arXiv subjects

Sunwoong Yang

Publications and source records attributed to Sunwoong Yang.

18 recordsLinked to original sources

arXiv2026

Distance-Aware Attention and Wall-Distance Expert Routing for Transformer-Based 3D Flow Prediction

Transformer surrogates for 3D flow prediction compress an industrial mesh into a small set of tokens from which every prediction point reads. Two operations follow: the retrieval step in which a point gathers information from the compressed representation, and the feed-forward layer that transforms what it retrieved. In current backbones both are blind to where the point sits in the flow. We condition both on wall-related physical signals. Distance-aware cross-attention (DA-CA) reshapes each volume query by its wall distance before retrieval, so that a point deep in the boundary layer draws different geometric information than one in the outer flow. Surface-volume mixture-of-experts (SVMoE) replaces the shared feed-forward layer with a small set of experts, routed by wall distance for volume points and by local geometry for surface points. Neither mechanism is tied to one architecture, so we apply both unchanged to AB-UPT and Transolver-3. On DrivAerML with 50 training cases, DA-CA reduces the volume pressure error by 10.1%, and DA-CA and SVMoE together reduce it by 12.5%; DA-CA improves the near-wall region at some cost in the far region, which SVMoE recovers, and the volume experts settle into near-wall, transition, and free-stream bands without routing supervision. Retrained on 300 cases, the conditioning improves every field quantity, reducing volume pressure and velocity errors by 33.1% and 18.6% on AB-UPT and by 21.4% and 21.3% on Transolver-3. Under Leave-One-Body-Out evaluation on DrivAerNet++, it reduces the volume pressure error on unseen body types by up to 14.2%.

cs.AI↗

arXiv2026

Enhancing Creativity in 3D Generative Design via a TRIZ-Inspired Text-to-CAD Framework

Recent advances in large language models (LLMs) have demonstrated significant potential in supporting engineering design tasks, including computer-aided design (CAD) automation. However, most existing LLM-based 3D CAD generation approaches primarily focus on geometric precision and instruction-following performance, often overlooking the fundamental aspect of creative design exploration. This study presents a TRIZ-inspired text-to-CAD framework that leverages LLMs to generate high-quality, editable CAD models while systematically exploring creative design alternatives. The framework integrates the Theory of Inventive Problem Solving (TRIZ)-embedding deep human insights from extensive patent records-into LLM prompting strategies, enabling autonomous generation of innovative CAD variants that address technical contradictions. Through a comprehensive three-stage pipeline of design generation, enhancement, and optimization, the framework produces structurally diverse CAD models from well-crafted prompts. The present study implements and evaluates the first two stages, while positioning the design optimization stage as future work. A product design case study (chair) demonstrates that the TRIZ-inspired text-to-CAD framework generates multiple creative design alternatives by systematically applying TRIZ inventive principles such as segmentation, anti-weight, dynamics, and composite materials, achieving 4.0-14.7% mass reduction across all enhanced designs while maintaining structural integrity. The key findings suggest that integrating systematic innovation methodologies with LLM-based 3D CAD generation bridges the gap between precision-focused synthesis and creativity-focused exploration, advancing toward autonomous design systems where AI makes design decisions independently, supporting human decision-making in human-AI collaborative design for engineering applications.

cs.LG↗

arXiv2026

DeepJEB++: Foundation Model-Driven Large-Scale 3D Engineering Dataset via 2D Latent Space Augmentation

Data-driven engineering design is constrained by the lack of large-scale 3D datasets that pair geometry with physics-based performance labels. In particular, existing 3D data augmentation techniques have limitations in preserving subtle and diverse geometric variations, and it remains difficult to automate the subsequent simulation-labeling process, where boundary conditions vary depending on the generated geometry. We present DeepJEB++, a foundation-model-driven data-augmentation framework that expands a small seed set of jet engine brackets into a large, simulation-labeled 3D dataset under constrained resources. Our key idea is to augment in the data-rich 2D latent space, then transfer to 3D. In Stage 1, we fine-tune a pretrained 2D latent diffusion model on multi-view renders and synthesize novel views by latent interpolation, retaining manufacturable designs through a vision-language-model (VLM) quality filter. In Stage 2, the validated images are lifted to 3D meshes by a domain-adapted generative foundation model. In Stage 3, an automated pipeline recognizes the load and bolt interfaces on each mesh and assigns finite-element labels -- mass, stress, and displacement -- without manual intervention. We assess augmentation quality along three intrinsic axes: manufacturability, label fidelity against the SimJEB ground truth, and distributional consistency. Starting from fewer than 400 seed designs, DeepJEB++ yields 15,360 simulation-labeled 3D brackets -- a 40x expansion -- using a single GPU per stage. The dataset will be made publicly available to support reproducible engineering-AI research.

