A FLOW FIELD SUPER-RESOLUTION RECONSTRUCTION METHOD BASED ON DIFFUSION MODEL

Han Yang; Zhu Junpeng; Guo Chunyu; Fan Yiwei; Wang Yonghao

doi:10.6052/0459-1879-23-167

Volume 55 Issue 10

Oct. 2023

Turn off MathJax

Article Contents

Abstract

References

Chinese Journal of Theoretical and Applied Mechanics > 2023 > 55(10): 2309-2320. > DOI: 10.6052/0459-1879-23-167 CSTR: 32045.14.0459-1879-23-167

Han Yang, Zhu Junpeng, Guo Chunyu, Fan Yiwei, Wang Yonghao. A flow field super-resolution reconstruction method based on diffusion model. Chinese Journal of Theoretical and Applied Mechanics, 2023, 55(10): 2309-2320. DOI: 10.6052/0459-1879-23-167

Citation:

PDF (3193 KB)

A FLOW FIELD SUPER-RESOLUTION RECONSTRUCTION METHOD BASED ON DIFFUSION MODEL

*.
College of Shipbuilding Engineering, Harbin Engineering University, Harbin 150001, China
†.
Qingdao Innovation and Development Base, Harbin Engineering University, Qingdao 266000, Shandong, China

Received Date: May 03, 2023
Accepted Date: July 27, 2023
Available Online: July 28, 2023

Graphical Abstract

Abstract

Abstract

Low-resolution flow field data contains limited information, which fails to fully capture the detailed evolutionary processes of the flow field. Especially for the random turbulent features and small-scale vortex details in turbulence, they are even more challenging to obtain, thereby restricting the in-depth investigation of flow field evolution mechanisms. In order to address this limitation and reconstruct high-resolution data from low-resolution flow fields, this paper proposes a generative diffusion model called FlowDiffusionNet for flow field super-resolution reconstruction. The model takes the low-resolution flow field data input as the constraint condition, and utilizes a denoising fraction matching method to reproduce high-resolution flow field data. FlowDiffusionNet's structural design takes into consideration both the low-frequency information and high-frequency spatial features of flow field data, employing a diffusion-based modeling technique to reconstruct the residuals for high-resolution data. The proposed model's architecture is amenable to transfer learning, allowing its application to degraded flow fields at different levels. The performance of FlowDiffusionNet is evaluated on various classical flow field datasets and compared against other methods such as bicubic interpolation, super-resolution generative adversarial network (SRGAN), and super-resolution convolutional neural network (SRCNN). The results demonstrate that the proposed method achieves the best reconstruction performance on various flow fields, especially for flow field data with small-scale vortex structures down sampled by a factor of 4, where the objective evaluation index structural similarity index measure (SSIM) reaches 0.999.
- super-resolution reconstruction of flow field,
- conditional diffusion model,
- deep learning,
- flow field residuals,
- transfer learning

FullText(HTML)

References (36)

