arXiv subjects

Sangseung Lee

Publications and source records attributed to Sangseung Lee.

14 recordsLinked to original sources

arXiv2026

UrbanFlow-3K: A Dataset of 3,000 Lattice-Boltzmann Simulations of Random Building Layouts

The analysis of flow around buildings has gained significant research interest across various domains, including pedestrian safety, pollutant dispersion, natural ventilation, and building energy efficiency. While these domains frequently include high-resolution computational fluid dynamics (CFD) data, predicting urban flow fields with machine learning (ML) models has emerged as a promising approach to overcome the prohibitive costs of CFD simulations. However, the availability of open-source datasets for training such ML models remains scarce. In particular, publicly available two-dimensional datasets of urban flow fields are nearly non-existent, despite their potential value for early development and debugging stages of data-driven models, before scaling to computationally expensive three-dimensional datasets. To bridge this gap, this study presents a comprehensive dataset consisting of 3,000 two-dimensional urban flow simulations conducted using a lattice-Boltzmann method across three distinct Reynolds numbers. The dataset contains the time-averaged velocity fields. A key feature of this dataset is its high geometric diversity: each layout incorporates between three and six buildings with randomized sizes, positions, and rotation angles ranging from 0° to 90°. This extensive variability enables the dataset to capture several critical flow characteristics, including wake formation, flow acceleration, shielding effects, and recirculation zones, across a wide range of orchestrated urban canopies. The large sample size and consistent simulation setup make the dataset particularly suitable for developing and benchmarking ML architectures. In addition, the dataset can support transfer-learning strategies in which models trained on large two-dimensional datasets are adapted to smaller and more computationally expensive three-dimensional datasets.

physics.flu-dyn↗

arXiv2025

Comparative Analysis of the Flow in a Realistic Human Airway

Accurate simulations of the flow in the human airway are essential for advancing diagnostic methods. Many existing computational studies rely on simplified geometries or turbulence models, limiting their simulation's ability to resolve flow features such shear-layer instabilities or secondary vortices. In this study, direct numerical simulations were performed for inspiratory flow through a detailed airway model which covers the nasal mask region to the 6th bronchial bifurcation. Simulations were conducted at two physiologically relevant \textsc{Reynolds} numbers with respect to the pharyngeal diameter, i.e., at Re_p=400 (resting) and Re_p=1200 (elevated breathing). These values characterize resting and moderately elevated breathing conditions. A lattice-Boltzmann method was employed to directly simulate the flow, i.e., no turbulence model was used. The flow field was examined across four anatomical regions: 1) the nasal cavity, 2) the naso- and oropharynx, 3) the laryngopharynx and larynx, and 4) the trachea and carinal bifurcation. The total pressure loss increased from 9.76 Pa at Re_p=400 to 41.93 Pa at Re_p=1200. The nasal cavity accounted for the majority of this loss for both Reynolds numbers, though its relative contribution decreased from 81.3% at Re_p=400 to 73.4% at Re_p=1200. At Re_p=1200, secondary vortices in the nasopharyngeal bend and turbulent shear-layers in the glottis jet enhanced the local pressure losses. In contrast, the carinal bifurcation mitigated upstream unsteadiness and stabilized the flow. A key outcome is the spatial correlation between the pressure loss and the onset of flow instabilities across the four regions. This yields a novel perspective on how the flow resistance and vortex dynamics vary with geometric changes and flow rate.

physics.flu-dyn↗

arXiv2025

Engineering application of physics-informed neural networks for Saint-Venant torsion

The Saint-Venant torsion theory is a classical theory for analyzing the torsional behavior of structural components, and it remains critically important in modern computational design workflows. Conventional numerical methods, including the finite element method (FEM), typically rely on mesh-based approaches to obtain approximate solutions. However, these methods often require complex and computationally intensive techniques to overcome the limitations of approximation, leading to significant increases in computational cost. The objective of this study is to develop a series of novel numerical methods based on physics-informed neural networks (PINN) for solving the Saint-Venant torsion equations. Utilizing the expressive power and the automatic differentiation capability of neural networks, the PINN can solve partial differential equations (PDEs) along with boundary conditions without the need for intricate computational techniques. First, a PINN solver was developed to compute the torsional constant for bars with arbitrary cross-sectional geometries. This was followed by the development of a solver capable of handling cases with sharp geometric transitions; variable-scaling PINN (VS-PINN). Finally, a parametric PINN was constructed to address the limitations of conventional single-instance PINN. The results from all three solvers showed good agreement with reference solutions, demonstrating their accuracy and robustness. Each solver can be selectively utilized depending on the specific requirements of torsional behavior analysis.

