Research On Dynamic Calculation Of Ship Navigation Risk In Complex Environment Waters

Maritime traffic safety has always been one of the most important research fields,because maritime accidents pose a major threat to the safety of waterway transport and hinder the high-quality development of waterway transportation.The continuous growth of trade leads to the rapid increase of shipping volume,which makes the maritime traffic busier,the navigation risk higher and the navigation environment more complex.Especially in the complex environments waters with high ship traffic density,various types of ships,and significant interaction and interference between ships,which brings huge supervision and management pressure to maritime supervisors.The scientific calculation of ship navigation risk is the core content of ship navigation safety and an important basis for maritime authorities to carry out safety supervision from the perspective of water navigation management.In this paper,taking the complex environments waters as the research objects,establish a dynamic domain model based on AIS data.The dynamic domain model is the basis of conflict probability detection.Predict the ship motion by considering the dynamic and uncertain characteristics of ship motion,and then calculate the conflict probability combined with the definition of conflict.Secondly,propose a method of ship cluster extraction from the perspective of water navigation management to divide regional ships into groups.Finally,based on the method of conflict probability detection and the regional cognition-oriented ship group model,propose the assessment framework of ship navigation risk in different local waters to realize the dynamic calculation of ship navigation risk in complex navigable waters.The specific research work of this paper is as follows:(1)Construct the model of ship conflict probability detect.Aiming at the effective perception of ship collision risk in the dynamic navigation scene of complex navigable waters,propose a calculation method of conflict probability.Based on AIS data mining,the relationship between ship lengths,speed,relative bear and extract the minimum safe distance,and a dynamic domain model is constructed.Then,considering the dynamics of ship navigation and the uncertainty of ship position,using the Monte Carlo simulation to quantify the conflict probability.The conflict detect method lays a foundation for the risk calculation of order ships in multi-ship encounter scenarios and ship group extraction from the perspective of water navigation management.(2)Ship cluster extraction from the perspective of water navigation management.In order to realize the fast and accurate grouping of ships in complex environments waters and facilitate the safety supervision of different local waters,a two-stage ship clusters division algorithm based on fast modularity optimization and spectral clustering algorithm is proposed.First,using the fast modularity optimization algorithm to divide the ship into communities,and according to the results of community division,the ship node and edge information graph based on ship conflict connectivity and spatial compactness is constructed.Combined with the conflict relationship between nodes and spatial location characteristics,calculate the difference as edge weight,and use the spectral clustering algorithm to realize the division of regional ship clusters.(3)Dynamic calculation of ship navigation risk.Based on the method of conflict detect and the ship clusters extraction from the perspective of water navigation management,the conflict probability detection and contribution identification of regional ships are carried out,and the risk quantification of different local waters and regions is realized.Furthermore,analysis the topological characteristics of the network of the ship navigation risk,obtain the conclusion that there are key ships in the water area.The integrated centrality model and node deletion method using to identify the key ships of regional water traffic.This paper focuses on three aspects related to water traffic safety,such as ship conflict probability detect,ship clusters extraction from the perspective of water navigation management,and quantification of ship navigation risk in different local waters and regions.It provides guidance for ship safe navigation and decision-making,and provides valuable reference for real-time and fine perception of ship navigation risk.It provides important support for improving the supervision level of maritime authorities and promoting the modernization of waterway supervision system and supervision capacity.