Water traffic supervision and early warning system and method based on machine learning technology

Abstract

The invention discloses a water traffic supervision and early warning system and method based on a machine learning technology. According to the system, big data and machine learning technologies aremainly utilized, historical ship track data are trained according to ship types and regions to obtain a ship navigation rule model which is more consistent with the actual navigation condition of theship, abnormal behaviors of the ships in the whole sea area in the administrative region are monitored in real time by utilizing the model, and finally, the abnormal behaviors are displayed to water traffic supervisors through an electronic nautical chart. Compared with a system for water traffic safety supervision and early warning based on rule setting, the system has the advantages of being higher in intelligent degree, complete in supervision water region coverage rate and low in early warning false alarm rate.

Description

technical field [0001] The invention relates to a water traffic supervision and early warning system and method based on machine learning technology. Background technique [0002] With the rapid development of my country's domestic and foreign trade, the number of ships put into ocean transportation is increasing, and the types of ships, navigation and operation methods are various. The complexity and difficulty of water traffic supervision are increasing. The relevant departments of water traffic The requirements for safety supervision are also getting higher and higher. Water traffic command and water search and rescue are one of the core contents of the maritime system. In these two tasks, the sooner the water accident is discovered and intervened, the better the rescue effect and the smaller the loss. Therefore, the ship behavior analysis and abnormal automatic alarm function become more and more important. [0003] The traditional water traffic supervision system condu...

AI Technical Summary

Problems solved by technology

[0003] The traditional water traffic supervision system conducts ship warning supervision by setting rules, but there are two problems in this method: First, experts are required to summarize the navigation rules of specific supervised waters and set the rules in detail in the system, but the actual In operation, it is difficult to set appropriate alarm parameters for each area in the supervised water area, and the rules are difficult to generalize; second, even if powerful and well-conceived alarm rules can be set individually for each area, it will often lead to frequent alarms in the system. In the face of a large number of alarm messages, the on-duty personnel are at a loss as to what to do

At present, there is no monitoring and early warning system that effectively utilizes ship historical track data to ensure ship navigation

Images