cs.LG↗

arXiv2024

Data-driven discovery of drag-inducing elements on a rough surface through convolutional neural networks

Understanding the influence of surface roughness on drag forces remains a significant challenge in fluid dynamics. This paper presents a convolutional neural network (CNN) that predicts drag solely by the topography of rough surfaces and is capable of discovering spatial patterns linked to drag-inducing structures. A CNN model was developed to analyze spatial information from the topography of a rough surface and predict the roughness function, $ΔU^+$, obtained from direct numerical simulation. This model enables the prediction of drag from rough surface data alone, which was not possible with previous methods owing to the large number of surface-derived parameters. Additionally, the retention of spatial information by the model enables the creation of a feature map that accentuates critical areas for drag prediction on rough surfaces. By interpreting the feature maps, we show that the developed CNN model is able to discover spatial patterns associated with drag distributions across rough surfaces, even without a direct training on drag distribution data. The analysis of the feature map indicates that, even without flow field information, the CNN model extracts the importance of the flow-directional slope and height of roughness elements as key factors in inducing pressure drag. This study demonstrates that CNN-based drag prediction is grounded in physical principles of fluid dynamics, underscoring the utility of CNNs in both predicting and understanding drag on rough surfaces.

physics.flu-dyn↗

arXiv2024

Drag prediction of rough-wall turbulent flow using data-driven regression

Efficient tools for predicting the drag of rough walls in turbulent flows would have a tremendous impact. However, methods for drag prediction rely on experiments or numerical simulations which are costly and time-consuming. Data-driven regression methods have the potential to provide a prediction that is accurate and fast. We assess the performance and limitations of linear regression, kernel methods and neural networks for drag prediction using a database of 1000 homogeneous rough surfaces. Model performance is evaluated using the roughness function obtained at friction-scaled Reynolds number 500. With two trainable parameters, the kernel method can fully account for nonlinear relations between $ΔU^+$ and surface statistics (roughness height, effective slope, skewness, etc). In contrast, linear regression cannot account for nonlinear correlations and display large errors and high uncertainty. Multilayer perceptron and convolutional neural networks demonstrate performance on par with the kernel method but have orders of magnitude more trainable parameters. For the current database size, the networks' capacity cannot be fully exploited, resulting in reduced generalizability and reliability. Our study provides insight into the appropriateness of different regression models for drag prediction. We also discuss the remaining steps before data-driven methods emerge as useful tools in applications.

physics.flu-dyn↗

arXiv2023

Prediction of equivalent sand-grain size and identification of drag-relevant scales of roughness -- a data driven approach

The purpose of the present work is to examine two possibilities; firstly, predicting equivalent sand-grain roughness size $k_s$ based on the roughness height probability density function and power spectrum leveraging machine learning as a regression tool, and secondly, extracting information about relevance of different roughness scales to skin-friction drag by interpreting the output of the trained data-driven model. The model is an ensemble neural network consisting of 50 deep neural networks. The data for the training of the model is obtained from direct numerical simulations (DNSs) of turbulent flow in plane channels over 85 irregular multi-scale roughness samples at friction Reynolds number Re$_τ=800$. The 85 roughness samples are selected from a repository of 4200 samples, covering a wide parameter space, through an active learning (AL) framework. The selection is made in several iterations, based on the informativeness of samples in the repository, quantified by the variance of ENN predictions. This AL framework aims to maximize the generalizability of the predictions with a certain amount of data. This is examined using three different testing data sets with different types of roughness, including 21 surfaces from the literature. The model yields an overall mean error of 5\% to 10\% on different testing data sets. Subsequently, a data interpretation technique, known as layer-wise relevance propagation, is applied to measure the contributions of different roughness wave-lengths to the predicted $k_s$. High-pass filtering is then applied to the roughness PS to exclude the wave-numbers identified as drag-irrelevant. The filtered rough surfaces are investigated using DNS, and it is demonstrated that, despite significant impact of filtering on the roughness topographical appearance and statistics, the skin-friction coefficient of the original roughness is successfully preserved.

physics.flu-dyn↗

arXiv2023

Effects of spatiotemporal correlations in wind data on neural network-based wind predictions