cs.LG↗

arXiv2026

A Multi-Agent System for Motor Design Optimization via an FEA-AI Hybrid Approach

This study presents a large language model (LLM)-based multi-agent framework for interior permanent magnet synchronous motor (IPMSM) design optimization that mitigates limitations of conventional workflows: expertise-dependent problem setup and data preparation, the prohibitive computational cost of finite element analysis (FEA), and the unreliability of AI surrogates in unexplored regions. To this end, we first introduce a Design agent that formulates the optimization problem in natural language, leveraging retrieval-augmented generation to improve answer accuracy on motor design problems from below 50% to 67-80%. Furthermore, a Training agent autonomously repairs improperly defined design spaces by reasoning over solver failure history, raising the success ratio of the geometry sampling from 28% to 84% for AI training. Additionally, to resolve cost and reliability simultaneously, an Optimization agent employs an uncertainty-aware FEA-AI hybrid model: the AI surrogate is the primary evaluator, and FEA is selectively invoked where predictive uncertainty is high. Under the same FEA budget, this hybrid model achieves up to 44% lower iron loss in single-objective and 22.5% higher hypervolume in multi-objective optimization than conventional FEA-only search. Under the same evaluation budget, it reduces computation time by 52-55% while retaining 90-92% of FEA-only hypervolume. Conversely, AI-only search converges to false optima, leaving half its Pareto designs infeasible. Notably, a controller agent adaptively updates the uncertainty threshold that triggers FEA each round, eliminating manual tuning and achieving 5.8% lower single objective iron loss than with a fixed threshold. These results establish domain specialized LLM agents with uncertainty-aware hybrid evaluation as a reliable, scalable paradigm for simulation-driven design automation.

cs.AI↗

arXiv2026

Terahertz magneto-nanoscopy of encapsulated monolayer graphene

This study investigates the nanoscale conductivity of encapsulated monolayer graphene at temperatures down to 5 K and magnetic fields of up to 1 T. We use the scattering-type scanning near-field optical microscopy (s-SNOM) technique to probe magnetic-field-dependent responses from graphene close to charge neutrality in the terahertz spectral region. We observe the near-perfect high-$q$ reflector behavior of graphene but with subtle changes by the presence of magnetic fields. Measurements align with calculations of the magneto-optical conductivity and the near-field spectroscopic contrast that describes the field-tunable cyclotron resonance of Dirac fermions. Our result provides an initial step toward understanding temperature and magnetic-field effects on nanoscale terahertz transport in two-dimensional quantum materials.

cond-mat.mes-hall↗

arXiv2025

Point-wise Diffusion Models for Physical Systems with Shape Variations: Application to Spatio-temporal and Large-scale system

This study introduces a novel point-wise diffusion model that processes spatio-temporal points independently to efficiently predict complex physical systems with shape variations. This methodological contribution lies in applying forward and backward diffusion processes at individual spatio-temporal points, coupled with a point-wise diffusion transformer architecture for denoising. Unlike conventional image-based diffusion models that operate on structured data representations, this framework enables direct processing of any data formats including meshes and point clouds while preserving geometric fidelity. We validate our approach across three distinct physical domains with complex geometric configurations: 2D spatio-temporal systems including cylinder fluid flow and OLED drop impact test, and 3D large-scale system for road-car external aerodynamics. To justify the necessity of our point-wise approach for real-time prediction applications, we employ denoising diffusion implicit models (DDIM) for efficient deterministic sampling, requiring only 5-10 steps compared to traditional 1000-step and providing computational speedup of 100 to 200 times during inference without compromising accuracy. In addition, our proposed model achieves superior performance compared to image-based diffusion model: reducing training time by 94.4% and requiring 89.0% fewer parameters while achieving over 28% improvement in prediction accuracy. Comprehensive comparisons against data-flexible surrogate models including DeepONet and Meshgraphnet demonstrate consistent superiority of our approach across all three physical systems. To further refine the proposed model, we investigate two key aspects: 1) comparison of final physical states prediction or incremental change prediction, and 2) computational efficiency evaluation across varying subsampling ratios (10%-100%).

