【摘要】：海洋平臺在遠離陸地、惡劣環(huán)境、極端天氣、網(wǎng)絡條件差等諸多苛刻的不利條件下實施監(jiān)測,監(jiān)測生成的流數(shù)據(jù)來源各異、數(shù)量巨大、快速實時、結構多樣、瞬時變化,管理技術要求越來越高,許多傳統(tǒng)管理方法和模式不能滿足需求,如何科學有效地管理海洋平臺監(jiān)測流數(shù)據(jù)研究成果和實踐經(jīng)驗相對較少,也正是當前亟待解決的問題。本文總結了海洋平臺監(jiān)測流數(shù)據(jù)的特點和相關特性的變化,國內(nèi)外流數(shù)據(jù)遠程傳輸、規(guī)范化處理、分布式存儲和管理系統(tǒng)的研究現(xiàn)狀以及存在的主要問題,提出了基于反饋機制和MD5碼檢測的流數(shù)據(jù)遠程傳輸技術,通過發(fā)送端傳輸?shù)臄?shù)據(jù)文件的MD5碼和接收端生成的MD5碼一致性檢驗,保證了流數(shù)據(jù)傳輸?shù)耐暾院鸵恢滦裕唤⒘艘粋�包含所有可能情況的數(shù)據(jù)規(guī)范處理方式的監(jiān)測流數(shù)據(jù)修復函數(shù)庫和數(shù)據(jù)格式信息表,實現(xiàn)了對監(jiān)測流數(shù)據(jù)錯誤、丟失、無序、噪聲等問題的處理和對格式、范圍、標準不一致統(tǒng)一規(guī)范；基于Hadoop搭建了監(jiān)測流數(shù)據(jù)分布式存儲平臺,為高效查詢分析數(shù)據(jù)提供服務。針對現(xiàn)實應用的實際需求,設計了海洋平臺監(jiān)測流數(shù)據(jù)管理系統(tǒng),對用戶管理權限、流數(shù)據(jù)遠程傳輸、數(shù)據(jù)規(guī)范化處理、分布式數(shù)據(jù)存儲、數(shù)據(jù)查詢等功能進行了實現(xiàn)。通過對海洋平臺監(jiān)測流數(shù)據(jù)相關管理問題的研究,為海洋平臺監(jiān)測流數(shù)據(jù)的管理應用提供了思路,對確保海洋平臺安全作業(yè)、推動海洋重大工程監(jiān)測技術的發(fā)展和應用具有重大意義。
[Abstract]:Offshore platforms are monitored under harsh adverse conditions, such as far from land, harsh environment, extreme weather, poor network conditions, and so on. The current data generated by the monitoring are of various sources, large quantities, fast and real-time, diverse structures, and instantaneous changes. The requirement of management technology is getting higher and higher, and many traditional management methods and models can not meet the demand. How to manage the monitoring flow data of offshore platform scientifically and effectively has less research results and practical experience, which is also the problem to be solved urgently. In this paper, the characteristics and related characteristics of monitoring stream data on offshore platform are summarized. The research status and main problems of remote transmission, standardized processing, distributed storage and management system of stream data at home and abroad are summarized. Based on the feedback mechanism and MD5 code detection, the streaming data remote transmission technology is proposed. The integrity and consistency of stream data transmission are ensured by checking the MD5 code of the data file transmitted by the sender and the MD5 code generated by the receiver. A monitor stream data repair function library and data format information table including all possible data canonical processing methods are established, which can deal with the problems of monitoring stream data error, loss, disorder, noise and so on. Based on Hadoop, a distributed storage platform for monitoring stream data is built, which provides a service for efficient query and analysis of data. In order to meet the practical requirements of practical applications, a data management system for offshore platform monitoring stream is designed. The functions of user management authority, stream data remote transmission, data normalization, distributed data storage, data query and so on are implemented. Through the research on the management of the monitoring current data of the offshore platform, this paper provides some ideas for the management and application of the monitoring current data of the offshore platform, and ensures the safe operation of the offshore platform. It is of great significance to promote the development and application of marine engineering monitoring technology.
【學位授予單位】：大連理工大學
【學位級別】：碩士
【學位授予年份】：2015
【分類號】：P715

【參考文獻】

相關期刊論文 前1條

1 彭高輝;王志良;;數(shù)據(jù)挖掘中的數(shù)據(jù)預處理方法[J];華北水利水電學院學報;2008年06期

相關碩士學位論文 前2條

1 劉少偉;基于云計算的科學工作流數(shù)據(jù)存儲策略研究[D];國防科學技術大學;2011年

2 苗潤華;基于聚類和孤立點檢測的數(shù)據(jù)預處理方法的研究[D];北京交通大學;2012年

，

本文編號：2127403