数据驱动的大型船舶靠泊行为监控设计

Design of Large Ship Berthing Behavior Monitoring Driven by Data

随着航运事业的发展,各航运和港口公司加快了码头的物流运作,导致了停泊作业仓促进行,使船舶发生碰撞事故的风险增加。船舶碰撞事故大多发生在靠泊过程中,而大型船舶一旦发生事故,将会造成人力物力的惨重损失。为解决以上问题,基于局部异常因子(LOF),使用船舶历史轨迹和对应的风、流和潮汐数据建立船舶靠泊行为模型,并对异常靠泊行为给出预警的系统设计。系统采用模块化的设计架构,提高船舶靠泊行为监控系统的数据处理效率,其中的AIS通信服务器可对船舶轨迹数据进行预处理,包括数据筛选和噪声去除,同时,建立了靠泊行为检测模型。

With the rapid development of shipping industry,the shipping and port companies speed up the logistics operation of the terminal,which leads to the rush of berthing operation and increases the risk of ship collision accidents.Most of the collision accidents occur in the process of ship berthing,and once large ships have accidents,it will cause heavy losses of human and material resources.In order to solve the above problems,based on the local anomalous factor(LOF),a ship berthing behavior model is established by using the ship’s historical trajectory and the corresponding wind,current and tide data,and an early warning system is designed for the abnormal berthing behavior.The system adopts modular design architecture to improve the data processing efficiency of ship berthing behavior monitoring system,in which the AIS communication server can preprocess the ship trajectory data,including data filtering and noise removal,the berthing behavior detection model is established.