional relational database as system database for high consistency,high structural configuration and management information.Use Mongo DB as comprehensive state database for massive, unstructured Optical fiber sensing monitoring data.(2) Analysis the characteristics of the optical fiber sensing monitoring data, design and implementation of the massive monitoring data storage scheme based on traditional relational database. In the new model, the multi-threaded receiver and Socket buffer adjustment mechanism based on UDP are used to establish reliable high-speed transmission model. At the same time, the two-levels cache based on producer-consumer algorithm, multi-threaded access mechanism based on table replacement algorithm and database write mechanism based on ADO.NET bulkcopy method are used to establish massive storage model in the relational database.(3) Design and implementation the massive storage model of fiber grating sensing monitoring data based on Mongo DB. using the FODO algorithm to improved the Mongo DB highly scalable cluster, achieve a platform independent, multi user diversity extensible storage service based on Web Service.(4) Design and implementation the massive data storage system for optical fiber sensing monitoring data and the large rotating bearing vibration monitoring platform based on those.Through the experimental comparison under hundred million data conditions and the actual working status of equipment condition monitoring system based on fiber grating sensor network, the efficiency, stability and practicality of the new model is verified.