耦合聚类的数据驱动稀薄流非线性本构计算方法

doi:10.7527/S1000-6893.2022.27708

Abstract

Abstract:

Owing to the rarefied nonequilibrium effect encountered in near space environment， the existing numerical methods are difficult to describe the multi-scale flow phenomena efficiently and accurately at the same time. By integrating the machine learning methods proposed in recent years， Data-driven Nonlinear Constitutive Relations （DNCR） presented a new data-driven modeling approach for solving the nonequilibrium problem. To further improve the generalization capability and model training efficiency of the DNCR method， Gaussian Mixture Model （GMM） and Sparse Principal Component Analysis （SPCA） for preprocessing the data of training set and prediction set are introduced to the DNCR method for the first time in this paper. The prediction results of relevant cases show that the dominant parameters in different flow fields are extracted by GMM and SPCA without relying on artificial experience， which could improve the interpretability and robustness of DNCR. On the other hand， GMM and SPCA can accurately cluster the data points under a large number of flow field samples to eliminate the interference of redundant information and reduce the training and prediction time of the regression model， which is crucial to the updating and maintenance of the model when adding new sample data in future. For prediction accuracy， the coupled DNCR could improve the computational efficiency without losing accuracy in simple flows， and could further elevate the precision significantly in complex flows coupled with different flow characteristics.

Key words: Gaussian mixture model, sparse principal component analysis, data driven, tree model, rarefied non-equilibrium flow

CLC Number:

V211.3

Shaobo YAO, Lijian JIANG, Wenwen ZHAO, Zheng LU, Changju WU, Weifang CHEN. Numerical method of data-driven rarefied nonlinear constitutive relations coupled with clustering[J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(S2): 40-53.

