The issue of resistance reduction through hull ventilation is of particular interest in contemporary research.This paper presents multiphase computational fluid dynamics(CFD)simulations with 2-DOF motion of a planing ...The issue of resistance reduction through hull ventilation is of particular interest in contemporary research.This paper presents multiphase computational fluid dynamics(CFD)simulations with 2-DOF motion of a planing hull.The original hull was modified by introducing a step to allow air ventilation.Following an assessment of the hull performance,a simulation campaign in calm water was conducted to characterize the hull at various forward speeds and air insufflation rates for a defined single step geometry.Geometric analysis of the air layer thickness beneath the hull for each simulated condition was performed using a novel method for visualizing local air thickness.Additionally,two new parameters were introduced to understand the influence of spray rails on the air volume beneath the hull and to indicate the primary direction of ventilated air escape.A validation campaign and an assessment of uncertainty of the simulation has been conducted.The features offered by the CFD methodology include the evaluation of the air layer thickness as a function of hull velocity and injection flow rate and the air volume distribution beneath the hull.The air injection velocity can be adjusted across various operating conditions,thereby preventing performance or efficiency loss during navigation.Based on these findings,the study highlights the benefits of air insufflation in reducing hull resistance for high-speed planing vessels.This work lays a robust foundation for future research and new promising topics,as the exploration of air insufflation continues to be a topic of contemporary interest within naval architecture and hydrodynamics.展开更多
基金supported by European Union funding(PON“Ricerca e Innovazione”2014‒2020).
文摘The issue of resistance reduction through hull ventilation is of particular interest in contemporary research.This paper presents multiphase computational fluid dynamics(CFD)simulations with 2-DOF motion of a planing hull.The original hull was modified by introducing a step to allow air ventilation.Following an assessment of the hull performance,a simulation campaign in calm water was conducted to characterize the hull at various forward speeds and air insufflation rates for a defined single step geometry.Geometric analysis of the air layer thickness beneath the hull for each simulated condition was performed using a novel method for visualizing local air thickness.Additionally,two new parameters were introduced to understand the influence of spray rails on the air volume beneath the hull and to indicate the primary direction of ventilated air escape.A validation campaign and an assessment of uncertainty of the simulation has been conducted.The features offered by the CFD methodology include the evaluation of the air layer thickness as a function of hull velocity and injection flow rate and the air volume distribution beneath the hull.The air injection velocity can be adjusted across various operating conditions,thereby preventing performance or efficiency loss during navigation.Based on these findings,the study highlights the benefits of air insufflation in reducing hull resistance for high-speed planing vessels.This work lays a robust foundation for future research and new promising topics,as the exploration of air insufflation continues to be a topic of contemporary interest within naval architecture and hydrodynamics.
文摘针对传统的神经网络模型因超参数众多,在实验中比对最优参数组合效率低下导致误差较大和反应速度慢的问题。本文提出一种基于北方苍鹰优化(Northern Goshawk Optimization,NGO)算法和双向门控循环单元神经网络(Bidirectional Gated Recurrent Unit, Bi-GRU)的船舶轨迹预测模型NGO-Bi-GRU(Northern Goshawk Optimization Bidirectional Gated Recurrent Unit)。利用NGO对Bi-GRU模型的学习率、隐藏节点和正则化系数进行寻优,然后将寻优得到的网络超参数代入Bi-GRU进行船舶轨迹预测。将该模型与长短时记忆神经网络(Long Short Term Memory, LSTM)和门控循环单元神经网络模型(Gated Recurrent Unit, GRU)以及使用该算法优化的长短期神经网络模型进行实验对比,将均方误差、均方根误差、平均绝对误差作为评价标准。结果表明,NGO-Bi-GRU模型在经度和纬度预测上误差较小、精确度较高且数值波动更加稳定。
