The invention discloses an individual ship collision risk calculation method based on big data, and belongs to the technical field of ship traffic. The method specifically comprises the steps of collecting ship historical data of an AIS and VTS of typical water areas in a certain period; obtaining ship static and dynamic traffic data; defining unavailable data and abnormal data through dissimilarity analysis; carrying out data cleaning, repairing and fusion; according to probability statistics and clustering analysis of the data, carrying out abstraction processing on hydro meteorology, navigation rules, steps and the like; and constructing a ship traffic flow evolution model. According to the method of the invention, collision risk identification is taken as an objective, a high-collision-risk ship group is constructed through a link prediction method; high-collision-risk ships are identified; a high-collision-risk ship identification model can be established through collision risk indexes in predicted time and space dimensions; and a ship collision risk identification method based on big data is formed, and the purposes of safety and high efficiency of ship traffic are achieved.
本发明中公开了一种基于大数据的个体船舶碰撞风险计算方法,涉及船舶交通技术领域;具体包括搜集一定时期内典型水域的AIS和VTS等船舶历史数据,获取船舶静态和动态交通数据,通过相异度分析定义不可用数据和异常数据,进行数据的清洗、修复、融合,根据数据的概率统计和聚类分析,同时对水文气象、通航规则和步骤等进行抽象化处理,构建船舶交通流演化模型。本发明中以碰撞风险识别为目标,通过链路预测方法构建高碰撞风险船舶群组,对高碰撞风险船舶进行识别,能够在预测的时间和空间维度内,通过碰撞风险性指标,建立高碰撞风险船舶识别模型,形成基于大数据的船舶碰撞风险识别方法,达到船舶交通的安全和高效的目的。
Individual ship collision risk calculation method based on big data
一种基于大数据的个体船舶碰撞风险计算方法
2021-08-10
Patent
Electronic Resource
Chinese
Ship collision risk calculation method considering uncertainty of ship position prediction
DOAJ | 2021
|European Patent Office | 2023
|European Patent Office | 2018
|European Patent Office | 2017
|European Patent Office | 2025
|