应用数据驱动算法提高涡粘模型分离流动模拟精度

doi:10.13229/j.cnki.jdxbgxb20210322

Abstract

Abstract:

Aiming at the key problem of prediction accuracy of eddy viscosity model in Reynolds averaged N-S equations for separation flow， a data-driven machine learning algorithm based on artificial intelligence is developed to overcome the problem that the traditional eddy viscosity model relying too much on empirical parameters for the separation boundary layer flow prediction， and the accuracy of numerical simulation is improved. According to the physical mechanism affecting turbulence evolution， the eddy viscosity model calculation results of separation flow are taken as the baseline， several variables characterized by the average flow state are selected as the input variables， the high fidelity database is constructed based on the calculation results of high-order Reynolds stress model， and the six variables decomposed from Reynolds stress are taken as output variables， the mapping relationship and prediction model from baseline flow field data to high fidelity data constructed by random forest regression algorithm are established. The results show that， the prediction accuracy of the model for the separation flow is significantly improved.

Key words: power machinery and engineering, data driven algorithm, eddy viscosity model, Reynolds stress, separation flow

CLC Number:

V231.3

Zhong-hua GU,Pei-gang YAN,Pan-hong LIU,Xiang-feng WANG. Applying data driven algorithm to promote prediction accuracy of separation boundary simulation with eddy viscosity model[J].Journal of Jilin University(Engineering and Technology Edition), 2022, 52(11): 2532-2541.

Figures/Tables 19

Table 1

Input variables"

输入特征	原始变量 $α$	正则化系数 $β$
$q 1$	$12 Ω 2 - S 2$	$S 2$
$q 2$	$k$	$12 U i U i$
$q 3$	$m i n k d 50 ν, 2$	无需正则化
$q 4$	$U k ? P ? x k$	$? P ? x j ? P ? x j U i U i$
$q 5$	$k ε$	$1 S$
$q 6$	$? P ? x i ? P ? x i$	$12 ρ ? U k 2 ? x k$
$q 7$	$U i U j ? U i ? x j$	$U l U l U i ? U i ? x j U k ? U k ? x j$
$q 8$	$u' i u' j ˉ$	$k$
$q 9$	$ν t$	$100 ν$

Table 1

Fig.1

Fig.2

Fig.3

Table 2

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

Fig.9

Fig.10

Fig.11

Fig.12

Table 3

Fig.13

Fig.14

Fig.15

Fig.16

References 19

1	Marusic I, Candler G, Interrante V, et al. Real time feature extraction for the analysis of turbulent flows[J]. Data Mining for Scientific and Engineering Applications, 2001(2): 223-238.
2	Spencer A, Rivlin R. Isotropic integrity bases for vectors and second-order tensors[J]. Archive for Rational Mechanics & Analysis, 1964, 18(1): 51-82.
3	Tracey B D, Duraisamy K, Alonso J J. A machine learning strategy to assist turbulence model development[C]∥53rd AIAA Aerospace Sciences Meeting, Kissimmee, USA, 2015: 1287.
4	Zhang Z J, Duraisamy K. Machine learning methods for data-driven turbulence modeling[C]∥22nd AIAA Computational Fluid Dynamics Conference, Dallas, USA, 2015: 2460.
5	Duraisamy K, Zhang Z J, Singh A P. New approaches in turbulence and transition modeling using data-driven techniques[C]∥53rd AIAA Aerospace Sciences Meeting, Kissimmee, USA, 2015: 1284.
6	Singh A P, Duraisamy K, Zhang Z J. Augmentation of turbulence models using field inversion and machine learning[C]∥55th AIAA Aerospace Sciences Meeting, Grapevine, USA, 2017: 0993.
7	Wang J X, Wu J L, Xiao H. Physics informed machine learning approach for reconstructing reynolds stress modeling discrepancies based on DNS data[J]. Phys Rev Fluids, 2017(3): 1-22.
8	Ling J, Kurzawski A, Templeton J. Reynolds averaged turbulence modelling using deep neural networks with embedded invariance[J]. Journal of Fluid Mechanics, 2016, 807: 155-166.
9	Ling J, Templeton J. Evaluation of machine learning algorithms for prediction of regions of high Reynolds averaged Navier-Stokes uncertainty[J]. Physics of Fluids, 2015, 27(8): 085103.
10	Xiao H, Wu J, Wang J, et al. Quantifying and reducing model-form uncertainties in reynolds-averaged navier-stokes equations: a data-driven, physics-informed bayesian approach[J]. Journal of Computational Physics, 2016, 324: 115-136.
11	Xiao H, Wu J, Wang J, et al. Physics-informed machine learning for predictive turbulence modeling: progress and perspectives[J]. 55th AIAA Aerospace Sciences Meeting, Grapevine, USA, 2017: 1712.
12	Gamaharam M, Hattoriy Y. Searching for turbulence models by artificial neural network[J]. Physical Review Fluids, 2017, 2(5): 054604.
13	Maulik R, San O, Rasheed A, et al. Sub-grid modelling for two-dimensional turbulence using neural networks[J]. Journal of Fluid Mechanics, 2019, 858: 122-144.
14	Wang Z, Luo K, Li D, et al. Investigations of data-driven closure for subgrid-scale stress in large-eddy simulation[J]. Physics of Fluids, 2018, 30(12): 125101.
15	Wu J, Xiao H, Paterson E. Physics-informed machine learning approach for augmenting turbulence models: a comprehensive framework[J]. Physical Review Fluids, 2018, 3: 074602.
16	Thompson Roney L, Sampaio Luiz Eduardo B, de Braganca Alves Felipe A V, et al. A methodology to evaluate statistical errors in DNS data of plane channel flows[J]. Computation of Fluids, 2016, 130: 1-7.
17	Banerjee S, Krahl R, Durst F, et al. Presentation of anisotropy properties of turbulence, invariants versus eigenvalue approaches[J]. Journal of Turbulence, 2007, 8(32): 1-27.
18	Kumar V, Frohnapfel B, Jovaanovic J. Anisotropy invariant reynolds stress model of turbulence and its application to attached and separated wall-bounded flows[J]. Flow Turbulence Combust, 2009, 83: 81-103.
19	颜培刚, 杜明杰. 燃气涡轮带肋冷却通道流动与换热的大涡模拟[J]. 哈尔滨工业大学学报, 2018, 50(1): 75-81.
	Yan Pei-gang, Du Ming-jie. Numerical research of flow and heat transfer in a ribbed channel based on les approach[J]. Journal of Harbin Institute of Technology, 2018, 50(1): 75-81.

