arXiv subjects

Seunghwa Ryu

Publications and source records attributed to Seunghwa Ryu.

At least 19 recordsLinked to original sources

arXiv2026

Predicting Space Groups of Double Perovskites by LLM with Dynamic Few-Shot Learning

Double perovskites (DPs) offer broad compositional tunability, but predicting the space groups (SGs) of stable structures remains difficult because available datasets are often strongly imbalanced toward dominant SG classes. We refer to dominant SG classes as major SGs and underrepresented classes as minor SGs. We introduce Dynamic and Diversity-enhanced Few-shot Retrieval and Rule-Guided Inference for Space-Group Prediction (DyRIS), an LLM-agent-based framework that predicts ranked SG candidates from a given DP composition. DyRIS uses diversity-enhanced dynamic few-shot prompting to retrieve relevant in-context examples while limiting the dominance of frequently represented SGs. It further incorporates rule-guided inference based on B/B' cation ordering, quantitative indicators, and major-SG bias control to refine and rank the final Top-3 SG candidates. We evaluate DyRIS on 3,528 thermodynamically filtered DP entries and compare it with composition-based and descriptor-based baselines. At a training-data ratio of 0.5, DyRIS achieves competitive overall accuracy while obtaining the best Overall Top-1 macro-F1 score and the best performance across all Minor-SG metrics. DyRIS improves Minor-SG Top-1 accuracy by 3.26 percentage points relative to CrabNet and achieves higher Minor-SG Top-3 accuracy than the strongest PyCaret-based baseline. Ablation studies show that diversity-enhanced retrieval, quantitative indicators, major-SG bias control, and B/B' ordering information each contribute to prediction performance. Additional experiments show that the final rule-guided inference step is not easily replaced by conventional classifier- or ranker-based models. These findings demonstrate the potential of combining retrieval-based LLM reasoning with crystallographic domain knowledge for SG prediction in imbalanced materials datasets.

cs.AI↗

arXiv2026

A Multi-Agent Framework for Automated Coarse-Grained Molecular Dynamics of Polymers

Coarse-grained (CG) molecular dynamics extends polymer simulation beyond the scales accessible to all-atom (AA) methods, but bottom-up CG modeling is laborious. The CG resolution is a design choice, so a transferable parameter set is generally not available and the potentials are derived anew for each polymer mapping. Here we present CGMas, a multi-agent framework that automates topology construction, equilibration, mapping, potential derivation, and validation from a natural-language specification of the polymer and target resolution. A large-language-model (LLM) reasoning agent infers the AA topology from polymer name, while layered self-correction resolves physical errors common to unsaturated, heteroatom-containing, and polar polymers. Downstream agents equilibrate the system, map it onto CG representation, derive potentials through Boltzmann inversion, and benchmark the model against its atomistic reference. CGMas completed all 27 homopolymer and copolymer tasks, matched the AA density to within 5% in 22, and reduced simulation from 38-88 min to 1 min, establishing agentic LLMs as a route to automated polymer coarse-graining.

cs.AI↗

arXiv2026

Uncertainty-Aware Structure-Property Mapping of Spinodoid Metamaterials via Heteroscedastic Gaussian Process Regression

Spinodoid metamaterials offer a broad, tunable design space for anisotropic mechanical properties, yet their structure-property relationships are commonly treated as representative mappings from cone-angle descriptors to single effective stiffness values. This deterministic view overlooks the stochastic nature of Gaussian random field (GRF)-based topology generation, where identical cone-angle descriptors can produce different morphology realizations and property scatter. Here, we present an uncertainty-aware structure-property mapping framework that reinterprets cone-angle descriptors as stochastic descriptors associated with input-dependent property distributions. Using heteroscedastic Gaussian process regression (GPR), the framework infers input-dependent predictive uncertainty from sparse one-realization-per-point data without requiring empirical variance labels at every design point. The results show that stiffness scatter differs across tensor components according to each component's mechanically active directions, and that parameter sets yielding identical mean stiffness can carry different aleatoric uncertainty. Applying this uncertainty to reliability-based design optimization (RBDO), we show that a deterministic optimum is highly susceptible to constraint violation once morphology-induced variability is considered, and that a homoscedastic RBDO formulation fails to meet the prescribed reliability target - only the heteroscedastic formulation satisfies the reliability target under the heteroscedastic uncertainty evaluation. This establishes uncertainty-aware surrogate modeling as essential for reliability-aware inverse design of spinodoid metamaterials; extending the framework to nonlinear responses remains for future work.