References

[1]	Yang L, Zhang ZL, Song Y, et al. Diffusion models: A comprehensive survey of methods and applications. arXiv, 2022: 2209.00796
[2]	Yu C, Bi X, Fan Y. Deep learning for fluid velocity field estimation: A review. Ocean Engineering, 2023, 271: 113693 doi: 10.1016/j.oceaneng.2023.113693
[3]	陈皓, 郭明明, 田野等. 卷积神经网络在流场重构研究中的进展. 力学学报, 2022, 54(9): 2343-2360 (Chen Hao, Guo Mingming, Tian Ye, et al. Progress of convolutional neural network in flow field reconstruction. Chinese Journal of Theoretical and Applied Mechanics, 2022, 54(9): 2343-2360 (in Chinese) Chen Hao, Guo Mingming, Tian Ye, et al. Progress of Convolutional Neural Network in Flow Field Reconstruction. Chinese Journal of Mechanics, 2022, 54(9): 2343-2360(in Chinese))
[4]	Guo X, Li W, Iorio F. Convolutional neural networks for steady flow approximation//22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco. Assoc Computing Machinery, 2016: 481-490
[5]	Lee Y, Yang H, Yin Z. PIV-DCNN: cascaded deep convolutional neural networks for particle image velocimetry. Experiments in Fluids, 2017, 58: 1-10 doi: 10.1007/s00348-016-2278-6
[6]	Cai S, Liang J, Gao Q, et al. Particle image velocimetry based on a deep learning motion estimator. IEEE Transactions on Instrumentation and Measurement, 2019, 69(6): 3538-3554
[7]	Lagemann C, Lagemann K, Mukherjee S, et al. Deep recurrent optical flow learning for particle image velocimetry data. Nature Machine Intelligence, 2021, 3(7): 641-651 doi: 10.1038/s42256-021-00369-0
[8]	Dong C, Loy CC, He K, et al. Image super-resolution using deep convolutional networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 38(2): 295-307
[9]	Kim J, Lee J K, Lee KM. Accurate image super-resolution using very deep convolutional networks//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 29th IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas. Institute of Electrical and Electronics Engineers Inc, 2016: 1646-1654
[10]	Ledig C, Theis L, Huszár F, et al. Photo-realistic single image super-resolution using a generative adversarial network//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 30th IEEE Conference on Computer Vision and Pattern Recognition, Honolulu. Institute of Electrical and Electronics Engineers Inc, 2017: 4681-4690
[11]	Tong T, Li G, Liu X, et al. Image super-resolution using dense skip connections//Proceedings of the IEEE international conference on computer vision, 16th IEEE International Conference on Computer Vision, Venice. Institute of Electrical and Electronics Engineers Inc, 2017: 4799-4807
[12]	Zhang Y, Li K, Li K, et al. Image super-resolution using very deep residual channel attention networks//Lecture Notes in Computer Science, 15th European Conference on Computer Vision (ECCV), Munich. Springer Verlag, 2018: 286-301
[13]	Lee J, Lee S, You D. Deep learning approach in multi-scale prediction of turbulent mixing-layer. arXiv, 2018: 1809.07021
[14]	Deng Z, He C, Liu Y, et al. Super-resolution reconstruction of turbulent velocity fields using a generative adversarial network-based artificial intelligence framework. Physics of Fluids, 2019, 31(12): 125111 doi: 10.1063/1.5127031
[15]	Wang X, Yu K, Wu S, et al. Esrgan: Enhanced super-resolution generative adversarial networks//Lecture Notes in Computer Science, 15th European Conference on Computer Vision, Munich, 2018. Springer Verlag, 2019: 63-79
[16]	Fukami K, Fukagata K, Taira K. Super-resolution reconstruction of turbulent flows with machine learning. Journal of Fluid Mechanics, 2019, 870: 106-120 doi: 10.1017/jfm.2019.238
[17]	Liu B, Tang J, Huang H, et al. Deep learning methods for super-resolution reconstruction of turbulent flows. Physics of Fluids, 2020, 32(2): 025105 doi: 10.1063/1.5140772
[18]	Kong C, Chang JT, Li YF, et al. Deep learning methods for super-resolution reconstruction of temperature fields in a supersonic combustor. AIP Advances, 2020, 10(11): 115021 doi: 10.1063/5.0030040
[19]	Ferdian E, Suinesiaputra A, Dubowitz DJ, et al. 4 DFlowNet: super-resolution 4D flow MRI using deep learning and computational fluid dynamics. Frontiers in Physics, 2020, 8: 138 doi: 10.3389/fphy.2020.00138
[20]	Fukami K, Fukagata K, Taira K. Machine-learning-based spatio-temporal super resolution reconstruction of turbulent flows. Journal of Fluid Mechanics, 2021, 909: A9 doi: 10.1017/jfm.2020.948
[21]	Bi X, Liu A, Fan Y, et al. FlowSRNet: A multi-scale integration network for super-resolution reconstruction of fluid flows. Physics of Fluids, 2022, 34(12): 127104 doi: 10.1063/5.0128435
[22]	Raissi M, Perdikaris P, Karniadakis GE. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. Journal of Computational Physics, 2019, 378: 686-707 doi: 10.1016/j.jcp.2018.10.045
[23]	Raissi M, Yazdani A, Karniadakis GE. Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations. Science, 2020, 367(6481): 1026-1030 doi: 10.1126/science.aaw4741
[24]	Shu D, Li Z, Farimani AB. A physics-informed diffusion model for high-fidelity flow field reconstruction. Journal of Computational Physics, 2023, 478: 111972 doi: 10.1016/j.jcp.2023.111972
[25]	Ho J, Jain A, Abbeel P. Denoising diffusion probabilistic models. Advances in Neural Information Processing Systems, 2020, 33: 6840-6851
[26]	Song Y, Sohl-Dickstein J, Kingma DP, et al. Score-based generative modeling through stochastic differential equations. arXiv, 2020: 2011.13456
[27]	Saharia C, Ho J, Chan W, et al. Image super-resolution via iterative refinement. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45(4): 4713-4726
[28]	Li H, Yang Y, Chang M, et al. Srdiff: Single image super-resolution with diffusion probabilistic models. Neurocomputing, 2022, 479: 47-59 doi: 10.1016/j.neucom.2022.01.029
[29]	Liu J, Tang J, Wu G. Residual feature distillation network for lightweight image super-resolution//Computer Vision–ECCV 2020 Workshops, Glasgow, 2020. Springer International Publishing, 2020: 41-55
[30]	Lu C, Zhou Y, Bao F, et al. Dpm-solver: A fast ode solver for diffusion probabilistic model sampling in around 10 steps. arXiv, 2022: 2206.00927
[31]	Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need//Proceedings of the 31st International Conference on Neural Information Processing Systems, December 2017: 6000-6010
[32]	Carlier J. Second set of fluid mechanics image sequences. European Project Fluid Image Analysis and Description, Project No. FP6-513663, 2005
[33]	Resseguier V, Mémin E, Chapron B. Geophysical flows under location uncertainty, Part II: Quasi-geostrophy and efficient ensemble spreading. Geophysical & Astrophysical Fluid Dynamics, 2017, 111(3): 177-208
[34]	Li Y, Perlman E, Wan M, et al. A public turbulence database cluster and applications to study Lagrangian evolution of velocity increments in turbulence. Journal of Turbulence, 2008, 9: N31
[35]	Song J, Meng C, Ermon S. Denoising diffusion implicit models. arXiv, 2020: 2010.02502
[36]	Keys R. Cubic convolution interpolation for digital image processing. IEEE Transactions on Acoustics, Speech, and Signal Processing, 1981, 29(6): 1153-1160 doi: 10.1109/TASSP.1981.1163711

