摘要
为了快速预报螺旋桨水动力性能,提出一种遗传算法优化BP神经网络(GA-BP)的螺旋桨水动力性能快速预测方法。采用计算流体力学(CFD)方法,使用FLUENT软件对P4119螺旋桨水动力性能进行仿真计算,结果表明CFD方法可替代敞水试验获取螺旋桨水动力性能。采用CFD方法计算100组不同系列螺旋桨的水动力性能,建立GA-BP神经网络螺旋桨水动力性能预测模型,以计算所得的100组数据为模型的学习样本,选择其中的80%为训练集,10%为验证集,10%为测试集,对比分析传统BP神经网络和GA-BP神经网络模型的预测效果。结果表明,GA-BP神经网络预测结果预测精度更高,能够满足快速预测螺旋桨水动力性能的要求。
In order to predict the hydrodynamic performance of propellers quickly,a genetic algorithm optimized BP neural network(GA-BP)method for rapid prediction of propeller hydrodynamic performance is proposed.The computational fluid dynamics(CFD)method is used to simulate the hydrodynamic performance of P4119 propeller using FLUENT software.The results show that the CFD method can replace the open water test to obtain the hydrodynamic performance of the propeller.Using CFD method to calculate the hydrodynamic performance of 100 different series of propellers,a GA-BP neural network propeller hydrodynamic performance prediction model is established.The calculated 100 sets of data are used as learning samples for the model,and 80%of them are selected as the training set,10%as the validation set,and 10%as the test set.The prediction performance of traditional BP neural network and GA-BP neural network models is compared and analyzed.The results indicate that the GA-BP neural network has higher prediction accuracy and can meet the requirements of quickly predicting the hydrodynamic performance of propellers.
作者
孙梦果
向祖权
SUN Mengguo;XIANG Zuquan(School of Naval Architecture,Ocean and Energy Power Engineering,Wuhan University of Technology,Wuhan 430063,China)
出处
《船舶工程》
CSCD
北大核心
2023年第10期69-75,141,共8页
Ship Engineering
基金
武汉理工大学威海研究院开放基金项目(WHYJY-KJ2021-002)
作者简介
孙梦果(1998—),男,硕士研究生。研究方向:船舶推进器水动力性能、船舶先进制造技术。
