Research On Ship Speed Optimization Based On Dynamic Characteristics Of Hull-Propeller-Engine

Ocean shipping is the most important mode of transportation in international trade,but the annual CO₂ emission of the shipping industry is up to 1 billion tons.Under the background of climate issues attracting more attention from countries around the world and peak carbon dioxide emission and carbon neutrality have become the international consensus,the shipping industry urgently needs to reduce carbon emissions to meet the increasingly strict regulatory requirements.Ship speed optimization is favored by ship owners because of its low investment and significant benefits.The research on ship speed optimization based on the dynamic characteristics of hull-propeller-engine can not only reduce fuel consumption of ship main engine by optimizing the speed of each section,but also fully consider the dynamic characteristics of ship navigation,which is of great significance for guiding the economic operation of ships.In this paper,a 300,000-ton ocean-going tanker is taken as the research object,and the variation law of different parameters with ship speed in the dynamic process is analyzed.The mathematical and simulation model of dynamic process is established based on the mathematical and physical relationship between hull-propeller-engine.The machine learning model of fuel consumption was established based on ship navigation data,and the method of random search and grid search was used to optimize the hyperparameters of the model,and the reliability of model prediction was evaluated by simulating the actual navigation environment.The ship speed is discretized and Python was used to independently develop the source program,and Gurobi optimization solver was used to solve the steady-state ship speed optimization problem,and the effects of sailing time and meteorological conditions on the optimization results were analyzed.On this basis,iterative calculation and dynamic programming are used to study dynamic speed optimization.The results show that the dynamic process simulation model can accurately capture and record the changes of various parameters during acceleration and deceleration,and all of them meet the engineering application error of about 10%.The relative errors of fuel consumption prediction models are all within 4%,which show the prediction accuracy of the models is high.Speed optimization can achieve better fuel saving effect,compared with the historical voyage,the fuel saving is 45.43t and the fuel saving rate can reach 6.44%;compared with the sailing on punctual constant speed,the fuel saving is 26.61t and the fuel saving rate is 3.88%.Since fuel consumption,sailing distance and duration of dynamic process account for a very small proportion in the whole voyage（0.378%,0.276%and 0.290%,respectively）,there is no significant difference between steady-state and dynamic speed optimization in terms of fuel saving effect.But dynamic speed optimization fully demonstrates the acceleration and deceleration process of the ship during actual voyage.The research work in this paper breaks through the limitation of only considering steady-state navigation,and provides feasible technical means for energy conservation and emission reduction of ships while fully considering the dynamic process of ship voyage.