[1]	Wen Guilin, Liu Jie, Chen Zijie, Wei Peng, Long Kai, Wang Hongxin, Rong Jianhua, Xie Yimin. A SURVEY OF NONLINEAR CONTINUUM TOPOLOGY OPTIMIZATION METHODS[J]. Chinese Journal of Theoretical and Applied Mechanics, 2022, 54(10): 2659-2675. DOI: 10.6052/0459-1879-22-179
[2]	Huang Bin, He Zhiyun, Zhang Heng. HYBRID PERTURBATION-GALERKIN METHOD FOR GEOMETRICAL NONLINEAR ANALYSIS OF TRUSS STRUCTURES WITH RANDOM PARAMETERS[J]. Chinese Journal of Theoretical and Applied Mechanics, 2019, 51(5): 1424-1436. DOI: 10.6052/0459-1879-19-099
[3]	Xingtian Liu, Shuhai Chen, Jiadeng Wang, Junfeng Shen. ANLYSIS OF THE DYNAMIC BEHAVIOR AND PERFORMANCE OF A VIBRATION ISOLATION SYSTEM WITH GEOMETRIC NONLINEAR FRICTION DAMPING[J]. Chinese Journal of Theoretical and Applied Mechanics, 2019, 51(2): 371-379. DOI: 10.6052/0459-1879-18-302
[4]	Wang Gang, Qi Zhaohui, Wang Jing. SUBSTRUCTURE METHODS OF GEOMETRIC NONLINEAR ANALYSIS FOR MEMBER STRUCTURES WITH HINGED SUPPORTS[J]. Chinese Journal of Theoretical and Applied Mechanics, 2014, 46(2): 273-283. DOI: 10.6052/0459-1879-13-345
[5]	Yijiang Peng, Yinghua Liu. Application of the base forces concept in geometrically nonlinear finite element method on complementary energy[J]. Chinese Journal of Theoretical and Applied Mechanics, 2008, 40(4): 496-501. DOI: 10.6052/0459-1879-2008-4-2007-340
[6]	Bin Huang, Jianchen Suo, Wenjun Huang. Geometrical nonlinear analysis of truss structures with random parameters utilizing recursive stochastic finite element method[J]. Chinese Journal of Theoretical and Applied Mechanics, 2007, 23(6): 835-842. DOI: 10.6052/0459-1879-2007-6-2007-165
[7]	TWO-NODE CURVED ELEMENT METHOD FOR GEOMETRICALLY NONLINEAR ANALYSIS OF TENSION STRUCTURES[J]. Chinese Journal of Theoretical and Applied Mechanics, 1999, 31(5): 633-639. DOI: 10.6052/0459-1879-1999-5-1995-075
[8]	DYNAMICS OF THIN ELASTIC PLATES IN LARGE OVERALL MOTIONS CONSIDERING GEOMETRIC NON-LINEARITY AND COUPLING DEFORMATION[J]. Chinese Journal of Theoretical and Applied Mechanics, 1999, 31(2): 243-249. DOI: 10.6052/0459-1879-1999-2-1995-025
[9]	A HIGHER ORDER NONLINEAR FEM FOR RC PLATES[J]. Chinese Journal of Theoretical and Applied Mechanics, 1998, 30(3): 379-384. DOI: 10.6052/0459-1879-1998-3-1995-140
[10]	GEOMETRICALLY NONLINEAR ANALYSES FOR 2-D PROBLEMS BASED ON THE INCOMPATIBLE FINITE ELEMENTS WITH INTERNAL PARAMETERS[J]. Chinese Journal of Theoretical and Applied Mechanics, 1993, 25(4): 505-513. DOI: 10.6052/0459-1879-1993-4-1995-673

