摘要
通过传统实验或CFD手段获取流场信息的方法往往需要耗费大量资源或时间,这在需要快速获取大量流场信息时产生的成本是无法接受的,发展比传统CFD更快速的流场预测方法具有重要意义。采用本征正交分解(POD)方法对样本流场进行模态分解,提取流场的主导模态;而后采用径向基函数神经网络(RBFN)响应POD基函数的系数,实现流场降阶预测模型的构建,并在模型中采用基于函数响应偏差的自适应抽样方法;通过某串列叶栅非定常流场数据对预测模型进行验证。结果表明:本文构建的POD-RBFN混合模型可以快速准确地预测出串列叶栅的流场参数分布;与静态采样相比,本文采用的自适应采样方法在采样效率上表现出明显优势,同样重构精度所需的样本数降低了25%左右。
The cost of obtaining a large amount of flow field information by traditional experiment or CFD is unacceptable,so it is of significance to develop faster forecasting calculation methods.Proper orthogonal decomposition(POD)is used to extract the dominant mode of the tandem flow field.Radial basis function network(RBFN)is used to respond to the coefficients of the POD basis functions to realize the construction of the reduced-order prediction model of the flow field.Then the adaptive sampling method is developed for the reduced order model.The prediction model is verified by the unsteady flow field data of a cascade.It is concluded that the hybrid method can be utilized to accurately predict the aerodynamics parameters and flow field of tandem cascade.Compared with static sampling,the number of samples required for adaptive sampling to achieve the same reconstruction accuracy is reduced by about 25%.
作者
尚珣
刘汉儒
杜亦璨
胡之颉
SHANG Xun;LIU Hanru;DU Yican;HU Zhijie(Yangtze River Delta Research Institute,Northwestern Polytechnical University,Taicang 215400,China;School of Power and Energy,Northwestern Polytechnical University,Xi’an 710129,China)
出处
《航空工程进展》
CSCD
2022年第5期86-94,共9页
Advances in Aeronautical Science and Engineering
基金
太仓市大院大所创新引领专项项目(TC2019DYDS09)。
关键词
降阶模型
本征正交分解
径向基函数神经网络
自适应抽样
压气机串列叶栅
reduced order model
proper orthogonal decomposition
radial basis function network
adaptive sampling
compressor tandem cascade
作者简介
尚珣(1998-),男,硕士研究生。主要研究方向:叶轮机械气动降阶方法;通信作者:刘汉儒(1985-),男,博士,副教授。主要研究方向:叶轮机械气动噪声及气动不稳定流动控制,hrliu@nwpu.edu.cn;杜亦璨(1997-),男,硕士研究生。主要研究方向:叶轮机械气动热力学;胡之颉(1997-),男,硕士研究生。主要研究方向:叶轮机械气动热力学。
