【摘要】：隨著工業(yè)4.0概念的提出,智能化成為工業(yè)發(fā)展的必然趨勢(shì)。把通信技術(shù)、網(wǎng)絡(luò)技術(shù)和傳感器技術(shù)相結(jié)合的物聯(lián)網(wǎng)技術(shù)將為工業(yè)智能化提供數(shù)據(jù)支撐。為滿足工業(yè)智能化的需求,工業(yè)生產(chǎn)環(huán)境中的物聯(lián)網(wǎng)設(shè)備數(shù)量將會(huì)逐漸增加,工業(yè)物聯(lián)網(wǎng)產(chǎn)生的數(shù)據(jù)量也會(huì)越來越大,如何有效處理工業(yè)生產(chǎn)中的海量物聯(lián)網(wǎng)數(shù)據(jù)成為亟待解決的問題。本文針對(duì)工業(yè)領(lǐng)域?qū)τ诖髷?shù)據(jù)處理技術(shù)的迫切需求,結(jié)合工業(yè)大數(shù)據(jù)的特點(diǎn)進(jìn)行系統(tǒng)研究和架構(gòu)設(shè)計(jì),實(shí)現(xiàn)了一套模塊化、可移植的工業(yè)能耗大數(shù)據(jù)分析平臺(tái)。首先,本文對(duì)無線傳感網(wǎng)絡(luò)領(lǐng)域和大數(shù)據(jù)領(lǐng)域的發(fā)展現(xiàn)狀進(jìn)行了簡要闡述,分析了工業(yè)生產(chǎn)和管理中對(duì)于大數(shù)據(jù)相關(guān)技術(shù)的迫切需求,并對(duì)能耗大數(shù)據(jù)分析平臺(tái)中使用的相關(guān)技術(shù)進(jìn)行了簡要介紹;隨后,詳細(xì)分析了用戶對(duì)于能耗大數(shù)據(jù)分析平臺(tái)的數(shù)據(jù)查詢功能和大數(shù)據(jù)分析功能的需求,以工業(yè)生產(chǎn)環(huán)境和工業(yè)大數(shù)據(jù)特性為背景對(duì)能耗大數(shù)據(jù)平臺(tái)的系統(tǒng)架構(gòu)進(jìn)行研究、設(shè)計(jì)和論證,最終將系統(tǒng)設(shè)計(jì)為由傳感網(wǎng)絡(luò)層、大數(shù)據(jù)平臺(tái)層和中間件層構(gòu)成的三層架構(gòu)模型;接著,完成了傳感網(wǎng)絡(luò)層的部署和實(shí)現(xiàn),以及適用于工業(yè)生產(chǎn)環(huán)境的高可用、可擴(kuò)展的大數(shù)據(jù)平臺(tái)層的選型、部署和優(yōu)化;然后,參考工廠現(xiàn)有條件和用戶需求對(duì)中間件層的必要性和可行性進(jìn)行了研究和論證,并完成了中間件層的模塊化設(shè)計(jì)和實(shí)現(xiàn);最后,在大數(shù)據(jù)平臺(tái)層的基礎(chǔ)上,通過中間件層的任務(wù)調(diào)度和功能銜接實(shí)現(xiàn)了能耗數(shù)據(jù)的統(tǒng)計(jì)查詢功能、即時(shí)查詢功能以及基于復(fù)雜模型的數(shù)據(jù)分析功能。本文最后在生產(chǎn)環(huán)境中驗(yàn)證了能耗大數(shù)據(jù)分析平臺(tái)的設(shè)計(jì)方案,并通過分析能耗數(shù)據(jù)提出了節(jié)能策略。
[Abstract]:With the introduction of the concept of industry 4.0, intelligence has become an inevitable trend of industrial development. The Internet of things, which combines communication technology, network technology and sensor technology, will provide data support for industrial intelligence. In order to meet the demand of industrial intelligence, the number of Internet of things equipment in the industrial production environment will gradually increase, and the amount of data generated by the industrial Internet of things will also become larger and larger. How to deal with mass Internet of things data in industrial production effectively becomes an urgent problem. In order to meet the urgent need of big data processing technology in the industrial field, this paper carries out systematic research and architecture design combined with the characteristics of industrial big data, and realizes a modular and portable big data analysis platform for industrial energy consumption. First of all, this paper briefly describes the development of wireless sensor network and big data, and analyzes the urgent needs of big data related technologies in industrial production and management. The related technologies used in the energy consumption big data analysis platform are briefly introduced, and then, the requirements of the users for the data query function and the big data analysis function of the energy consumption big data analysis platform are analyzed in detail. Based on the background of industrial production environment and industrial big data characteristics, the system architecture of energy consumption big data platform is studied, designed and demonstrated. Finally, the system is designed as sensor network layer. Then, the deployment and implementation of the sensor network layer, and the selection, deployment and optimization of the highly usable and extensible big data platform layer suitable for industrial production environment are completed. Then, the necessity and feasibility of the middleware layer are studied and demonstrated with reference to the existing conditions and user requirements of the factory, and the modularization design and implementation of the middleware layer are completed. Finally, on the basis of the big data platform layer, Through the task scheduling and function connection of middleware layer, the statistical query function of energy consumption data, the real-time query function and the data analysis function based on complex model are realized. Finally, the design scheme of energy consumption big data analysis platform is verified in the production environment, and the energy saving strategy is put forward by analyzing the energy consumption data.
【學(xué)位授予單位】：北京交通大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TN929.5;TP311.13