physics.comp-ph↗

arXiv2025

Node Assigned physics-informed neural networks for thermal-hydraulic system simulation: CVH/FL module

Severe accidents (SAs) in nuclear power plants have been analyzed using thermal-hydraulic (TH) system codes such as MELCOR and MAAP. These codes efficiently simulate the progression of SAs, while they still have inherent limitations due to their inconsistent finite difference schemes. The use of empirical schemes incorporating both implicit and explicit formulations inherently induces unidirectional coupling in multi-physics analyses. The objective of this study is to develop a novel numerical method for TH system codes using physics-informed neural network (PINN). They have shown strength in solving multi-physics due to the innate feature of neural networks-automatic differentiation. We propose a node-assigned PINN (NA-PINN) that is suitable for the control volume approach-based system codes. NA-PINN addresses the issue of spatial governing equation variation by assigning an individual network to each nodalization of the system code, such that spatial information is excluded from both the input and output domains, and each subnetwork learns to approximate a purely temporal solution. In this phase, we evaluated the accuracy of the PINN methods for the hydrodynamic module. In the 6 water tank simulation, PINN and NA-PINN showed maximum absolute errors of 1.678 and 0.007, respectively. It should be noted that only NA-PINN demonstrated acceptable accuracy. To the best of the authors' knowledge, this is the first study to successfully implement a system code using PINN. Our future work involves extending NA-PINN to a multi-physics solver and developing it in a surrogate manner.

cs.LG↗

arXiv2025

Rigid-Deformation Decomposition AI Framework for 3D Spatio-Temporal Prediction of Vehicle Collision Dynamics

This study presents a rigid-deformation decomposition framework for vehicle collision dynamics that mitigates the spectral bias of implicit neural representations, that is, coordinate-based neural networks that directly map spatio-temporal coordinates to physical fields. We introduce a hierarchical architecture that decouples global rigid-body motion from local deformation using two scale-specific networks, denoted as RigidNet and DeformationNet. To enforce kinematic separation between the two components, we adopt a frozen-anchor training strategy combined with a quaternion-incremental scheme. This strategy alleviates the kinematic instability observed in joint training and yields a 29.8% reduction in rigid-body motion error compared with conventional direct prediction schemes. The stable rigid-body anchor improves the resolution of high-frequency structural buckling, which leads to a 17.2% reduction in the total interpolation error. Loss landscape analysis indicates that the decomposition smooths the optimization surface, which enhances robustness to distribution shifts in angular extrapolation and yields a 46.6% reduction in error. To assess physical validity beyond numerical accuracy, we benchmark the decomposed components against an oracle model that represents an upper bound on performance. The proposed framework recovers 92% of the directional correlation between rigid and deformation components and 96% of the spatial deformation localization accuracy relative to the oracle, while tracking the temporal energy dynamics with an 8 ms delay. These results demonstrate that rigid-deformation decomposition enables accurate and physically interpretable predictions for nonlinear collision dynamics.

cs.CE↗

arXiv2025

Data-Efficient Deep Operator Network for Unsteady Flow: A Multi-Fidelity Approach with Physics-Guided Subsampling

This study presents an enhanced multi-fidelity Deep Operator Network (DeepONet) framework for efficient spatio-temporal flow field prediction when high-fidelity data is scarce. Key innovations include: a merge network replacing traditional dot-product operations, achieving 50.4% reduction in prediction error and 7.57% accuracy improvement while reducing training time by 96%; a transfer learning multi-fidelity approach that freezes pre-trained low-fidelity networks while making only the merge network trainable, outperforming alternatives by up to 76% and achieving 43.7% better accuracy than single-fidelity training; and a physics-guided subsampling method that strategically selects high-fidelity training points based on temporal dynamics, reducing high-fidelity sample requirements by 40% while maintaining comparable accuracy. Comprehensive experiments across multiple resolutions and datasets demonstrate the framework's ability to significantly reduce required high-fidelity dataset size while maintaining predictive accuracy, with consistent superior performance against conventional benchmarks.

