Study On The Approach Of Ship-bridge Collision Early-warning In Bridge Waterways Considering The Ship Behavior

With the rapid development of the country’s transportation industry,inland navigation is becoming more and more important in the water transportation system.Large-scale ships and more bridges deteriorate the inland navigation environment,especially in continuous bridge waterways,where the increase of uncertainties in ship navigation not only increased the risk of collisions between bridges and ships,but also affected the navigation of ships in varying degrees.Once the accident occurs,it will bring serious obstruction to the ship’s navigation.Therefore,it is important to carry out research on ship behavior in this specific area,analyze the causes and establish realtime warning model for ship-bridge touch,which is of great practical significance for guaranteeing the navigation safety of ships in bridge area waters.Automatic Identification System(AIS)is usually regarded as an significant navigation facility in maritime situational,thus the data is of great significance to the research of ship behavior.The dissertation takes the waterway of the continuous bridge waterways in Wuhan section of the Yangtze River as the research object.On this basis,the risk factors of early warning were studied from the point of view of ship-bridge collision accidents and ship behavior.During the analysis of the causes of the bridge accident based on the Fuzzy Fault Tree method,the statistical analysis of the 12 million AIS data of bridge area in Wuhan section of 2015 was conducted.The early warning risk factors such as ship length,age,speed,heading angle,and bridge distance were extracted,and a risk index indicator system for ship bridge touch in continuous bridge area was established.Meanwhile,a real-time early warning model of bridge touch was established which based on the theory of Fuzzy Logic Theory.The model was verified by simulations from 3 accident cases and 6 safe navigation cases.The main research work of the dissertation includes:(1)The extraction of risk factors based on bridge accidents.The potential risk factors are obtained by analyzing the data collected from the bridge accidents in the typical bridge waterways in China,on the basis of which a Fuzzy Fault Tree model for ship-bridges touch was established.Meanwhile,qualitative analysis and quantitative analysis were carried out to obtain the risk factors with a high degree of impact on the top event.(2)The extraction of risk factors based on the characteristics of the ship behavior.Based on the AIS data in Wuhan section of the Yangtze River,the influence of different bridge waterways,upstream and downstream,water period and ship length on ship behavior characteristics(including speed and heading angle) were statistically analyzed to achieve the extraction of ship behavioral risk factors based on AIS information.(3)The establishment and simulation of ship-bridge touch early-warning model.Based on the extracted risk factors,an Early-warning indicator system was established,and the Fuzzy Inference Subsystems at all levels were constructed with the ship’s dynamic navigation states in the bridge waterways.Moreover, the membership function of the risk factor and the inference rules of the subsystems were built on the basis of the navigation regulations and the expertise.Finally,the Fuzzy Inference Toolkit of matlab was used to realize the simulation and case verification of the early warning model.