physics.comp-ph↗

arXiv2026

Physics-Informed Discovery of Yield Functions in Plasticity via Convex Neural Representations

Identifying anisotropic yield functions remains challenging since yielding is not directly observed in full-field mechanical measurements, directional calibration can require many loading directions, and selecting an appropriate analytical form is nontrivial. This study proposes a physics-informed framework for discovering yield functions from full-field displacement data and reaction force data, without stress observations, plastic strain measurements, direct yield surface data, or a prescribed parametric yield function. The framework identifies the yield function as a mechanically constrained constitutive component inside elastoplastic stress integration, rather than through direct stress-space supervision. The yield function is represented by a convex neural network that enforces convexity and positive homogeneity of degree one while imposing the assumed tension-compression symmetry, and this neural yield function is trained with a differentiable stress update and a physics-informed force equilibrium loss across multiple loading cases. The proposed framework is validated using finite element (FE) benchmark studies with von Mises, Hill 1948, and Yld2000-2d yield functions, assessing yield contour agreement, displacement-noise sensitivity, identifiability through plastically active stress states, epistemic uncertainty, and polynomial-surrogate deployment. This study provides a mechanics-constrained pathway for discovering anisotropic yield functions from displacement and force data while keeping the identified component within the structure of elastoplastic stress integration.

cs.LG↗

arXiv2026

A Hybrid GNN-FEM Framework for Phase-Field Fracture Simulation. Physics-Preserving Hybridization for Generalizable Surrogate Modeling

Scientific machine learning (SciML) has emerged as a promising approach for accelerating simulations of complex physical systems, yet achieving physically consistent and generalizable predictions for nonlinear, history-dependent problems remains a central challenge. In this study, we propose a hybrid GNN--FEM framework for efficient and generalizable phase-field fracture modeling. While phase-field approaches provide a robust variational framework for simulating complex crack evolution, their high computational cost limits practical applications because they require solving coupled, nonlinear, and history-dependent systems within an incremental finite element procedure. To address this challenge, a graph neural network surrogate is integrated into the conventional staggered scheme, replacing the phase-field update at each load increment while retaining the FEM-based displacement solver to enforce mechanical equilibrium and boundary conditions. By preserving the incremental solution structure, the framework remains consistent with history-dependent fracture evolution without requiring the surrogate to approximate the full solution trajectory. This selective surrogate strategy emphasizes the identification of a physically meaningful and incrementally structured learning target, rather than relying on brute-force data generation to learn the full fracture process. The proposed framework achieves strong generalization across varying geometries, loading conditions, material properties, and discretizations through dimensionless feature design, a graph-based formulation on mesh-based domains, and a physics-informed loss derived from the governing phase-field equation. Numerical experiments demonstrate that the hybrid approach reduces computational cost while maintaining accuracy compared with conventional FEM, and exhibits robust predictive performance across diverse problem settings.

cs.LG↗

arXiv2026

Critical evaluation of PINN for FWD inverse analysis and differentiable FEM as an alternative

Automatic-differentiation-based inverse analysis methods, including physics-informed neural networks (PINNs) and differentiable programming, have recently shown great promise due to their ability to compute accurate gradients and convergence efficiency. However, their applicability to falling weight deflectometer (FWD) backcalculation remains unexplored. This study critically evaluates PINN-based inverse analysis for a multilayer pavement system and investigates differentiable finite element method (DiffFEM) as an alternative based on a synthetic benchmark. The standard PINN does not recover layer moduli because of the sharp domain discontinuities inherent to layered pavement systems. Although we use an extended PINN with domain decomposition (XPINN), which shows better performance on discontinuous domains, its performance remains highly sensitive to loss weighting and network architecture, and degrades under measurement noise. By contrast, DiffFEM consistently achieves more accurate, stable, and computationally efficient inversion results. These results indicate that DiffFEM, which enforces the governing physics as a hard constraint, yields better accuracy, robustness, and computational efficiency than PINN-based approaches, in which the governing physics is imposed as a soft constraint through the loss function. More broadly, the findings suggest that the choice between PINN- and DiffFEM-based inverse analysis needs careful consideration, with DiffFEM offering practical advantages when an efficient and robust differentiable forward solver is available.

