摘要
小型舰船在航行中,机库后方产生的不稳定回流区会影响舰载机在甲板上的作业。为了改善舰面流场,保障舰载机在甲板上的安全起降,基于改进的简化护卫舰外形(MSFS),首先结合计算流体力学和人工神经网络方法,预测不同的机库长度L和宽度W对应的回流区大小,建立两种神经网络下的回流区响应面,然后在该响应面上使用粒子群算法得到优化的舰船结构。结果表明:回流区长度随着L和W的变化呈现出明显的非线性变化;与数值模拟结果相比,误差逆传播神经网络得到的回流区长度最大相对误差约为0.9%,径向基函数神经网络得到的最大相对误差约为2.2%;两种神经网络下得到的预测回流区响应面在整体上较为相似,且在两个预测响应面上优化得到的最小回流区长度都为1.93H,和数值模拟结果之间的相对误差约为0.5%。
When a small ship moves forward,the unstable recirculation zone behind the hangar will affect a ship-borne helicopter’s operation over the deck.In order to improve the flow field and ensure the safe take off and landing of the ship-borne helicopter,the modified simplified frigate shape(MSFS)was used to predict the size of the recirculation zone corresponding to different hangar’s length L and width W,combined with computational fluid dynamics method and artificial neural network method.The response surfaces of the recir⁃culation zone for the two kinds of neural networks were established.Then the particle swarm optimization al⁃gorithm was employed to obtain the optimized ship structure based on these response surfaces.The predicted and numerical results show that the length of the recirculation zone changes nonlinearly with the change of L and W.Compared with numerical results,the maximum relative error obtained by error back propagation(BP)neural network is about 0.9%,and that gained from radial basis function(RBF)neural network is about 2.2%.The response surfaces of the recirculation zone predicted by the two neural networks are similar.The mini⁃mum length of the recirculation zone is 1.93H on the two predicted response surfaces,and the relative error compared with numerical results is about 0.5%.
作者
李通
王逸斌
赵宁
邓思强
LI Tong;WANG Yi-bin;ZHAO Ning;DENG Si-qiang(Systems Engineering Research Institute,CSSC,Beijing 100094,China;Key Laboratory of Unsteady Aerodynamics and Flow Control,Ministry of Industry and Information Technology,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)
出处
《船舶力学》
EI
CSCD
北大核心
2023年第1期10-22,共13页
Journal of Ship Mechanics
基金
南京航空航天大学研究生跨学科创新基金(KXKCXJJ202003)
国家数值风洞工程资助项目(NNW2018-ZT1A02)
江苏高校优势学科建设工程资助项目。
关键词
计算流体力学
数值模拟
人工神经网络
优化
舰船尾流场
回流区
computational fluid dynamics
numerical simulation
artificial neural network
optimization
ship airwake
recirculation zone
