| With the rapid development of the global economy and the deepening of industrialization,the global ecological environment has further deteriorated.At present,the power fuel used in maritime transportation is ship fuel oil,which is an energy source with very high sulfur content and carbon emissions.To this end,countries around the world have invested a lot of research in order to optimize the energy structure of their offshore ships.At present,the development of LNG as a ship power fuel is relatively mature,on the one hand,due to the high calorific value of LNG fuel,followed by the low toxicity and harmful emissions of the fuel,it has become the most popular energy alternative in the maritime industry.In order to realize energy conservation and emission reduction and energy structure adjustment in the maritime industry.Since 2012,the Chinese government departments have also successively issued relevant policies to promote the "oil to gas" work of ships.In this era,in order to make the "oil to gas" of ships proceed smoothly,it is important to accurately predict the gas demand of LNG-fueled ships in ports and guide the planning and construction of shorebased supporting facilities for LNG-fueled ships in a more scientific way.This article is mainly divided into four parts.In the first part,the grey correlation analysis method is used to extract the main influencing factors affecting port throughput.In the second part,the DE-RBF prediction model and three control group models are constructed,the case verification is carried out,the model evaluation index is set,the highest fitting predictive model is compared and analyzed,and the future throughput is predicted by the model.In the third part,a regression model between oil and LNG fuel consumption in water transportation and port cargo and container throughput is established.In the fourth part,the predicted future throughput is imported into the regression model,and the gas volume demand of LNG-fueled ships in the port is calculated by combining the substitution rate indicators determined by the current "oil to gas" strength.Specifically,taking Dalian Port in Liaoning Province as an example,this thesis uses the grey correlation analysis method to extract the factors affecting the large throughput of ports based on the quarterly data of port throughput from 2012 to 2022,taking into account the port’s own throughput and the total GDP and import and export value of different port hinterland regions.The extracted factors were imported as feature data samples and throughput data,and the gray prediction model,BP neural network model,RBF neural network model and DE-RBF neural network model were verified by example,and the average relative error,average absolute error,mean squared error and determining coefficient were used as evaluation indicators to compare and analyze the prediction accuracy of the four models,and the results showed that the prediction results of the DE-RBF prediction model had the highest degree of fitting with the true value.The model was used to predict the throughput of Dalian Port in the next three years.Then,according to the fuel consumption of water transportation,the consumption of gasoline,diesel,fuel oil and natural gas is converted into the consumption of standard coal through the standard coal coefficient,and the regression relationship between it and the port cargo and container throughput is verified,and the regression mathematical model is established by polynomial approximation.Finally,the forecast results of the throughput of Dalian Port in the next three years will be imported into the regression model,and the energy consumption measured in standard coal will be obtained,and then the energy consumption converted into LNG will be calculated according to the standard coal coefficient of LNG,and finally the substitution rate index will be determined in combination with the current "oil to gas" strength,and the gas demand of LNG-fueled ships in Dalian Port is calculated. |