This paper investigates the influence of incorporating spatiotemporal wind data on the performance of wind forecasting neural networks. While previous studies have shown that including spatial data enhances the accuracy of such models, limited research has explored the impact of different spatial and temporal scales of input wind data on the learnability of neural network models. In this study, convolutional neural networks (CNNs) are employed and trained using various scales of spatiotemporal wind data. The research demonstrates that using spatiotemporally correlated data from the surrounding area and past time steps for training a CNN favorably affects the predictive performance of the model. The study proposes correlation analyses, including autocorrelation and Pearson correlation analyses, to unveil the influence of spatiotemporal wind characteristics on the predictive performance of different CNN models. The spatiotemporal correlations and performances of CNN models are investigated in three regions: Korea, the USA, and the UK. The findings reveal that regions with smaller deviations of autocorrelation coefficients (ACC) are more favorable for CNNs to learn the regional and seasonal wind characteristics. Specifically, the regions of Korea, the USA, and the UK exhibit maximum standard deviations of ACCs of 0.100, 0.043, and 0.023, respectively. The CNNs wind prediction performances follow the reverse order of the regions: UK, USA, and Korea. This highlights the significant impact of regional and seasonal wind conditions on the performance of the prediction models.

cs.LG↗

arXiv2021

Predicting drag on rough surfaces by transfer learning of empirical correlations

Recent developments in neural networks have shown the potential of estimating drag on irregular rough surfaces. Nevertheless, the difficulty of obtaining a large high-fidelity dataset to train neural networks is deterring their use in practical applications. In this study, we propose a transfer learning framework to model the drag on irregular rough surfaces even with a limited amount of direct numerical simulations. We show that transfer learning of empirical correlations, reported in the literature, can significantly improve the performance of neural networks for drag prediction. This is because empirical correlations include `approximate knowledge' of the drag dependency in high-fidelity physics. The `approximate knowledge' allows neural networks to learn the surface statistics known to affect drag more efficiently. The developed framework can be applied to applications where acquiring a large dataset is difficult, but empirical correlations have been reported.

physics.flu-dyn↗

arXiv2019

Mechanisms of a Convolutional Neural Network for Learning Three-dimensional Unsteady Wake Flow

Convolutional neural networks (CNNs) have recently been applied to predict or model fluid dynamics. However, mechanisms of CNNs for learning fluid dynamics are still not well understood, while such understanding is highly necessary to optimize the network or to reduce trial-and-errors during the network optmization. In the present study, a CNN to predict future three-dimensional unsteady wake flow using flow fields in the past occasions is developed. Mechanisms of the developed CNN for prediction of wake flow behind a circular cylinder are investigated in two flow regimes: the three-dimensional wake transition regime and the shear-layer transition regime. Feature maps in the CNN are visualized to compare flow structures which are extracted by the CNN from flow at the two flow regimes. In both flow regimes, feature maps are found to extract similar sets of flow structures such as braid shear-layers and shedding vortices. A Fourier analysis is conducted to investigate mechanisms of the CNN for predicting wake flow in flow regimes with different wave number characteristics. It is found that a convolution layer in the CNN integrates and transports wave number information from flow to predict the dynamics. Characteristics of the CNN for transporting input information including time histories of flow variables is analyzed by assessing contributions of each flow variable and time history to feature maps in the CNN. Structural similarities between feature maps in the CNN are calculated to reveal the number of feature maps that contain similar flow structures. By reducing the number of feature maps that contain similar flow structures, it is also able to successfully reduce the number of parameters to learn in the CNN by 85\% without affecting prediction performances.

physics.flu-dyn↗

arXiv2019

Data-driven prediction of unsteady flow fields over a circular cylinder using deep learning

Unsteady flow fields over a circular cylinder are trained and predicted using four different deep learning networks: convolutional neural networks with and without consideration of conservation laws, generative adversarial networks with and without consideration of conservation laws. Flow fields at future occasions are predicted based on information of flow fields at previous occasions. Deep learning networks are trained first using flow fields at Reynolds numbers of 100, 200, 300, and 400, while flow fields at Reynolds numbers of 500 and 3000 are predicted using the trained deep learning networks. Physical loss functions are proposed to explicitly impose information of conservation of mass and momentum to deep learning networks. An adversarial training is applied to extract features of flow dynamics in an unsupervised manner. Effects of the proposed physical loss functions, adversarial training, and network sizes on the prediction accuracy are analyzed. Predicted flow fields using deep learning networks are in favorable agreement with flow fields computed by numerical simulations.

