地球信息科学学报 ›› 2022, Vol. 24 ›› Issue (9): 1688-1700.doi: 10.12082/dqxxkx.2022.210710

• 地理信息技术在海事领域的应用 • 上一篇    下一篇

基于多源海事数据的大型船舶精细化分段乘潮研究

张新宇1,*(), 郭文强1, 王婧贇1, 杨炳栋2   

  1. 1.大连海事大学海上智能交通研究团队,大连 116026
    2.黄骅港引航站,沧州 061000
  • 收稿日期:2021-11-05 修回日期:2022-02-11 出版日期:2022-09-25 发布日期:2022-11-25
  • 作者简介:张新宇(1978— ),男,辽宁大连人,博士,教授,主要从事船舶交通组织优化、海事大数据,无人船自主航行研究。E-mail: zhangxy@dlmu.edu.cn
  • 基金资助:
    国家自然科学基金项目(51779028)

Research on the Refined Segmental Tide Riding of Large Vessels based on Multi-Source Maritime Data

ZHANG Xinyu1,*(), GUO Wenqiang1, WANG Jingyun1, YANG Bingdong2   

  1. 1. DaLian Maritime University, Maritime Intelligent Transportation Research Team, Dalian 116026, China
    2. Huanghua Port Pilot Station, Cangzhou 061000, China
  • Received:2021-11-05 Revised:2022-02-11 Online:2022-09-25 Published:2022-11-25
  • Contact: *ZHANG Xinyu, E-mail: zhangxy@dlmu.edu.cn
  • Supported by:
    National Natural Science Foundation of China(51779028)

摘要:

针对大型船舶长航道乘潮进港窗口期时长不充足问题,本文提出了基于船舶自动识别系统(Automatic Identification System,AIS)数据、港口潮汐数据、官方电子海图数据和航道地理位置数据等多源海事数据的大型船舶长航道精细化分段乘潮模型。首先,基于AIS数据采用K中心点算法对大型船舶乘潮航行行为特征进行挖掘,识别出大型船舶乘潮航迹关键点,计算大型船舶乘潮航行行为变化关键船位点。接着,结合长航道地理环境特征和大型船舶航行行为特征对长航道进行精细化分段,在此基础上基于港口潮汐数据构建大型船舶精细化分段乘潮窗口期计算模型。其次,设计乘潮历时自适应排列算法求解大型船舶乘潮最长窗口期;然后,以黄骅港综合港区航道为例验证了本文所提出的精细化分段乘潮模型。最后,基于电子海图数据利用地理信息系统平台实现大型船舶精细化分段乘潮三维动态推演,进一步验证大型船舶精细化分段乘潮航行的安全性。结果表明,该模型能够有效增加大型船舶乘潮进港窗口期时长,提高大型船舶乘潮进港效率,可为港航管理部门制定大型船舶进港计划提供理论指导。

关键词: 多源海事数据, K中心点算法, 精细化, 分段乘潮, 窗口期, 历时排列, 地理信息系统, 三维动态推演

Abstract:

Aiming at the problem of insufficient time for large vessels to enter the port by tide in the long channel, this paper proposed a refined segmented tide riding model for large vessels based on multi-source maritime data such as AIS data, port tide data, electronic chart data, and so on. Firstly, based on the AIS data, the K-medoids algorithm was used to mine the characteristics of the vessel's tide riding behavior, identify the key points of the vessel's tide riding trajectory, and calculate the key vessel position points of the vessel's tide-riding behavior change. Then, combing the geographical environment characteristics of the long channel and the characteristics of the navigation behavior of the vessels, the long channel was refined and segmented. Based on the port tide data, a refined calculation model of the vessel's segmented tide riding window period was constructed. Secondly, we designed an adaptive arrangement algorithm for the tide riding duration to solve the longest window period of the vessel's tide riding. Taking the Huanghua Port integrated port area as an example, the refined segmented tide riding model proposed in this paper was verified. Finally, based on the electronic chart data, the geographic information system platform was used to realize the fine segmented three-dimensional (3D) dynamic deduction of the vessel's tide riding, to further verify the safety of the vessel's tide navigation. The results show that the model can effectively increase the window period for vessels to enter the port by tide and improve the efficiency of ships entering the port by tide. This study can provide theoretical guidance for the port and shipping management departments to formulate vessel entry plans.

Key words: multi-source maritime data, K-medoids, refinement, ride the tide in stages, window period, duration sort, GIS, 3D dynamic deduction