Mingyang Wang, Enzhi Wang, Xiaoli Liu, Congcong Wang
{"title":"基于循环一致对抗网络和少量学习的瓦瑟斯坦距离引导的三维多孔介质重建技术","authors":"Mingyang Wang, Enzhi Wang, Xiaoli Liu, Congcong Wang","doi":"10.1007/s11242-024-02099-4","DOIUrl":null,"url":null,"abstract":"<div><p>Numerical simulation studies of water–rock interaction mechanisms and pore-scale multiphase flow properties often require high computational efficiency and realistic geometries to enable a fast and accurate description of hydrodynamic behavior. In this paper, we have chosen to use deep learning models to achieve these requirements, firstly by using encoder structures to refine the image segmentation of void-solid structures on complex geometries of scanning electron microscopy (SEM) images of porous media through few-shot learning (FSL), not only obtaining an accuracy of 0.97, but also reducing the amount of annotation work. We then focus on pore-scale three-dimensional (3D) structural reconstruction using the unpaired image-to-image translation method, optimizing the cycle-consistent adversarial network (cycle-GAN) model via sliced Wasserstein distance (SWD) to transfer marine sedimentary sandstone features to the initial image, and the geometric stochastic reconstruction problems are transformed into optimization problems. Subsequently, the computational efficiency was improved by a factor of 21 by implementing the lattice Boltzmann simulation method (LBM) accelerated by GPU through compute-unified device architecture (CUDA). The flow field distribution and absolute permeability of the extracted 2D samples and the reconstructed 3D porous media structure were simulated. The results showed that our method could rapidly and accurately reconstruct the 3D structures of a given feature, ensuring statistical equivalence between the 3D reconstructed structures and 2D samples. We solve the problem of extrapolation-based 3D reconstruction of porous media and significantly reduce the time spent on structure extraction and numerical calculations.</p></div>","PeriodicalId":804,"journal":{"name":"Transport in Porous Media","volume":null,"pages":null},"PeriodicalIF":2.7000,"publicationDate":"2024-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Sliced Wasserstein Distance-Guided Three-Dimensional Porous Media Reconstruction Based on Cycle-Consistent Adversarial Network and Few-Shot Learning\",\"authors\":\"Mingyang Wang, Enzhi Wang, Xiaoli Liu, Congcong Wang\",\"doi\":\"10.1007/s11242-024-02099-4\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>Numerical simulation studies of water–rock interaction mechanisms and pore-scale multiphase flow properties often require high computational efficiency and realistic geometries to enable a fast and accurate description of hydrodynamic behavior. In this paper, we have chosen to use deep learning models to achieve these requirements, firstly by using encoder structures to refine the image segmentation of void-solid structures on complex geometries of scanning electron microscopy (SEM) images of porous media through few-shot learning (FSL), not only obtaining an accuracy of 0.97, but also reducing the amount of annotation work. We then focus on pore-scale three-dimensional (3D) structural reconstruction using the unpaired image-to-image translation method, optimizing the cycle-consistent adversarial network (cycle-GAN) model via sliced Wasserstein distance (SWD) to transfer marine sedimentary sandstone features to the initial image, and the geometric stochastic reconstruction problems are transformed into optimization problems. Subsequently, the computational efficiency was improved by a factor of 21 by implementing the lattice Boltzmann simulation method (LBM) accelerated by GPU through compute-unified device architecture (CUDA). The flow field distribution and absolute permeability of the extracted 2D samples and the reconstructed 3D porous media structure were simulated. The results showed that our method could rapidly and accurately reconstruct the 3D structures of a given feature, ensuring statistical equivalence between the 3D reconstructed structures and 2D samples. We solve the problem of extrapolation-based 3D reconstruction of porous media and significantly reduce the time spent on structure extraction and numerical calculations.</p></div>\",\"PeriodicalId\":804,\"journal\":{\"name\":\"Transport in Porous Media\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":2.7000,\"publicationDate\":\"2024-07-02\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Transport in Porous Media\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://link.springer.com/article/10.1007/s11242-024-02099-4\",\"RegionNum\":3,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ENGINEERING, CHEMICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Transport in Porous Media","FirstCategoryId":"5","ListUrlMain":"https://link.springer.com/article/10.1007/s11242-024-02099-4","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, CHEMICAL","Score":null,"Total":0}
Sliced Wasserstein Distance-Guided Three-Dimensional Porous Media Reconstruction Based on Cycle-Consistent Adversarial Network and Few-Shot Learning
Numerical simulation studies of water–rock interaction mechanisms and pore-scale multiphase flow properties often require high computational efficiency and realistic geometries to enable a fast and accurate description of hydrodynamic behavior. In this paper, we have chosen to use deep learning models to achieve these requirements, firstly by using encoder structures to refine the image segmentation of void-solid structures on complex geometries of scanning electron microscopy (SEM) images of porous media through few-shot learning (FSL), not only obtaining an accuracy of 0.97, but also reducing the amount of annotation work. We then focus on pore-scale three-dimensional (3D) structural reconstruction using the unpaired image-to-image translation method, optimizing the cycle-consistent adversarial network (cycle-GAN) model via sliced Wasserstein distance (SWD) to transfer marine sedimentary sandstone features to the initial image, and the geometric stochastic reconstruction problems are transformed into optimization problems. Subsequently, the computational efficiency was improved by a factor of 21 by implementing the lattice Boltzmann simulation method (LBM) accelerated by GPU through compute-unified device architecture (CUDA). The flow field distribution and absolute permeability of the extracted 2D samples and the reconstructed 3D porous media structure were simulated. The results showed that our method could rapidly and accurately reconstruct the 3D structures of a given feature, ensuring statistical equivalence between the 3D reconstructed structures and 2D samples. We solve the problem of extrapolation-based 3D reconstruction of porous media and significantly reduce the time spent on structure extraction and numerical calculations.
期刊介绍:
-Publishes original research on physical, chemical, and biological aspects of transport in porous media-
Papers on porous media research may originate in various areas of physics, chemistry, biology, natural or materials science, and engineering (chemical, civil, agricultural, petroleum, environmental, electrical, and mechanical engineering)-
Emphasizes theory, (numerical) modelling, laboratory work, and non-routine applications-
Publishes work of a fundamental nature, of interest to a wide readership, that provides novel insight into porous media processes-
Expanded in 2007 from 12 to 15 issues per year.
Transport in Porous Media publishes original research on physical and chemical aspects of transport phenomena in rigid and deformable porous media. These phenomena, occurring in single and multiphase flow in porous domains, can be governed by extensive quantities such as mass of a fluid phase, mass of component of a phase, momentum, or energy. Moreover, porous medium deformations can be induced by the transport phenomena, by chemical and electro-chemical activities such as swelling, or by external loading through forces and displacements. These porous media phenomena may be studied by researchers from various areas of physics, chemistry, biology, natural or materials science, and engineering (chemical, civil, agricultural, petroleum, environmental, electrical, and mechanical engineering).