Figures/Tables 18

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Table 1

Fig. 6

Table 2

Table 3

Fig. 7

Table 4

Fig. 8

Table 5

Fig. 9

Table 6

Fig. 10

Fig. 11

Fig. 12

References 29

1	TSIEN H S. Superaerodynamics, mechanics of rarefied gases[J]. Journal of the Aeronautical Sciences, 1946, 13(12): 653-664.
2	BHATNAGAR P L, GROSS E P, KROOK M. A model for collision processes in Gases. I. Small amplitude processes in charged and neutral one-component systems[J]. Physical Review, 1954, 94(3): 511-525.
3	XU K, HUANG J C. A unified gas-kinetic scheme for continuum and rarefied flows[J]. AIP Conference Proceedings, 2011, 1333(1): 525-530.
4	LIU S, YU P, XU K, et al. Unified gas-kinetic scheme for diatomic molecular simulations in all flow regimes[J]. Journal of Computational Physics, 2014, 259: 96-113.
5	周恒, 张涵信. 空气动力学的新问题[J]. 中国科学: 物理学力学天文学, 2015, 45(10): 109-113.
	ZHOU H, ZHANG H X. New problems of aerodynamics[J]. Scientia Sinica (Physica, Mechanica ＆ Astronomica), 2015, 45(10): 109-113 (in Chinese).
6	张伟伟, 寇家庆, 刘溢浪. 智能赋能流体力学展望[J]. 航空学报, 2021, 42(4): 524689.
	ZHANG W W, KOU J Q, LIU Y L. Prospect of artificial intelligence empowered fluid mechanics[J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(4): 524689 (in Chinese).
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]. Physical Review Fluids, 2017, 2(3): 1-22.
8	RABAULT J, KUCHTA M, JENSEN A, et al. Artificial neural networks trained through deep reinforcement learning discover control strategies for active flow control[J]. Journal of Fluid Mechanics, 2019, 865: 281-302.
9	SEKAR V, KHOO B C. Fast flow field prediction over airfoils using deep learning approach[J]. Physics of Fluids, 2019, 31(5): 57103.
10	LI Z, KOVACHKI N, AZIZZADENESHELI K, et al. Fourier neural operator for parametric partial differential equations[DB/OL]. arXiv preprint: 2010.08895, 2020.
11	BAR-SINAI Y, HOYER S, HICKEY J, et al. Learning data-driven discretizations for partial differential equations[J]. Proceedings of the National Academy of Sciences of the United States of America, 2019, 116(31): 15344-15349.
12	KOCHKOV D, SMITH J A, ALIEVA A, et al. Machine learning-accelerated computational fluid dynamics[J]. Proceedings of the National Academy of Sciences of the United States of America, 2021, 118(21): e2101784118.
13	RAISSI M, PERDIKARIS P, KARNIADAKIS G E. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations[J]. Journal of Computational Physics, 2019, 378: 686-707.
14	HU L, XIANG Y, ZHAN J, et al. Aerodynamic data predictions based on multi-task learning[J]. Applied Soft Computing, 2021, 116: 108369.
15	ZHANG J, MA W. Data-driven discovery of governing equations for fluid dynamics based on molecular simulation[J]. Journal of Fluid Mechanics, 2020, 892: A5.
16	XING H Y, ZHANG J, MA W J, et al. Using gene expression programming to discover macroscopic governing equations hidden in the data of molecular simulations[J]. Physics of Fluids, 2022, 34(5): 057109.
17	ZHAO W W, JIANG L J, YAO S B, et al. Data-driven nonlinear constitutive relations for rarefied flow computations[J]. Advances in Aerodynamics, 2021, 3(1): 540-558.
18	李廷伟, 张莽, 赵文文, 等. 面向稀薄流非线性本构预测的机器学习方法[J]. 航空学报, 2021, 42(4): 524386.
	LI T W, ZHANG M, ZHAO W W, et al. Machine learning method for correction of rarefied nonlinear constitutive relations[J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(4): 524386 (in Chinese).
19	蒋励剑,赵文文,陈伟芳,等. 旋转不变的数据驱动稀薄非线性本构计算方法[J/OL].航空学报, (2021-10-14)[2022-06-30]. .
	JIANG L J, ZHAO W W, CHEN W F, et al. Data-driven rarefied nonlinear constitutive relations based on rotation in-variants[J/OL]. Acta Aeronautica et As-tronautica Sinica, (2021-10-14)[2022-06-30]. .
20	SAFAVIAN S R, LANDGREBE D. A survey of decision tree classifier methodology[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1991, 21(3): 660-674.
21	孙吉贵, 刘杰, 赵连宇. 聚类算法研究[J]. 软件学报, 2008, 19(1): 48-61.
	SUN J G, LIU J, ZHAO L Y. Clustering algorithms research[J]. Journal of Software, 2008, 19(1): 48-61 (in Chinese).
22	贺玲, 吴玲达, 蔡益朝. 数据挖掘中的聚类算法综述[J]. 计算机应用研究, 2007, 24(1): 10-13.
	HE L, WU L D, CAI Y C. Survey of clustering algorithms in data mining[J]. Application Research of Computers, 2007, 24(1): 10-13 (in Chinese).
23	SUNG H G. Gaussian mixture regression and classification[D]. Houston: Rice University, 2004: 23-47.
24	ZOU H, HASTIE T, TIBSHIRANI R. Sparse principal component analysis[J]. Journal of Computational and Graphical Statistics, 2006, 15(2): 265-286.
25	TSIEN H S. Superaerodynamics, mechanics of rarefied gases[M]. Collected Works of H.S. Tsien (1938-1956). Amsterdam: Elsevier, 2012: 406-429.
26	HARTIGAN J A, WONG M A. Algorithm AS 136: A K-means clustering algorithm[J]. Applied Statistics, 1979, 28(1): 100.
27	WOLD S, ESBENSEN K, GELADI P. Principal component analysis[J]. Chemometrics and Intelligent Laboratory Systems, 1987, 2(1-3): 37-52.
28	CALLAHAM J L, KOCH J V, BRUNTON B W, et al. Learning dominant physical processes with data-driven balance models[J]. Nature Communications, 2021, 12: 1016.
29	GEURTS P, ERNST D, WEHENKEL L. Extremely randomized trees[J]. Machine Learning, 2006, 63(1): 3-42.