physics.flu-dyn↗

arXiv2018

Prediction of typhoon tracks using a generative adversarial network with observational and meteorological data

Tracks of typhoons are predicted using a generative adversarial network (GAN) with observational data in form of satellite images and meteorological data from a reanalysis database. Time series of images of typhoons which occurred in the Korean Peninsula in the past are used to train the neural network. The trained GAN is employed to produce a 6-hour-advance track of a typhoon for which the GAN was not trained. The predicted image favorably identifies the future location of the typhoon center as well as the deformed cloud structures. The errors between predicted and real typhoon centers are measured quantitatively in kilometers. 65.5 % of all typhoon center predictions have an error of less than 80 km, 31.5 % lie within a range of 80 - 120 km and the remaining 3.0 % are above 120 km. The overall error is 67.2 km, compared to 95.6 km when only observational data are used as input. The cloud structure prediction is evaluated qualitatively. It is shown that the GAN is able to predict trends in cloud motion. It is found that adding physically meaningful meteorological data to satellite images improves the sharpness of predicted images.

physics.ao-ph↗

arXiv2018

Deep learning approach in multi-scale prediction of turbulent mixing-layer

Achievement of solutions in Navier-Stokes equation is one of challenging quests, especially for its closure problem. For achievement of particular solutions, there are variety of numerical simulations including Direct Numerical Simulation (DNS) or Large Eddy Simulation (LES). These methods analyze flow physics through efficient reduced-order modeling such as proper orthogonal decomposition or Koopman method, showing prominent fidelity in fluid dynamics. Generative adversarial network (GAN) is a reprint of neurons in brain as combinations of linear operations, using competition between generator and discriminator. Current paper propose deep learning network for prediction of small-scale movements with large-scale inspections only, using GAN. Therefore DNS result of three-dimensional mixing-layer was filtered blurring out the small-scaled structures, then is predicted of its detailed structures, utilizing Generative Adversarial Network (GAN). This enables multi-resolution analysis being asked to predict fine-resolution solution with only inspection of blurry one. Within the grid scale, current paper present deep learning approach of modeling small scale features in turbulent flow. The presented method is expected to have its novelty in utilization of unprocessed simulation data, achievement of 3D structures in prediction by processing 3D convolutions, and predicting precise solution with less computational costs.

physics.comp-ph↗

arXiv2018

Typhoon track prediction using satellite images in a Generative Adversarial Network

Tracks of typhoons are predicted using satellite images as input for a Generative Adversarial Network (GAN). The satellite images have time gaps of 6 hours and are marked with a red square at the location of the typhoon center. The GAN uses images from the past to generate an image one time step ahead. The generated image shows the future location of the typhoon center, as well as the future cloud structures. The errors between predicted and real typhoon centers are measured quantitatively in kilometers. 42.4% of all typhoon center predictions have absolute errors of less than 80 km, 32.1% lie within a range of 80 - 120 km and the remaining 25.5% have accuracies above 120 km. The relative error sets the above mentioned absolute error in relation to the distance that has been traveled by a typhoon over the past 6 hours. High relative errors are found in three types of situations, when a typhoon moves on the open sea far away from land, when a typhoon changes its course suddenly and when a typhoon is about to hit the mainland. The cloud structure prediction is evaluated qualitatively. It is shown that the GAN is able to predict trends in cloud motion. In order to improve both, the typhoon center and cloud motion prediction, the present study suggests to add information about the sea surface temperature, surface pressure and velocity fields to the input data.

physics.ao-ph↗

arXiv2017

Prediction of laminar vortex shedding over a cylinder using deep learning

Unsteady laminar vortex shedding over a circular cylinder is predicted using a deep learning technique, a generative adversarial network (GAN), with a particular emphasis on elucidating the potential of learning the solution of the Navier-Stokes equations. Numerical simulations at two different Reynolds numbers with different time-step sizes are conducted to produce training datasets of flow field variables. Unsteady flow fields in the future at a Reynolds number which is not in the training datasets are predicted using a GAN. Predicted flow fields are found to qualitatively and quantitatively agree well with flow fields calculated by numerical simulations. The present study suggests that a deep learning technique can be utilized for prediction of laminar wake flow in lieu of solving the Navier-Stokes equations.

physics.flu-dyn↗