基于AIS数据的船舶航线挖掘

发布时间：2018-01-01 06:16

本文关键词：基于AIS数据的船舶航线挖掘　出处：《大连海事大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着航海信息技术的迅速进步,船舶自动识别系统产生的航行数据(简称AIS数据)呈现爆发式增长。这些海量的AIS数据蕴含有丰富的物流、航运、海上交通等有用知识。传统的数据分析方法已不能满足现代航运业快速发展的要求,急需有效的理论与技术对这些大量的AIS数据进行分析和应用。数据挖掘和聚类作为一种从海量数据中挖掘有用的、具有潜在效用知识的理论与方法,已经成为海量AIS数据的主要分析技术。本文以船舶轨迹为研究对象,数据挖掘和聚类技术为方法,针对航线规划中的相关问题,致力于开展构造基于AIS数据的航线规划服务系统。降低航线设计时的人为错误,同时为航路规划提供辅助决策提高监管效率,优化船舶通航。本文的主要工作如下:(1)AIS数据的解码与处理。在介绍AIS信息类型、内容更新时间的基础上,分析了 AIS原始信息的构成和主要语句结构,并对AIS信息进行了解码。设计了AIS数据库,并对错误的AIS信息进行了筛选。提出了缺失的AIS信息插值方法。获得清洁可用的AIS信息。(2)当轨迹数据达到大数据量级时会对数据的存储、传输和处理带来极大的困难和挑战。本文对经典的Douglas-Peucker算法进行了研究,并结合船舶航线设计时的偏航极限和船舶领域知识对压缩阈值进行了探究。提出了基于速度和航向约束的船舶轨迹数据压缩方法,并对两种算法的效果进行了实验对比。(3)根据船舶海上的运动特征,研究了船舶航向和航速变化规律,为研究船舶行为特征提供了新的依据。同时设计了在线识别船舶的搁置行为的算法,对航线规划中的船舶热点区域有一定的参考价值。(4)采用地理网格技术,对AIS数据进行了划分,挖掘出了热点航迹段和典型的航线。并提出了转向点和热点航迹段的数据库建立方法,对船舶航线规划有一定的参考意义。
[Abstract]:With the rapid development of navigation information technology, the navigation data (AIS data) generated by the automatic ship identification system (AIS) is explosive. These massive AIS data contain abundant logistics and shipping. Traditional data analysis methods can not meet the requirements of the rapid development of modern shipping industry. There is an urgent need for effective theory and technology to analyze and apply these large amounts of AIS data. Data mining and clustering is a useful theory and method for mining useful and potentially useful knowledge from massive data. This paper takes the ship trajectory as the research object, data mining and clustering technology as the method, aiming at the related problems in route planning. It is committed to construct a route planning service system based on AIS data to reduce the human error in route design and to provide auxiliary decision for route planning to improve the efficiency of supervision. The main work of this paper is as follows: decoding and processing of AIS data. Based on the introduction of AIS information type and content update time. This paper analyzes the composition and main sentence structure of AIS original information, and decodes the AIS information, and designs the AIS database. The missing AIS information interpolation method is proposed. The clean and usable AIS information is obtained. (2) when the track data reaches the level of large amount of data, the data will be stored. Transmission and processing bring great difficulties and challenges. The classical Douglas-Peucker algorithm is studied in this paper. Combined with the yaw limit of ship course design and ship domain knowledge, the compression threshold is explored, and a method of ship trajectory data compression based on speed and course constraints is proposed. The effects of the two algorithms are compared experimentally. 3) according to the characteristics of ship motion at sea, the variation of ship course and speed is studied. It provides a new basis for studying the characteristics of ship behavior, and designs an algorithm to identify the shelving behavior of ships online. There is a certain reference value to the hot spot area of ship in route planning. (4) the AIS data are divided by using the technology of geographical grid. The hot track section and the typical route are excavated and the method of establishing the database of the turning point and the hot track section is put forward which has certain reference significance for the ship route planning.
【学位授予单位】：大连海事大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP311.13;U675.7

【相似文献】