The present study proposes a sub-grid scale model for the one-dimensional Burgers turbulence based on the neuralnetwork and deep learning method.The filtered data of the direct numerical simulation is used to establis...The present study proposes a sub-grid scale model for the one-dimensional Burgers turbulence based on the neuralnetwork and deep learning method.The filtered data of the direct numerical simulation is used to establish thetraining data set,the validation data set,and the test data set.The artificial neural network(ANN)methodand Back Propagation method are employed to train parameters in the ANN.The developed ANN is applied toconstruct the sub-grid scale model for the large eddy simulation of the Burgers turbulence in the one-dimensionalspace.The proposed model well predicts the time correlation and the space correlation of the Burgers turbulence.展开更多
In present study, the subgrid scale （SGS） stress and dissipation for multiscale formulation of large eddy simulation are analyzed using the data of turbulent channel flow at Ret = 180 obtained by direct numerical si...In present study, the subgrid scale （SGS） stress and dissipation for multiscale formulation of large eddy simulation are analyzed using the data of turbulent channel flow at Ret = 180 obtained by direct numerical simulation. It is found that the small scale SGS stress is much smaller than the large scale SGS stress for all the stress components. The dominant contributor to large scale SGS stress is the cross stress between small scale and subgrid scale motions, while the cross stress between large scale and subgrid scale motions make major contributions to small scale SGS stress. The energy transfer from resolved large scales to subgrid scales is mainly caused by SGS Reynolds stress, while that between resolved small scales and subgrid scales are mainly due to the cross stress. The multiscale formulation of SGS models are evaluated a priori, and it is found that the small- small model is superior to other variants in terms of SGS dissipation.展开更多
Based on preliminary theoretical analysis and numerical experiment, it is found that land surface heterogeneity plays an important role in the models turbulent flux calculation. In nearly neutral atmosphere conditions...Based on preliminary theoretical analysis and numerical experiment, it is found that land surface heterogeneity plays an important role in the models turbulent flux calculation. In nearly neutral atmosphere conditions, variation coefficient of sub-scale roughness length, cell-average roughness, and reference height are main factors affecting the calculation of grid turbulent fluxes. The first factor has a determinant role on calculation deviation. The relative error generated by roughness heterogeneity could be more than 40% in some cases in certain areas （e.g., in vegetation-climate transition belt）. Selecting a specific reference height may improve the calculation of turbulent flux. In stable or unstable atmosphere conditions, with sensible heat flux as an example, analysis shows that the discrepancy is correlated to the sub-grid distributions of mean wind velocity, potential temperature gradient between land surface and reference levels, and atmosphere stability near surface layer caused by the heterogeneity of land surface roughness. The calculation of turbulent flux is the most sensitive to stability in the above three factors. The above analysis shows that it is necessary to make a further consideration for the calculation deviation of the turbulent fluxes brought from land surface heterogeneity in the present numerical models.展开更多
The effects of Reynolds number on both large-scale and small-scale turbulence properties are investigated in a square jet issuing from a square pipe. The detailed velocity fields were measured at five different exit R...The effects of Reynolds number on both large-scale and small-scale turbulence properties are investigated in a square jet issuing from a square pipe. The detailed velocity fields were measured at five different exit Reynolds numbers of 8 × 10^3 〈 Re 〈 5 × 10^4. It is found that both large-scale properties （e.g,, rates of mean velocity decay and spread） and small-scale properties （e.g., the dimensionless dissipation rate constant A = εL/（u^2）^3/2） are dependent on Re for Re ≤ 3 ×10^4 or Reλ ≤ 190, but virtually become Re-independent with increasing Re or Reλ. In addition, for Reλ 〉 190, the value ofA = εL/（u^2）^3/2 in the present square jet converges to 0.5, which is consistent with the observation in direct numerical simulations of box turbulence, but lower than that in circular jet, plate wake flows, and grid turbulence. The discrepancies in critical Reynolds number and A = εL/（u^2）^3/2 among different turbulent flows most likely result from the flow type and initial conditions.展开更多
Subgrid nonlinear interaction and energy transfer are analyzed using direct numerical simulations of isotropic turbulence. Influences of cutoff wave number at different ranges of scale on the energetics and dynamics h...Subgrid nonlinear interaction and energy transfer are analyzed using direct numerical simulations of isotropic turbulence. Influences of cutoff wave number at different ranges of scale on the energetics and dynamics have been investigated. It is observed that subgrid-subgrid interaction dominates the turbulent dynamics when cut-off wave number locates in the energy-containing range while resolved-subgrid interaction dominates if it is in the dissipation range. By decomposing the subgrid energy transfer and nonlinear interaction into ‘forward’ and ‘backward’ groups according to the sign of triadic interaction, we find that individually each group has very large contribution, but the net of them is much smaller, implying that tremendous cancellation happens between these two groups.展开更多
基金supported by the National Key R&D Program of China(Grant No.2022YFB3303500).