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 張紅;王曉明;曹潔;馬彥宏;郭義戎;王a\;;Hadoop云平臺(tái)MapReduce模型優(yōu)化研究[J];計(jì)算機(jī)工程與應(yīng)用;2016年22期

2 高見文;薛行貴;羅杰;姜源;吳啟武;;基于迭代式MapReducede的海量數(shù)據(jù)并行聚類算法研究[J];中國科技論文;2016年14期

3 吳卉男;;大數(shù)據(jù)系統(tǒng)和分析技術(shù)綜述[J];信息記錄材料;2016年03期

4 冷建飛;高旭;朱嘉平;;多元線性回歸統(tǒng)計(jì)預(yù)測(cè)模型的應(yīng)用[J];統(tǒng)計(jì)與決策;2016年07期

5 洪萬帆;蘇淑靖;;基于無線傳感器網(wǎng)絡(luò)的數(shù)據(jù)采集系統(tǒng)設(shè)計(jì)[J];電子器件;2016年01期

6 王慶福;;線性回歸算法在個(gè)性化推薦中應(yīng)用研究[J];電腦編程技巧與維護(hù);2015年23期

7 李振;賈瑞玉;;一種改進(jìn)的K-means蟻群聚類算法[J];計(jì)算機(jī)技術(shù)與發(fā)展;2015年12期

8 程雙江;李世平;鄔肖敏;鄭堂;;基于改進(jìn)混合核SVM的非線性組合預(yù)測(cè)[J];計(jì)量技術(shù);2015年10期

9 武霞;董增壽;孟曉燕;;基于大數(shù)據(jù)平臺(tái)hadoop的聚類算法K值優(yōu)化研究[J];太原科技大學(xué)學(xué)報(bào);2015年02期

10 劉旺鎖;王平波;顧雪峰;;混合高斯參數(shù)估計(jì)的兩種EM算法比較[J];聲學(xué)技術(shù);2014年06期

相關(guān)博士學(xué)位論文 前1條

1 花小朋;非平行超平面分類器算法研究[D];中國礦業(yè)大學(xué);2015年

相關(guān)碩士學(xué)位論文 前2條

1 童冰彬;基于分布式海量數(shù)據(jù)處理的協(xié)作無線傳感器網(wǎng)絡(luò)架構(gòu)[D];南京郵電大學(xué);2015年

2 朱敏明;基于YARN框架下并行化計(jì)算的研究[D];西北師范大學(xué);2015年

，

本文編號(hào)：2197289