【作者】 柏堃；

【导师】 赵德鹏；

【作者基本信息】 大连海事大学 ， 交通信息工程及控制， 2013， 博士

【摘要】 航运是一种有着悠久历史的运输方式。改革开放以来,航运业以其成本低、运量大、占地少、污染小等比较优势,为我国经济的发展及现代化建设贡献了重要力量。进入21世纪以来,现代科学技术的发展,尤其是信息技术、网络技术及数据通信技术的发展和应用对人类社会产生了深远的影响。信息化是合理利用各类航运资源、改善航运管理、保障航运安全、提高航运经济社会效益、实现航运可持续发展的有效手段及合理途径。中国航运业信息化建设从上个世纪80年代起步,经历了管理信息系统开发、电子数据交换和国际互联网应用等阶段。在30多年的信息化发展过程中,一方面,航运信息化建设已经取得一定的成绩,在航运生产及经营管理中发挥了重要的作用；另一方面,在航运业快速发展的同时,由于环境因素和自然条件等的限制,运输业务效率普遍较为低下,同时生产运输安全也存在着较为严重的隐患。在我国社会经济快速发展趋势下,航运业的需求结构将发生深刻变化,增长方式将发生由“数量扩张性”向“质量效益型”的重大转变。为适应十二五期间国民经济发展趋势,航运业必须在需求结构和发展方式发生深刻转变的情况下,针对行业发展中面临的主要矛盾和问题,进一步提高运输能力和服务效率。航运业是一个极易吸纳新技术的行业。物联网是新一代信息技术的重要组成部分,是继互联网和移动通信技术之后的又一次信息产业的革命性发展。一方面,航运信息化建设已在航运要素感知、信息传输、数据处理和行业应用等方面取得了一定的建设成果,另一方面,物联网的产业及相关技术的发展离不开具体行业应用的依托和支持。因此,本文尝试运用物联网相关技术来研究航运信息化建设,阐述了船联网的概念并提出了层次化的船联网体系结构,同时在船联网背景下,展开了关于构建船联网的重要技术射频识别(RFID)技术的数据处理相关研究,主要包括船联网RFID数据的清洗方法的研究及船联网RFID数据存储模型的研究。首先,本文分析了物联网及其相关概念,总结了物联网的发展情况、不同组织及学者对于物联网的理解及定义、物联网的主流体系架构及物联网相关关键技术。根据物联网的相关概念并结合航运业务具体需求,本文阐述了船联网的概念及船联网的基本特征。通过分析物联网的核心与基本特征及船联网的本质,本文提出了船联网三层体系结构,同时结合具体航运业务对各个层次具体功能及组成进行详细地分析与阐述。其次,船联网的发展必然带来海量数据的处理和存储问题,因此数据处理问题也就成为船联网需要研究的关键问题之一。对于射频识别技术在航运信息化相关的业务应用中所遇到的原始数据所存在的多读、漏读和冗余问题,本文有针对性地提出了一套对RFID原始数据进行有效清洗的方法。在本文的方法中通过三个步骤来对RFID数据进行清洗：首先提出了基于标签应答率的方法对原始数据中的多读数据进行判断并清洗；其次在研究经典SMURF数据清洗算法的基础上对其进行优化改进,对原始数据中的漏读数据进行处理；再次通过单点冗余清除和合并冗余清除两种方法对原始数据中的冗余数据进行清洗合并,最后通过模拟及真实数据实验对算法的性能进行分析。通过上述三个步骤的有效处理从而取得干净数据,为进一步的数据处理及相关上层应用提供了良好的数据基础。最后,本文分析了船联网RFID相关业务中涉及的实体及其关系,同时根据业务实体的动静态关系引入并借鉴了动态关系ER模型(DRER)对数据进行建模,将实体及实体间的动静态关系映射成相应的关系数据库表格,并在此基础上结合案例探讨了应用层的相关查询问题并将其转换为具体的SQL查询语句,展现了该模型强大的查询分析功能及动态语义表达功能。更多还原

【Abstract】 Shipping is a time-honored transportation means. Since the implementation of openness and reform policy, shipping contributes to our national economic development and modernization construction with its advantages of low costs, large transportation quantity, small occupation area and low pollution. As human society entered the21st century, the development of modern science and technology, especially the development and application of information technology, network technology and data communication technology, has a significant influence on human society. Informationization is an effective measure to make use of shipping resources, improve shipping management, guarantee shipping security, advance shipping economic&social benefits and achieve shipping sustainable development.China shipping informationization construction begun in the1980s, and has underwent the stages of management information system developing, eletronic data exchange and Internet application and so on. In the course of30years development, on one hand, shipping informationization construction has achieved a great deal and has played an important role in shipping production and management; on the other hand, shipping has a low efficiency because of the limitation of natural conditions and environmental elements. At the same time, shipping security has a serious hidden danger. With the rapid development of community economic&technology, the demand structure of shipping will undergo profound changes. The growth mode will change from quantity-increasing type to quality-benefit type. In order to adapt to this development trend, shipping should further enhance the transport capability and service efficiency to solve the main contradictions and problems which faces the development of shipping.Shipping is an industry which could easily absorb new technologies. Internet of Things (IOT) is an important aspect of new generation information technology and it is a revolutionary advance after Internet and mobile communication technology. On one hand, shipping informationization construction has made certain achievement in shipping elements sensing, information transmission, data processing and industry application and so on. On the other hand, the development of IOT is highly dependant on support of shipping and other industries. So, this paper tries to advance the shipping informationization construction with technologies of IOT and explains the conception of Internet of Ships and the architecture of Internet of Ships. On the basis of Internet of Ships, this paper researches on the data process method of Radio Frequency Identification (RFID), mainly including RFID data cleaning and modeling RFID data in the frame of Internet of Ships.Firstly, this paper analyzes the IOT and its relating concept, summarizes the development of IOT, the definition of IOT from different organization and scholars, the architecture and key technologies of IOT. According to the concept of IOT and combining the concrete demands of shipping, this paper proposes the conception of Internet of Ships and explains the essence of Internet of Ships. At the same time, this paper proposes the three-layered architecture of Internet of Ships and analyzes the function and makeup of each layer in combination with concrete shipping business.Secondly, data processing technologies become one of the key problems of Internet of Ships as the development of Internet of Ships brings the problems of mass data process and storage. With respect to the false positive, false negative and redundancy of original RFID data produced in shipping informationization related business, this paper puts forward a set of original RFID data cleaning methods. We achieve the RFID data cleaning in three steps:first of all, based on the tag response rate we clean the false positive data; then we study the SMURF data cleaning algorithm and improve it, so the improved algorithm is used to cleaning false negative RFID data; once again we clean the redundancy data with single redundancy cleaning method and combination redundancy cleaning method; last, we analyze the performance of our methods with analog data. The clean data provides a good foundation for up-layer application.Lastly, this paper analyzes the entities and relationship of these entities in the relative business of Internet of Ships. Meantime, this paper models RFID data with Dynamic Relationship Entity-Relationship (DRER) model according to the static and dynamic relationships between entities. Then, this paper maps relationship into relationship database. Besides, some inquiries about upper layer applications are conducted based on detail cases and these inquiries are converted to corresponding SQL query statements. It is proved that this model has a powerful query and analyse function and can express complicated dynamic semantics.更多还原