文摘The present study proposes a sub-grid scale model for the one-dimensional Burgers turbulence based on the neuralnetwork and deep learning method.The filtered data of the direct numerical simulation is used to establish thetraining data set,the validation data set,and the test data set.The artificial neural network(ANN)methodand Back Propagation method are employed to train parameters in the ANN.The developed ANN is applied toconstruct the sub-grid scale model for the large eddy simulation of the Burgers turbulence in the one-dimensionalspace.The proposed model well predicts the time correlation and the space correlation of the Burgers turbulence.
基金supported by the National Natural Science Foundation of China(10472053 and 10772098)
文摘In present study, the subgrid scale （SGS） stress and dissipation for multiscale formulation of large eddy simulation are analyzed using the data of turbulent channel flow at Ret = 180 obtained by direct numerical simulation. It is found that the small scale SGS stress is much smaller than the large scale SGS stress for all the stress components. The dominant contributor to large scale SGS stress is the cross stress between small scale and subgrid scale motions, while the cross stress between large scale and subgrid scale motions make major contributions to small scale SGS stress. The energy transfer from resolved large scales to subgrid scales is mainly caused by SGS Reynolds stress, while that between resolved small scales and subgrid scales are mainly due to the cross stress. The multiscale formulation of SGS models are evaluated a priori, and it is found that the small- small model is superior to other variants in terms of SGS dissipation.
基金Cooperative Project (2007DFB20210) funded by the Ministry of Science and Technology of the People's Republic of Chinathe Key Project of Basic Scientific Research +1 种基金Operation fund of Chinese Academy of Meteorological Sciences (2008Z006)the Independent Research Project of LaSW (2008LASWZI04,2009LASWZF02)
文摘Based on preliminary theoretical analysis and numerical experiment, it is found that land surface heterogeneity plays an important role in the models turbulent flux calculation. In nearly neutral atmosphere conditions, variation coefficient of sub-scale roughness length, cell-average roughness, and reference height are main factors affecting the calculation of grid turbulent fluxes. The first factor has a determinant role on calculation deviation. The relative error generated by roughness heterogeneity could be more than 40% in some cases in certain areas （e.g., in vegetation-climate transition belt）. Selecting a specific reference height may improve the calculation of turbulent flux. In stable or unstable atmosphere conditions, with sensible heat flux as an example, analysis shows that the discrepancy is correlated to the sub-grid distributions of mean wind velocity, potential temperature gradient between land surface and reference levels, and atmosphere stability near surface layer caused by the heterogeneity of land surface roughness. The calculation of turbulent flux is the most sensitive to stability in the above three factors. The above analysis shows that it is necessary to make a further consideration for the calculation deviation of the turbulent fluxes brought from land surface heterogeneity in the present numerical models.
基金supported by the Fundamental Research Funds for the Central Universities,China(3132015027)the general science research project of the education department of Liaoning Province,China(L2013198)the Natural Science Foundation of Liaoning Province,China(2014025012)
文摘The effects of Reynolds number on both large-scale and small-scale turbulence properties are investigated in a square jet issuing from a square pipe. The detailed velocity fields were measured at five different exit Reynolds numbers of 8 × 10^3 〈 Re 〈 5 × 10^4. It is found that both large-scale properties （e.g,, rates of mean velocity decay and spread） and small-scale properties （e.g., the dimensionless dissipation rate constant A = εL/（u^2）^3/2） are dependent on Re for Re ≤ 3 ×10^4 or Reλ ≤ 190, but virtually become Re-independent with increasing Re or Reλ. In addition, for Reλ 〉 190, the value ofA = εL/（u^2）^3/2 in the present square jet converges to 0.5, which is consistent with the observation in direct numerical simulations of box turbulence, but lower than that in circular jet, plate wake flows, and grid turbulence. The discrepancies in critical Reynolds number and A = εL/（u^2）^3/2 among different turbulent flows most likely result from the flow type and initial conditions.
文摘Subgrid nonlinear interaction and energy transfer are analyzed using direct numerical simulations of isotropic turbulence. Influences of cutoff wave number at different ranges of scale on the energetics and dynamics have been investigated. It is observed that subgrid-subgrid interaction dominates the turbulent dynamics when cut-off wave number locates in the energy-containing range while resolved-subgrid interaction dominates if it is in the dissipation range. By decomposing the subgrid energy transfer and nonlinear interaction into ‘forward’ and ‘backward’ groups according to the sign of triadic interaction, we find that individually each group has very large contribution, but the net of them is much smaller, implying that tremendous cancellation happens between these two groups.
