| Since the 21st century,the ocean has become increasingly prominent in the global strategic position,and shipping has also developed rapidly.However,the development of shipping has led to aggravation of the ship’s air pollution.Among the constituents of air pollution emissions from ships,PM10,PM2.5,SOx,and NOx are extremely harmful to the human body.China has always been a nation which is great power in shipping and has ranked first in the world’s port cargo throughput ranking for 17 consecutive years.Among China’s inland waterways,Yangtze River has always ranked first and has a large shipping throughput.In the port area along the Yangtze River,it is easy to cause severe air pollution disasters.If the future air pollution emissions from ships in the port can be predicted,traffic control can be carried out on related ships in advance to avoid air pollution disasters caused by ship congestion.Based on AIS data,this paper uses a ship emission inventory model based on the dynamic method to realize real-time calculation of ship air pollution emissions.In addition,the ship trajectory prediction model based on Naive Bayes algorithm is used in this paper to realize the function of predicting the ship’s pollutant emissions from ports along the Yangtze River.Based on these two core functions,this paper designs and implements a ship pollutant emission prediction system for the port area along the Yangtze River.The main work of this paper is as followers:1.Research theory and technology:Researched ship emission inventory theory,ship trajectory prediction theory,big data technology and server related technologies.And explained the core technology points used by this system.2.Functional module design:Based on business needs,this paper designed the functions of single ship pollution monitoring,grid pollution monitoring,regional pollution,port pollution monitoring,etc.And elaborate on specific function points.3.Technical architecture design:The technical architecture mainly includes four layers:data layer,computing support layer,service layer,and application layer.And elaborate the main implementation logic of each layer. |