Related Articles 15

[1]	Yan ZHANG,Wei LIU,Shu-yong ZHANG,Yi-qiang PEI,Meng-meng DONG,Jing QIN. Optimization on thermal load of combustion chamber on two/four⁃stroke switchable diesel engine [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(3): 504-514.
[2]	Dong TANG,Yu-bin HAN,Lun HUA,Jin-chong PAN,Sheng LIU. Effect of lubricating oil ash on performance of gasoline particle filter in direct injection gasoline engine [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(11): 2501-2507.
[3]	Wen-bo ZHAO,Yu-jie LI,Jun DENG,Li-guang LI,Zhi-jun WU. Needle motion and its influence on in-nozzle flow and spray jet characteristics [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(10): 2234-2243.
[4]	Zhi-jun LI,Hao LIU,Li-peng ZHANG,Zhen-guo LI,Yuan-kai SHAO,Zhi-yang LI. Simulation on influence of microstructure of the wall on deep bed filtration of particulate filter [J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(2): 422-434.
[5]	Jian WANG,Xin XU,Han GU,Duo-jun ZHANG,Sheng-ji LIU. Heating characteristics of DOC based on exhaust thermal management of diesel engine [J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(2): 408-416.
[6]	Chang-qing SONG,Wen-miao CHEN,Jun LI,Da-wei QU,Hao CUI. Effects of single and dual ignition on combustion characteristics of natural gas under different equivalence ratios [J]. Journal of Jilin University(Engineering and Technology Edition), 2019, 49(6): 1929-1935.
[7]	Yi-xiao ZHU,Xiao-min HE,Yi JIN. Effects of radial strut width on flow structure ofsingle⁃cavity trapped vortex combustor [J]. Journal of Jilin University(Engineering and Technology Edition), 2019, 49(6): 1936-1944.
[8]	Chang-cheng LIU,Zhong-chang LIU,Jing TIAN,Yun XU,Ze-yu YANG. In⁃cylinder exergy destruction during combustion process ofheavy⁃duty turbocharged diesel engine [J]. Journal of Jilin University(Engineering and Technology Edition), 2019, 49(6): 1911-1919.
[9]	Qiao WANG,Wan-chen SUN,Liang GUO,Peng CHENG,Lu-yan FAN,Guo-liang LI. Effects of butanol/diesel blends on combustion and particulate emission characteristics of compression ignition engine [J]. Journal of Jilin University(Engineering and Technology Edition), 2019, 49(6): 1920-1928.
[10]	Xiao-yu HU,Guo-xiang LI,Shu-zhan BAI,Ke SUN,Si-yuan LI. Modified boiling heat transfer model considering roughness and material of heating surface [J]. Journal of Jilin University(Engineering and Technology Edition), 2019, 49(6): 1945-1950.
[11]	De⁃jun WANG,Zhi⁃chao LYU,Qi⁃ming WANG,Jian⁃rui ZHANG,Jian⁃nan DING. Cylinder pressure identification based on EKF and frequency⁃amplitude modulation Fourier series [J]. Journal of Jilin University(Engineering and Technology Edition), 2019, 49(4): 1174-1185.
[12]	Peng⁃fei ZANG,Zhe WANG,Yang GAO,Chen⁃le SUN. Investigation of integrated control strategy for stable operation of linear generator/engine system [J]. Journal of Jilin University(Engineering and Technology Edition), 2019, 49(3): 798-804.
[13]	JIANG Tao,LIN Xue-dong,LI De-gang,YANG Miao,TANG Xue-lin. Effect of control parameters on heat release rate with ANN method [J]. Journal of Jilin University(Engineering and Technology Edition), 2018, 48(6): 1747-1754.
[14]	LIN Xue-dong, JIANG Tao, XU Tao, LI De-gang, GUO Liang. Control strategy of high pressure pump in starting condition of high pressure common rail diesel engine [J]. Journal of Jilin University(Engineering and Technology Edition), 2018, 48(5): 1436-1443.
[15]	LI Zhi-jun, WANG Hao, HE Li, CAO Li-juan, ZHANG Yu-chi, ZHAO Xin-shun. Soot distribution features and its influence factors in catalytic diesel particulate filte [J]. Journal of Jilin University(Engineering and Technology Edition), 2018, 48(5): 1466-1474.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

Comments

Recommended 0

No Suggested Reading articles found!

方案	训练集	测试集	目标函数
1	k-?模型，Re=5600	k-?模型，Re=4200	DNS
2	SST模型，Re=5600	SST模型，Re=4200
3	k-?模型，Re=7000	k-?模型，Re=10 595

Applying data driven algorithm to promote prediction accuracy of separation boundary simulation with eddy viscosity model

RICH HTML

PDF (PC)