physics.flu-dyn↗

arXiv2025

AI-powered Digital Twin of the Ocean: Reliable Uncertainty Quantification for Real-time Wave Height Prediction with Deep Ensemble

Environmental pollution and fossil fuel depletion have prompted the need for renewable energy-based power generation. However, its stability is often challenged by low energy density and non-stationary conditions. Wave energy converters (WECs), in particular, need reliable real-time wave height prediction to address these issues caused by irregular wave patterns, which can lead to the inefficient and unstable operation of WECs. In this study, we propose an AI-powered reliable real-time wave height prediction model that integrates long short-term memory (LSTM) networks for temporal prediction with deep ensemble (DE) for robust uncertainty quantification (UQ), ensuring high accuracy and reliability. To further enhance the reliability, uncertainty calibration is applied, which has proven to significantly improve the quality of the quantified uncertainty. Using real operational data from an oscillating water column-wave energy converter (OWC-WEC) system in Jeju, South Korea, the model achieves notable accuracy (R2 > 0.9), while increasing uncertainty quality by over 50% through simple calibration technique. Furthermore, a comprehensive parametric study is conducted to explore the effects of key model hyperparameters, offering valuable guidelines for diverse operational scenarios, characterized by differences in wavelength, amplitude, and period. These results demonstrate the model's capability to deliver reliable predictions, facilitating digital twin of the ocean.

cs.LG↗

arXiv2024

Model-Agnostic AI Framework with Explicit Time Integration for Long-Term Fluid Dynamics Prediction

This study addresses the critical challenge of error accumulation in spatio-temporal auto-regressive (AR) predictions within scientific machine learning models by exploring temporal integration schemes and adaptive multi-step rollout strategies. We introduce the first implementation of the two-step Adams-Bashforth method specifically tailored for data-driven AR prediction, leveraging historical derivative information to enhance numerical stability without additional computational overhead. To validate our approach, we systematically evaluate time integration schemes across canonical 2D PDEs before extending to complex Navier-Stokes cylinder vortex shedding dynamics. Additionally, we develop three novel adaptive weighting strategies that dynamically adjust the importance of different future time steps during multi-step rollout training. Our analysis reveals that as physical complexity increases, such sophisticated rollout techniques become essential, with the Adams-Bashforth scheme demonstrating consistent robustness across investigated systems and our best adaptive approach delivering an 89% improvement over conventional fixed-weight methods while maintaining similar computational costs. For the complex Navier-Stokes vortex shedding problem, despite using an extremely lightweight graph neural network with just 1,177 trainable parameters and training on only 50 snapshots, our framework accurately predicts 350 future time steps reducing mean squared error from 0.125 (single-step direct prediction) to 0.002 (Adams-Bashforth with proposed multi-step rollout). Our integrated methodology demonstrates an 83% improvement over standard noise injection techniques and maintains robustness under severe spatial constraints; specifically, when trained on only a partial spatial domain, it still achieves 58% and 27% improvements over direct prediction and forward Euler methods, respectively.

cs.LG↗

arXiv2024

Physics-Constrained Graph Neural Networks for Spatio-Temporal Prediction of Drop Impact on OLED Display Panels

This study aims to predict the spatio-temporal evolution of physical quantities observed in multi-layered display panels subjected to the drop impact of a ball. To model these complex interactions, graph neural networks have emerged as promising tools, effectively representing objects and their relationships as graph structures. In particular, MeshGraphNets (MGNs) excel in capturing dynamics in dynamic physics simulations using irregular mesh data. However, conventional MGNs often suffer from non-physical artifacts, such as the penetration of overlapping objects. To resolve this, we propose a physics-constrained MGN that mitigates these penetration issues while maintaining high level of accuracy in temporal predictions. Furthermore, to enhance the model's robustness, we explore noise injection strategies with varying magnitudes and different combinations of targeted components, such as the ball, the plate, or both. In addition, our analysis on model stability in spatio-temporal predictions reveals that during the inference, deriving next time-step node positions by predicting relative changes (e.g., displacement or velocity) between the current and future states yields superior accuracy compared to direct absolute position predictions. This approach consistently shows greater stability and reliability in determining subsequent node positions across various scenarios. Building on this validated model, we evaluate its generalization performance by examining its ability to extrapolate with respect to design variables. Furthermore, the physics-constrained MGN serves as a near real-time emulator for the design optimization of multi-layered OLED display panels, where thickness variables are optimized to minimize stress in the light-emitting materials. It outperforms conventional MGN in optimization tasks, demonstrating its effectiveness for practical design applications.

