Aerodynamic surrogate modeling mostly relies only on integrated loads data obtained from simulation or experiment,while neglecting and wasting the valuable distributed physical information on the surface.To make full ...Aerodynamic surrogate modeling mostly relies only on integrated loads data obtained from simulation or experiment,while neglecting and wasting the valuable distributed physical information on the surface.To make full use of both integrated and distributed loads,a modeling paradigm,called the heterogeneous data-driven aerodynamic modeling,is presented.The essential concept is to incorporate the physical information of distributed loads as additional constraints within the end-to-end aerodynamic modeling.Towards heterogenous data,a novel and easily applicable physical feature embedding modeling framework is designed.This framework extracts lowdimensional physical features from pressure distribution and then effectively enhances the modeling of the integrated loads via feature embedding.The proposed framework can be coupled with multiple feature extraction methods,and the well-performed generalization capabilities over different airfoils are verified through a transonic case.Compared with traditional direct modeling,the proposed framework can reduce testing errors by almost 50%.Given the same prediction accuracy,it can save more than half of the training samples.Furthermore,the visualization analysis has revealed a significant correlation between the discovered low-dimensional physical features and the heterogeneous aerodynamic loads,which shows the interpretability and credibility of the superior performance offered by the proposed deep learning framework.展开更多
The placement of pressure taps on the surface of the wind tunnel test model is an important means toobtain the surface pressure distribution.However,limited by space location and experimental cost,it isdifficult to ar...The placement of pressure taps on the surface of the wind tunnel test model is an important means toobtain the surface pressure distribution.However,limited by space location and experimental cost,it isdifficult to arrange enough pressure measuring taps on the surface of complex models to obtain completepressure distribution information,thus it is impossible to obtain accurate lift and moment characteristicsthrough integration.The paper proposes a refined reconstruction method of airfoil surface pressure basedon compressed sensing,which can reconstruct the pressure distribution with high precision with lesspressure measurement data.Tests on typical airfoil subsonic flow around flow show that the accuracyof lift and moment after the pressure integration reconstructed by 4-8 measuring points can meet therequirements of the national military standard.The algorithm is robust to noise,and provides a new ideafor obtaining accurate force data from sparse surface pressure tests in engineering.展开更多
Recent analytical solutions for peridynamic(PD)models of transient diffusion and elastodynamics allow us to revisit convergence of 1D PD models to their classical counterparts.We find and explain the reasons for some ...Recent analytical solutions for peridynamic(PD)models of transient diffusion and elastodynamics allow us to revisit convergence of 1D PD models to their classical counterparts.We find and explain the reasons for some interesting differences between the convergence behavior for transient diffusion and elastodynamics models.Except for very early times,PD models for transient diffusion converge monotonically to the classical one.In contrast,for elastodynamic problems this convergence is more complex,with some periodic/almost-periodic characteristics present.These special features are investigated for sine waves used as initial conditions.The analysis indicates that the convergence behavior of PD solutions to the classical one can be understood in terms of convergence properties for models using the Fourier series expansion terms of a particular initial condition.The results obtained show new connections between PD models and their corresponding classical versions.展开更多
基金supported by the National Natural Science Foundation of China(Nos.92152301,12072282)。
文摘Aerodynamic surrogate modeling mostly relies only on integrated loads data obtained from simulation or experiment,while neglecting and wasting the valuable distributed physical information on the surface.To make full use of both integrated and distributed loads,a modeling paradigm,called the heterogeneous data-driven aerodynamic modeling,is presented.The essential concept is to incorporate the physical information of distributed loads as additional constraints within the end-to-end aerodynamic modeling.Towards heterogenous data,a novel and easily applicable physical feature embedding modeling framework is designed.This framework extracts lowdimensional physical features from pressure distribution and then effectively enhances the modeling of the integrated loads via feature embedding.The proposed framework can be coupled with multiple feature extraction methods,and the well-performed generalization capabilities over different airfoils are verified through a transonic case.Compared with traditional direct modeling,the proposed framework can reduce testing errors by almost 50%.Given the same prediction accuracy,it can save more than half of the training samples.Furthermore,the visualization analysis has revealed a significant correlation between the discovered low-dimensional physical features and the heterogeneous aerodynamic loads,which shows the interpretability and credibility of the superior performance offered by the proposed deep learning framework.
基金by the foundation of National Key Laboratory of Science and Technology on Aerodynamic Design and Research(Grant 614220119040101)the National Natural Science Foundation of China(Grants 91852115 and 12072282)+1 种基金the National Numerical Wind tunnel Project(Grant NNW2018-ZT1B01)the Seed Foundation of Innovation and Creation for Graduate Student in Northwestern Polytechnical University(Grant CX2020195).
文摘The placement of pressure taps on the surface of the wind tunnel test model is an important means toobtain the surface pressure distribution.However,limited by space location and experimental cost,it isdifficult to arrange enough pressure measuring taps on the surface of complex models to obtain completepressure distribution information,thus it is impossible to obtain accurate lift and moment characteristicsthrough integration.The paper proposes a refined reconstruction method of airfoil surface pressure basedon compressed sensing,which can reconstruct the pressure distribution with high precision with lesspressure measurement data.Tests on typical airfoil subsonic flow around flow show that the accuracyof lift and moment after the pressure integration reconstructed by 4-8 measuring points can meet therequirements of the national military standard.The algorithm is robust to noise,and provides a new ideafor obtaining accurate force data from sparse surface pressure tests in engineering.
基金supported by the Fundamental Research Funds for the Central Universities(HUST:YCJJ202203014 and No.2021GCRC021)the Natural Science Foundation of China(No.11802098).
文摘Recent analytical solutions for peridynamic(PD)models of transient diffusion and elastodynamics allow us to revisit convergence of 1D PD models to their classical counterparts.We find and explain the reasons for some interesting differences between the convergence behavior for transient diffusion and elastodynamics models.Except for very early times,PD models for transient diffusion converge monotonically to the classical one.In contrast,for elastodynamic problems this convergence is more complex,with some periodic/almost-periodic characteristics present.These special features are investigated for sine waves used as initial conditions.The analysis indicates that the convergence behavior of PD solutions to the classical one can be understood in terms of convergence properties for models using the Fourier series expansion terms of a particular initial condition.The results obtained show new connections between PD models and their corresponding classical versions.