cs.CE↗

arXiv2026

Morphology-, Noise-, and Resolution-Robust Ultrasound Elasticity Imaging with Fourier Neural Operators

Ultrasound-based elasticity imaging is a non-invasive technique for estimating tissue stiffness fields from displacement fields obtained by comparing ultrasound signals before and after compression. While recent deep learning approaches have enabled faster and more accurate elasticity estimation compared to traditional methods, several challenges remain for clinical translation. In this study, we employ finite element simulations of free-hand palpation to investigate the applicability of the Fourier neural operator (FNO). Four practical scenarios were investigated: (1) prediction across diverse lesion morphologies, (2) generalization to cases with lesion counts differing from those in the training data, (3) robustness to noise in measured displacement fields, and (4) resilience to variations in ultrasound device resolution. Across these tasks, FNO consistently outperformed baseline models such as U-Net and DeepONet in predictive accuracy and generalization, while maintaining robustness under noise and resolution changes. Validated through in silico simulations, these findings demonstrate the potential of FNO as a framework that could facilitate translation of elasticity imaging toward clinical practice.

physics.med-ph↗

arXiv2025

Enhanced Conditional Generation of Double Perovskite by Knowledge-Guided Language Model Feedback

Double perovskites (DPs) are promising candidates for sustainable energy technologies due to their compositional tunability and compatibility with low-energy fabrication, yet their vast design space poses a major challenge for conditional materials discovery. This work introduces a multi-agent, text gradient-driven framework that performs DP composition generation under natural-language conditions by integrating three complementary feedback sources: LLM-based self-evaluation, DP-specific domain knowledge-informed feedback, and ML surrogate-based feedback. Analogous to how knowledge-informed machine learning improves the reliability of conventional data-driven models, our framework incorporates domain-informed text gradients to guide the generative process toward physically meaningful regions of the DP composition space. Systematic comparison of three incremental configurations, (i) pure LLM generation, (ii) LLM generation with LLM reasoning-based feedback, and (iii) LLM generation with domain knowledge-guided feedback, shows that iterative guidance from knowledge-informed gradients improves stability-condition satisfaction without additional training data, achieving over 98% compositional validity and up to 54% stable or metastable candidates, surpassing both the LLM-only baseline (43%) and prior GAN-based results (27%). Analyses of ML-based gradients further reveal that they enhance performance in in-distribution (ID) regions but become unreliable in out-of-distribution (OOD) regimes. Overall, this work provides the first systematic analysis of multi-agent, knowledge-guided text gradients for DP discovery and establishes a generalizable blueprint for MAS-driven generative materials design aimed at advancing sustainable technologies.

cs.AI↗

arXiv2025

Real-Time Structural Health Monitoring with Bayesian Neural Networks: Distinguishing Aleatoric and Epistemic Uncertainty for Digital Twin Frameworks

Reliable real-time analysis of sensor data is essential for structural health monitoring (SHM) of high-value assets, yet a major challenge is to obtain spatially resolved full-field aleatoric and epistemic uncertainties for trustworthy decision-making. We present an integrated SHM framework that combines principal component analysis (PCA), a Bayesian neural network (BNN), and Hamiltonian Monte Carlo (HMC) inference, mapping sparse strain gauge measurements onto leading PCA modes to reconstruct full-field strain distributions with uncertainty quantification. The framework was validated through cyclic four-point bending tests on carbon fiber reinforced polymer (CFRP) specimens with varying crack lengths, achieving accurate strain field reconstruction (R squared value > 0.9) while simultaneously producing real-time uncertainty fields. A key contribution is that the BNN yields robust full-field strain reconstructions from noisy experimental data with crack-induced strain singularities, while also providing explicit representations of two complementary uncertainty fields. Considered jointly in full-field form, the aleatoric and epistemic uncertainty fields make it possible to diagnose at a local level, whether low-confidence regions are driven by data-inherent issues or by model-related limitations, thereby supporting reliable decision-making. Collectively, the results demonstrate that the proposed framework advances SHM toward trustworthy digital twin deployment and risk-aware structural diagnostics.

cs.LG↗

arXiv2025

IM-Chat: A Multi-agent LLM Framework Integrating Tool-Calling and Diffusion Modeling for Knowledge Transfer in Injection Molding Industry