physics.comp-ph↗

arXiv2024

Enhancing Graph U-Nets for Mesh-Agnostic Spatio-Temporal Flow Prediction

This study aims to overcome the limitations of conventional deep-learning approaches based on convolutional neural networks in complex geometries and unstructured meshes by exploring the potential of Graph U-Nets for unsteady flow-field prediction. We present a comprehensive investigation of Graph U-Nets, originally developed for classification tasks, now tailored for mesh-agnostic spatio-temporal forecasting of fluid dynamics. Our focus is on enhancing their performance through systematic hyperparameter tuning and architectural modifications. We propose novel approaches to improve mesh-agnostic spatio-temporal prediction of transient flow fields using Graph U-Nets, enabling accurate prediction on diverse mesh configurations. Key enhancements to the Graph U-Net architecture, including the Gaussian-mixture-model convolutional operator and noise injection approaches, provide increased flexibility in modeling node dynamics: the former reduces prediction error by 95\% compared to conventional convolutional operators, while the latter improves long-term prediction robustness, resulting in an error reduction of 86\%. We demonstrate the effectiveness of these enhancements in both transductive and inductive learning settings, showcasing the adaptability of Graph U-Nets to various flow conditions and mesh structures. This work contributes to the field of reduced-order modeling for computational fluid dynamics by establishing Graph U-Nets as a viable and flexible alternative to convolutional neural networks, capable of accurately and efficiently predicting complex fluid flow phenomena across diverse scenarios.

cs.LG↗

arXiv2024

Data-Driven Physics-Informed Neural Networks: A Digital Twin Perspective

This study explores the potential of physics-informed neural networks (PINNs) for the realization of digital twins (DT) from various perspectives. First, various adaptive sampling approaches for collocation points are investigated to verify their effectiveness in the mesh-free framework of PINNs, which allows automated construction of virtual representation without manual mesh generation. Then, the overall performance of the data-driven PINNs (DD-PINNs) framework is examined, which can utilize the acquired datasets in DT scenarios. Its scalability to more general physics is validated within parametric Navier-Stokes equations, where PINNs do not need to be retrained as the Reynolds number varies. In addition, since datasets can be often collected from different fidelity/sparsity in practice, multi-fidelity DD-PINNs are also proposed and evaluated. They show remarkable prediction performance even in the extrapolation tasks, with $42\sim62\%$ improvement over the single-fidelity approach. Finally, the uncertainty quantification performance of multi-fidelity DD-PINNs is investigated by the ensemble method to verify their potential in DT, where an accurate measure of predictive uncertainty is critical. The DD-PINN frameworks explored in this study are found to be more suitable for DT scenarios than traditional PINNs from the above perspectives, bringing engineers one step closer to seamless DT realization.

physics.flu-dyn↗

arXiv2024

Multi-objective Generative Design Framework and Realization for Quasi-serial Manipulator: Considering Kinematic and Dynamic Performance