Cited By

Cited by

Periodical cited type(7)

1.	江巍，尹豪，吴剑，汤艳春，李坤鹏，郑宏. 基于S-R和分解定理的二维几何非线性问题的虚单元法求解. 工程力学. 2024(08): 23-35 .
2.	宋彦琦，石博康，李向上. 基于S-R和分解定理的无网格法在功能梯度板中的应用. 上海大学学报(自然科学版). 2022(04): 702-714 .
3.	杨健生，曾治平，韦冬炎，彭林欣. 基于无网格法的非均匀弹性地基上变厚度加筋板弯曲与固有频率分析. 计算力学学报. 2021(03): 364-370 .
4.	覃霞，刘珊珊，谌亚菁，彭林欣. 基于遗传算法的弹性地基加肋板肋梁无网格优化分析. 力学学报. 2020(01): 93-110 . 本站查看
5.	肖国峰. 具有稳定数值解的三维谐振子. 计算力学学报. 2020(01): 119-130 .
6.	王莉华，李溢铭，褚福运. 基于分区径向基函数配点法的大变形分析. 力学学报. 2019(03): 743-753 . 本站查看
7.	刘斌，李冬明，谢佳萱. 非保守荷载大变形分析的无网格方法. 武汉理工大学学报. 2019(09): 71-77 .

Other cited types(5)

Get Citation

PDF

XML

Article Metrics

Article views (1066) PDF downloads (272) Cited by(12)

A FLOW FIELD SUPER-RESOLUTION RECONSTRUCTION METHOD BASED ON DIFFUSION MODEL

Abstract

References

Related Articles

Cited by

Periodical cited type(7)

Other cited types(5)

Catalog

Article Metrics

Related

Download Center

Author Center

A FLOW FIELD SUPER-RESOLUTION RECONSTRUCTION METHOD BASED ON DIFFUSION MODEL

Abstract

References

Related Articles

Cited by

Periodical cited type(7)

Other cited types(5)

Catalog

Article Metrics

Related

Download Center

Author Center

Export File

Citation

Format

Content