The injection molding industry faces critical challenges in preserving and transferring field knowledge, particularly as experienced workers retire and multilingual barriers hinder effective communication. This study introduces IM-Chat, a multi-agent framework based on large language models (LLMs), designed to facilitate knowledge transfer in injection molding. IM-Chat integrates both limited documented knowledge (e.g., troubleshooting tables, manuals) and extensive field data modeled through a data-driven process condition generator that infers optimal manufacturing settings from environmental inputs such as temperature and humidity, enabling robust and context-aware task resolution. By adopting a retrieval-augmented generation (RAG) strategy and tool-calling agents within a modular architecture, IM-Chat ensures adaptability without the need for fine-tuning. Performance was assessed across 100 single-tool and 60 hybrid tasks for GPT-4o, GPT-4o-mini, and GPT-3.5-turbo by domain experts using a 10-point rubric focused on relevance and correctness, and was further supplemented by automated evaluation using GPT-4o guided by a domain-adapted instruction prompt. The evaluation results indicate that more capable models tend to achieve higher accuracy, particularly in complex, tool-integrated scenarios. In addition, compared with the fine-tuned single-agent LLM, IM-Chat demonstrated superior accuracy, particularly in quantitative reasoning, and greater scalability in handling multiple information sources. Overall, these findings demonstrate the viability of multi-agent LLM systems for industrial knowledge workflows and establish IM-Chat as a scalable and generalizable approach to AI-assisted decision support in manufacturing.

cs.AI↗

arXiv2025

Thermal Conductivity Estimation of Thermoelectric Materials with Uncertainty Quantification Using Bayesian Physics-Informed Neural Networks

Characterizing the temperature-dependent thermal conductivity is challenging because the property varies strongly with temperature and reliable heat flow measurement, not just temperature sensing, is difficult under experimental conditions. Here, we present a physics informed deep learning framework that infers conductivity solely from sparse electric potential measurements. We first develop a deterministic physics-informed neural network (PINN) that embeds coupled thermoelectric transport equations as soft constraints, enabling simultaneous recovery of spatial temperature, voltage, and conductivity profiles without temperature data. The deterministic PINN achieves accurate inference under noise-free conditions, yet its predictions degrade when measurement noise is introduced. To address this, we extend the framework to a Bayesian PINN, which models network parameters probabilistically and employs Hamiltonian Monte Carlo (HMC) sampling for posterior inference. This extension produces robust thermal conductivity estimates and, importantly, provides credible intervals that quantify uncertainty from sparse and noisy data. Numerical experiments confirm that the Bayesian PINN not only preserves predictive accuracy under noise but also reveals inference bias and enables uncertainty aware interpretation of material properties. Together, the deterministic and Bayesian formulations establish a scalable and generalizable alternative to conventional methods for determining temperature-dependent properties, offering physics- consistent and risk-aware property inference for thermoelectric systems and other functional materials where direct temperature sensing is impractical

physics.comp-ph↗

arXiv2025

Physics-Informed Neural Networks for Programmable Origami Metamaterials with Controlled Deployment

Origami-inspired structures provide unprecedented opportunities for creating lightweight, deployable systems with programmable mechanical responses. However, their design remains challenging due to complex nonlinear mechanics, multistability, and the need for precise control of deployment forces. Here, we present a physics-informed neural network (PINN) framework for both forward prediction and inverse design of conical Kresling origami (CKO) without requiring pre-collected training data. By embedding mechanical equilibrium equations directly into the learning process, the model predicts complete energy landscapes with high accuracy while minimizing non-physical artifacts. The inverse design routine specifies both target stable-state heights and separating energy barriers, enabling freeform programming of the entire energy curve. This capability is extended to hierarchical CKO assemblies, where sequential layer-by-layer deployment is achieved through programmed barrier magnitudes. Finite element simulations and experiments on physical prototypes validate the designed deployment sequences and barrier ratios, confirming the robustness of the approach. This work establishes a versatile, data-free route for programming complex mechanical energy landscapes in origami-inspired metamaterials, offering broad potential for deployable aerospace systems, morphing structures, and soft robotic actuators.

cond-mat.soft↗

arXiv2025

Physics-informed neural network framework for solving forward and inverse flexoelectric problems

