CN-120654322-B - Drag reduction prediction model design method and related device based on longitudinal heterogeneous ship formation

CN120654322BCN 120654322 BCN120654322 BCN 120654322BCN-120654322-B

Abstract

The invention relates to a drag reduction prediction model design method and a related device based on longitudinal heterogeneous ship formation, belonging to the technical field of ship engineering and hydrodynamics, wherein the drag reduction prediction model design method comprises the steps of setting longitudinal heterogeneous ship formation numerical simulation parameters in preset simulation software; and predicting the formation resistance of the ship group to be formed by adopting a mixed graph neural network prediction model based on the action rule, wherein the mixed graph neural network prediction model comprises a collaborative framework encoded by a GATv2 dynamic attention mechanism and GINE geometric features. The invention can predict and optimize the drag reduction effect of the longitudinal heterogeneous ship formation and provides scientific basis and intelligent decision tool for the ship formation configuration.

Inventors

MA YONG
ZHU PENGXIANG
YAO YONG
YAN XINPING

Assignees

武汉理工大学

Dates

Publication Date: 20260512
Application Date: 20250605

Claims (3)

1. A drag reduction prediction model design method based on longitudinal heterogeneous ship formation is characterized by comprising the following steps: Setting longitudinal heterogeneous ship formation numerical simulation parameters in preset simulation software; Adopting a control variable method to design numerical simulation scenes of different formation configurations, and determining ship resistance coefficients of longitudinal heterogeneous ship formations under each formation configuration, wherein the formation configuration comprises the number, speed, spacing and ship types of the formation ships; analyzing the action rules of the flow field speed of the bow and the traveling wave on the shearing resistance and the pressure difference resistance of the longitudinal heterogeneous ships in different formation configurations based on the ship resistance coefficients of the longitudinal heterogeneous ships in different formation configurations; predicting formation resistance of the formation ship group by adopting a mixed graph neural network prediction model based on the action rule, wherein the mixed graph neural network prediction model comprises a GATv dynamic attention mechanism and GINE geometric feature coding cooperative framework; Setting longitudinal heterogeneous ship formation numerical simulation parameters in preset simulation software, wherein the longitudinal heterogeneous ship formation numerical simulation parameters comprise: In a STAR-CCM+ simulation environment, 8 monitoring points are arranged in front of a ship in a continuous arrangement mode, each monitoring point is spaced by 0.5m and used for counting bow flow field data, 40 monitoring points are arranged behind the ship at the end of formation, and each monitoring point is spaced by 0.5m and used for counting the change of a traveling wave; the numerical simulation scene adopting the control variable method to design different formation configurations comprises the following steps: under the condition that the ship speeds and the ship pitches are the same, different ship numbers are configured for the longitudinal heterogeneous ship formation; Under the condition that the number of vessels and the vessel spacing are the same, configuring different vessel speeds for the longitudinal heterogeneous vessel formation; Under the condition that the number of vessels and the speed of the vessels are the same, configuring different vessel spacing for the longitudinal heterogeneous vessel formation; the numerical simulation scene adopting the control variable method to design different formation configurations specifically comprises the following steps: Constructing a heterogeneous ship formation model by adopting KCS ships and SERIES60 ships, and configuring different ship type combinations for the longitudinal heterogeneous ship formation under the conditions of the determined ship quantity, the same ship speed and the same ship spacing; The analysis of the action rules of the bow flow field speed and the traveling wave on the shearing resistance and the pressure difference resistance based on the ship resistance coefficients of the longitudinal heterogeneous ship formation under different formation configurations comprises the following steps: In the STAR-CCM+ simulation process, the change of the flow field speed of the bow and the change of the wave height of the traveling wave with time are counted through monitoring points, and the average value of the change of the flow field speed of the bow and the change of the wave height of the traveling wave with time is calculated and displayed in a columnar graph; the method for predicting the formation resistance of the formation ship group by adopting the mixed graph neural network prediction model based on the action rule comprises the following steps: the hybrid graph neural network prediction model comprises GATv layers, PNA layers and GINE layers of parallel architecture, is used for multi-level feature extraction and analyzes model prediction precision; the analytical model prediction accuracy comprises: three data enhancement measures are adopted to solve the problem of scarcity of CFD simulation data, and prediction accuracy of different formation scales is quantitatively analyzed through thermodynamic diagrams.
2. An electronic device comprising a memory and a processor, wherein, The memory is used for storing programs; the processor, coupled to the memory, is configured to execute the program stored in the memory to implement the steps in the drag reduction predictive model design method based on longitudinal heterogeneous vessel formation of claim 1.
3. A computer readable storage medium storing a computer readable program or instructions which when executed by a processor is capable of carrying out the steps of the drag reduction predictive model design method based on longitudinal heterogeneous vessel formation of claim 1.

