{"title":"深度矢量量化器的降维湍流数据","authors":"M. Momenifar, Enmao Diao, V. Tarokh, A. Bragg","doi":"10.1080/14685248.2022.2060508","DOIUrl":null,"url":null,"abstract":"Analysing large-scale data from simulations of turbulent flows is memory intensive, requiring significant resources. This major challenge highlights the need for data compression techniques. In this study, we apply a physics-informed Deep Learning technique based on vector quantisation to generate a discrete, low-dimensional representation of data from simulations of three-dimensional turbulent flows. The deep learning framework is composed of convolutional layers and incorporates physical constraints on the flow, such as preserving incompressibility and global statistical characteristics of the velocity gradients. The accuracy of the model is assessed using statistical, comparison-based similarity and physics-based metrics. The training data set is produced from Direct Numerical Simulation of an incompressible, statistically stationary, isotropic turbulent flow. The performance of this lossy data compression scheme is evaluated not only with unseen data from the stationary, isotropic turbulent flow, but also with data from decaying isotropic turbulence, a Taylor–Green vortex flow, and a turbulent channel flow. Defining the compression ratio (CR) as the ratio of original data size to the compressed one, the results show that our model based on vector quantisation can offer CR with a mean square error (MSE) of , and predictions that faithfully reproduce the statistics of the flow, except at the very smallest scales where there is some loss. Compared to the recent study of Glaws. et al. [Deep learning for in situ data compression of large turbulent flow simulations. Phys Rev Fluids. 2020;5(11):114602], which was based on a conventional autoencoder (where compression is performed in a continuous space), our model improves the CR by more than 30%, and reduces the MSE by an order of magnitude. Our compression model is an attractive solution for situations where fast, high quality and low-overhead encoding and decoding of large data are required.","PeriodicalId":49967,"journal":{"name":"Journal of Turbulence","volume":null,"pages":null},"PeriodicalIF":1.5000,"publicationDate":"2021-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"8","resultStr":"{\"title\":\"Dimension reduced turbulent flow data from deep vector quantisers\",\"authors\":\"M. Momenifar, Enmao Diao, V. Tarokh, A. Bragg\",\"doi\":\"10.1080/14685248.2022.2060508\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Analysing large-scale data from simulations of turbulent flows is memory intensive, requiring significant resources. This major challenge highlights the need for data compression techniques. In this study, we apply a physics-informed Deep Learning technique based on vector quantisation to generate a discrete, low-dimensional representation of data from simulations of three-dimensional turbulent flows. The deep learning framework is composed of convolutional layers and incorporates physical constraints on the flow, such as preserving incompressibility and global statistical characteristics of the velocity gradients. The accuracy of the model is assessed using statistical, comparison-based similarity and physics-based metrics. The training data set is produced from Direct Numerical Simulation of an incompressible, statistically stationary, isotropic turbulent flow. The performance of this lossy data compression scheme is evaluated not only with unseen data from the stationary, isotropic turbulent flow, but also with data from decaying isotropic turbulence, a Taylor–Green vortex flow, and a turbulent channel flow. Defining the compression ratio (CR) as the ratio of original data size to the compressed one, the results show that our model based on vector quantisation can offer CR with a mean square error (MSE) of , and predictions that faithfully reproduce the statistics of the flow, except at the very smallest scales where there is some loss. Compared to the recent study of Glaws. et al. [Deep learning for in situ data compression of large turbulent flow simulations. Phys Rev Fluids. 2020;5(11):114602], which was based on a conventional autoencoder (where compression is performed in a continuous space), our model improves the CR by more than 30%, and reduces the MSE by an order of magnitude. Our compression model is an attractive solution for situations where fast, high quality and low-overhead encoding and decoding of large data are required.\",\"PeriodicalId\":49967,\"journal\":{\"name\":\"Journal of Turbulence\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":1.5000,\"publicationDate\":\"2021-03-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"8\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Turbulence\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://doi.org/10.1080/14685248.2022.2060508\",\"RegionNum\":4,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"MECHANICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Turbulence","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1080/14685248.2022.2060508","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"MECHANICS","Score":null,"Total":0}
Dimension reduced turbulent flow data from deep vector quantisers
Analysing large-scale data from simulations of turbulent flows is memory intensive, requiring significant resources. This major challenge highlights the need for data compression techniques. In this study, we apply a physics-informed Deep Learning technique based on vector quantisation to generate a discrete, low-dimensional representation of data from simulations of three-dimensional turbulent flows. The deep learning framework is composed of convolutional layers and incorporates physical constraints on the flow, such as preserving incompressibility and global statistical characteristics of the velocity gradients. The accuracy of the model is assessed using statistical, comparison-based similarity and physics-based metrics. The training data set is produced from Direct Numerical Simulation of an incompressible, statistically stationary, isotropic turbulent flow. The performance of this lossy data compression scheme is evaluated not only with unseen data from the stationary, isotropic turbulent flow, but also with data from decaying isotropic turbulence, a Taylor–Green vortex flow, and a turbulent channel flow. Defining the compression ratio (CR) as the ratio of original data size to the compressed one, the results show that our model based on vector quantisation can offer CR with a mean square error (MSE) of , and predictions that faithfully reproduce the statistics of the flow, except at the very smallest scales where there is some loss. Compared to the recent study of Glaws. et al. [Deep learning for in situ data compression of large turbulent flow simulations. Phys Rev Fluids. 2020;5(11):114602], which was based on a conventional autoencoder (where compression is performed in a continuous space), our model improves the CR by more than 30%, and reduces the MSE by an order of magnitude. Our compression model is an attractive solution for situations where fast, high quality and low-overhead encoding and decoding of large data are required.
期刊介绍:
Turbulence is a physical phenomenon occurring in most fluid flows, and is a major research topic at the cutting edge of science and technology. Journal of Turbulence ( JoT) is a digital forum for disseminating new theoretical, numerical and experimental knowledge aimed at understanding, predicting and controlling fluid turbulence.
JoT provides a common venue for communicating advances of fundamental and applied character across the many disciplines in which turbulence plays a vital role. Examples include turbulence arising in engineering fluid dynamics (aerodynamics and hydrodynamics, particulate and multi-phase flows, acoustics, hydraulics, combustion, aeroelasticity, transitional flows, turbo-machinery, heat transfer), geophysical fluid dynamics (environmental flows, oceanography, meteorology), in physics (magnetohydrodynamics and fusion, astrophysics, cryogenic and quantum fluids), and mathematics (turbulence from PDE’s, model systems). The multimedia capabilities offered by this electronic journal (including free colour images and video movies), provide a unique opportunity for disseminating turbulence research in visually impressive ways.