外形	来流压力P/Pa	来流马赫数Ma
1.1	0.186 2	3
1.1	0.186 2	5
1.1	0.279 3	5
1.1	0.465 5	5
1.2	0.058 2	10
1.2	0.038 8	10

模型	平均训练时间/s	平均预测时间/s
GMM+SPAC	7.12	0.014
原始回归模型	127.97	2.34
聚类后回归模型1	4.88	0.44
聚类后回归模型2	16.64	预测集无此簇数据
聚类后回归模型3	28.26	1.15
聚类后回归模型4	52.66	1.18
聚类后回归模型5	8.48	预测集无此簇数据

外形	来流压力P/Pa	来流马赫数Ma
2.1	0.05	3
2.2	0.05	3
2.2	0.06	3

[1]	Tianhe GAO, Kuo TIAN, Lei HUANG, Shu ZHANG, Zengcong LI. Data⁃driven shape⁃topology optimization method for curved shells [J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(2): 428806-428806.
[2]	Qian YANG, Yanzhe WANG, Di YANG, Zezhong LI, Weiwei QU. Prediction and planning of automatic laying speed for fiber reinforced composite materials based on data⁃driven model [J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(10): 429313-429313.
[3]	CHEN Bo, YUE Kai, WANG Rusheng, HU Mingnan. Learning-based multi-rate multi-sensor fusion localization method [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(S1): 726904-726904.
[4]	JIANG Lijian, ZHAO Wenwen, CHEN Weifang, YAO Shaobo. Data-driven rarefied nonlinear constitutive relations based on rotation invariants [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(12): 126256-126256.
[5]	WANG Shaoping, GENG Yixuan, SHI Cun. Life estimation of aircraft hydraulic pump based on failure physics and data driven [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(10): 527347-527347.
[6]	LI Tingwei, ZHANG Mang, ZHAO Wenwen, CHEN Weifang, JIANG Lijian. Machine learning method for correction of rarefied nonlinear constitutive relations [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(4): 524386-524386.
[7]	YAN Chongyang, ZHANG Yufei, CHEN Haixin. Application of field inversion based on discrete adjoint method in turbulence modeling [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(4): 524695-524695.
[8]	CHEN Zhijie, TANG Jinhui, WANG Chong, CHENG Jizeng, CAO Shan, SHAO Xin. Using artificial intelligence in airspace system to improve airspace hierarchical governance capability [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(4): 525018-525018.
[9]	MENG Songhe, YE Yumei, YANG Qiang, HUANG Zhen, XIE Weihua. Digital twin and its aerospace applications [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2020, 41(9): 23615-023615.
[10]	LI Baozhu, DONG Yunlong, DING Hao, GUAN Jian. Anti-bias track association algorithm based on Gaussian mixture model [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2019, 40(6): 322650-322650.
[11]	WUNIRI Qiqige, LI Xiaoping, YANG Fan, MA Shilong, WANG Huamao. A virtual test method for satellite system level verification and case study [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2017, 38(7): 320768-320768.
[12]	LIU Shuaiqi, HU Shaohai, XIAO Yang. SAR Image De-noising Based on Complex Shearlet Transform Domain Gaussian Mixture Model [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2013, 34(1): 173-180.
[13]	Zhang Lei;Li Xingshan;Yu Jinsong;Dai Jing. A Fault Prognostic Algorithm Based on Gaussian Mixture Model Particle Filter [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2009, 30(2): 319-324.
[14]	Lu Feng;Huang Jinquan. Engine Component Performance Prognostics Based on Decision Fusion [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2009, 30(10): 1795-1800.

Numerical method of data-driven rarefied nonlinear constitutive relations coupled with clustering

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 18

References 29

Related Articles 14

Recommended Articles

Metrics

Comments