Description

Drag reduction prediction model design method and related device based on longitudinal heterogeneous ship formation Technical Field The invention relates to the technical field of ship engineering and hydrodynamics, in particular to a drag reduction prediction model design method and a related device based on longitudinal heterogeneous ship formation. Background With the acceleration of globalization process, shipping has become a core mainstay of international trade, with over 90% of the global commercial traffic being accomplished by the ocean, whose position in the transportation of goods is not replaceable. However, rapid developments in the shipping industry are also accompanied by significant environmental challenges. According to International Maritime Organization (IMO) data, carbon dioxide emissions in the 2018 shipping industry account for approximately 2.9% of the total world, and sulfur oxides and nitrogen oxides account for 11% and 15%, respectively. Excessive carbon dioxide emission not only exacerbates global warming, but also causes frequent extreme weather and sea level rise, endangers ecological system balance, and sulfur oxides and nitrogen oxides cause acid rain and air pollution, thus forming serious threat to soil, vegetation and human health. Therefore, how to reduce the navigation resistance of the ship so as to realize the energy conservation and emission reduction of shipping has become an important issue of green traffic development. In the prior art, the energy conservation and emission reduction of shipping are generally realized by means of fuel substitution, exhaust gas aftertreatment and the like, but the energy consumption of shipping cannot be fundamentally reduced by the means. Disclosure of Invention Therefore, it is necessary to provide a design method and a related device for drag reduction prediction model based on longitudinal heterogeneous ship formation, so as to solve the problem of higher energy consumption and emission of the existing shipping ships by a new thought without modification. In order to address the above challenges and achieve the goals of drag reduction and consumption reduction, in a first aspect, the present invention provides a drag reduction prediction model design method based on longitudinal heterogeneous ship formation, including: Setting longitudinal heterogeneous ship formation numerical simulation parameters in preset simulation software; Adopting a control variable method to design numerical simulation scenes of different formation configurations, and determining ship resistance coefficients of longitudinal heterogeneous ship formations under each formation configuration, wherein the formation configuration comprises the number, speed, spacing and ship types of the formation ships; analyzing the action rules of the flow field speed of the bow and the traveling wave on the shearing resistance and the pressure difference resistance of the longitudinal heterogeneous ships in different formation configurations based on the ship resistance coefficients of the longitudinal heterogeneous ships in different formation configurations; And predicting formation resistance of the ship group to be formed by adopting a mixed graph neural network prediction model based on the action rule, wherein the mixed graph neural network prediction model comprises a cooperative framework encoded by GATv dynamic attention mechanisms and GINE geometric features. In one possible implementation manner, setting longitudinal heterogeneous ship formation numerical simulation parameters in preset simulation software includes: In the STAR-CCM+ simulation environment, 8 monitoring points are arranged in front of the ship in a continuous mode, each monitoring point is spaced by 0.5m and used for counting bow flow field data, 40 monitoring points are arranged behind the ship at the end of formation, and each monitoring point is spaced by 0.5m and used for counting changes of traveling waves. In one possible implementation, the numerical simulation scenario of different formation configurations is designed by adopting a control variable method, including: under the condition that the ship speeds and the ship spaces are the same, different ship numbers are configured for the longitudinal heterogeneous ship formation model; under the condition that the number of vessels and the vessel spacing are the same, configuring different vessel speeds for the longitudinal heterogeneous vessel formation model; And under the condition that the number of the ships and the speeds of the ships are the same, configuring different ship pitches for the longitudinal heterogeneous ship formation model. In one possible embodiment, analyzing the law of action of bow flow field velocity and traveling wave on shear resistance and differential pressure resistance based on ship resistance coefficients of longitudinal heterogeneous ship formation under different formation configurations, comprises: In the