This paper proposes a framework that optimizes the linkage mechanism of the quasi-serial manipulator for target tasks. This process is explained through a case study of 2-degree-of-freedom linkage mechanisms, which significantly affect the workspace of the quasi-serial manipulator. First, a vast quasi-serial mechanism is generated with a workspace satisfying a target task and it converts it into a 3D CAD model. Then, the workspace and required torque performance of each mechanism are evaluated through kinematic and dynamic analysis. A deep learning-based surrogate model is leveraged to efficiently predict mechanisms and performance during the optimization process. After model training, a multi-objective optimization problem is formulated under the mechanical and dynamic conditions of the manipulator. The design goal of the manipulator is to recommend quasi-serial mechanisms with optimized kinematic (workspace) and dynamic (joint torque) performance that satisfies the target task. To investigate the underlying physics from the obtained Pareto solutions, various data mining techniques are performed to extract design rules that can provide practical design guidance. Finally, the manipulator was designed in detail for realization with 3D printed parts, including topology optimization. Also, the task-based optimized manipulator is verified through a payload test. Based on these results, the proposed framework has the potential for other real applications as realized cases and provides a reasonable design plan through the design rule extraction.

cs.RO↗

arXiv2023

Towards Reliable Uncertainty Quantification via Deep Ensembles in Multi-output Regression Task

This study aims to comprehensively investigate the deep ensemble approach, an approximate Bayesian inference, in the multi-output regression task for predicting the aerodynamic performance of a missile configuration. To this end, the effect of the number of neural networks used in the ensemble, which has been blindly adopted in previous studies, is scrutinized. As a result, an obvious trend towards underestimation of uncertainty as it increases is observed for the first time, and in this context, we propose the deep ensemble framework that applies the post-hoc calibration method to improve its uncertainty quantification performance. It is compared with Gaussian process regression and is shown to have superior performance in terms of regression accuracy ($\uparrow55\sim56\%$), reliability of estimated uncertainty ($\uparrow38\sim77\%$), and training efficiency ($\uparrow78\%$). Finally, the potential impact of the suggested framework on the Bayesian optimization is briefly examined, indicating that deep ensemble without calibration may lead to unintended exploratory behavior. This UQ framework can be seamlessly applied and extended to any regression task, as no special assumptions have been made for the specific problem used in this study.

cs.LG↗

arXiv2022

Physics-aware Reduced-order Modeling of Transonic Flow via $β$-Variational Autoencoder

Autoencoder-based reduced-order modeling (ROM) has recently attracted significant attention, owing to its ability to capture underlying nonlinear features. However, two critical drawbacks severely undermine its scalability to various physical applications: entangled and therefore uninterpretable latent variables (LVs) and the blindfold determination of latent space dimension. In this regard, this study proposes the physics-aware ROM using only interpretable and information-intensive LVs extracted by $β$-variational autoencoder, which are referred to as physics-aware LVs throughout this paper. To extract these LVs, their independence and information intensity are quantitatively scrutinized in a two-dimensional transonic flow benchmark problem. Then, the physical meanings of the physics-aware LVs are thoroughly investigated and we confirmed that with appropriate hyperparameter $β$, they actually correspond to the generating factors of the training dataset, Mach number and angle of attack. To the best of the authors' knowledge, our work is the first to practically confirm that $β$-variational autoencoder can automatically extract the physical generating factors in the field of applied physics. Finally, physics-aware ROM, which utilizes only physics-aware LVs, is compared with conventional ROMs, and its validity and efficiency are successfully verified.

physics.flu-dyn↗

arXiv2022

Inverse design optimization framework via a two-step deep learning approach: application to a wind turbine airfoil

The inverse approach is computationally efficient in aerodynamic design as the desired target performance distribution is prespecified. However, it has some significant limitations that prevent it from achieving full efficiency. First, the iterative procedure should be repeated whenever the specified target distribution changes. Target distribution optimization can be performed to clarify the ambiguity in specifying this distribution, but several additional problems arise in this process such as loss of the representation capacity due to parameterization of the distribution, excessive constraints for a realistic distribution, inaccuracy of quantities of interest due to theoretical/empirical predictions, and the impossibility of explicitly imposing geometric constraints. To deal with these issues, a novel inverse design optimization framework with a two-step deep learning approach is proposed. A variational autoencoder and multi-layer perceptron are used to generate a realistic target distribution and predict the quantities of interest and shape parameters from the generated distribution, respectively. Then, target distribution optimization is performed as the inverse design optimization. The proposed framework applies active learning and transfer learning techniques to improve accuracy and efficiency. Finally, the framework is validated through aerodynamic shape optimizations of the wind turbine airfoil. Their results show that this framework is accurate, efficient, and flexible to be applied to other inverse design engineering applications.

cs.LG↗