Flexoelectricity, the coupling between strain gradients and electric polarization, poses significant computational challenges due to its governing fourth-order partial differential equations that require C1-continuous solutions. To address these issues, we propose a physics-informed neural network (PINN) framework grounded in an energy-based formulation that treats both forward and inverse problems within a unified architecture. The forward problem is recast as a saddle-point optimization of the total potential energy, solved via the deep energy method (DEM), which circumvents the direct computation of high-order derivatives. For the inverse problem of identifying unknown flexoelectric coefficients from sparse measurements, we introduce an additional variational loss that enforces stationarity with respect to the electric potential, ensuring robust and stable parameter inference. The framework integrates finite element-based numerical quadrature for stable energy evaluation and employs hard constraints to rigorously enforce boundary conditions. Numerical results for both direct and converse flexoelectric effects show excellent agreement with mixed-FEM solutions, and the inverse model accurately recovers material parameters from limited data. This study establishes a unified, mesh-compatible, and scalable PINN approach for high-order electromechanical problems, offering a promising alternative to traditional simulation techniques.

physics.comp-ph↗

arXiv2025

Physics-Informed Neural Operators for Generalizable and Label-Free Inference of Temperature-Dependent Thermoelectric Properties

Accurate characterization of temperature-dependent thermoelectric properties (TEPs), such as thermal conductivity and the Seebeck coefficient, is essential for reliable modeling and efficient design of thermoelectric devices. However, their nonlinear temperature dependence and coupled transport behavior make both forward simulation and inverse identification difficult, particularly under sparse measurement conditions. In this study, we develop a physics-informed machine learning approach that employs physics-informed neural networks (PINN) for solving forward and inverse problems in thermoelectric systems, and neural operators (PINO) to enable generalization across diverse material systems. The PINN enables field reconstruction and material property inference by embedding governing transport equations into the loss function, while the PINO generalizes this inference capability across diverse materials without retraining. Trained on simulated data for 20 p-type materials and evaluated on 60 unseen materials, the PINO model demonstrates accurate and label-free inference of TEPs using only sparse field data. The proposed framework offers a scalable, generalizable, and data-efficient approach for thermoelectric property identification, paving the way for high-throughput screening and inverse design of advanced thermoelectric materials.

cond-mat.mtrl-sci↗

arXiv2025

Data-driven Design of Isotropic and High-Stiffness TPMS-based Amorphousness-Induced Architected Material (TAAM)

For their excellent stiffness-to-weight characteristics, triply periodic minimal surfaces (TPMS) are widely adopted in architected materials. However, their geometric regularity often leads to elastic anisotropy, limiting their effectiveness under complex loading. To address this, we propose TPMS-based amorphousness-induced architected materials (TAAMs), which incorporate controllable geometric disorder as a tunable design variable. This concept of designable amorphousness broadens the geometric design space, enabling the simultaneous optimization of stiffness and isotropy. A data-driven framework integrating computational homogenization with multi-objective Bayesian optimization is employed to discover high-performance TAAMs. Selected designs were fabricated using fused deposition modeling and validated through uniaxial compression tests, showing strong agreement with numerical predictions. Compared to conventional TPMS, TAAMs exhibit significantly improved elastic isotropy while maintaining high stiffness across various relative densities. This approach offers a robust and scalable pathway for developing architected materials tailored to applications requiring isotropic performance, such as biomedical implants, protective systems, and aerospace components.

physics.comp-ph↗

arXiv2025

DRL-Based Injection Molding Process Parameter Optimization for Adaptive and Profitable Production

Plastic injection molding remains essential to modern manufacturing. However, optimizing process parameters to balance product quality and profitability under dynamic environmental and economic conditions remains a persistent challenge. This study presents a novel deep reinforcement learning (DRL)-based framework for real-time process optimization in injection molding, integrating product quality and profitability into the control objective. A profit function was developed to reflect real-world manufacturing costs, incorporating resin, mold wear, and electricity prices, including time-of-use variations. Surrogate models were constructed to predict product quality and cycle time, enabling efficient offline training of DRL agents using soft actor-critic (SAC) and proximal policy optimization (PPO) algorithms. Experimental results demonstrate that the proposed DRL framework can dynamically adapt to seasonal and operational variations, consistently maintaining product quality while maximizing profit. Compared to traditional optimization methods such as genetic algorithms, the DRL models achieved comparable economic performance with up to 135x faster inference speeds, making them well-suited for real-time applications. The framework's scalability and adaptability highlight its potential as a foundation for intelligent, data-driven decision-making in modern manufacturing environments.

cs.AI↗

arXiv2025

Physics-Informed Neural Network-Based Discovery of Hyperelastic Constitutive Models from Extremely Scarce Data

The discovery of constitutive models for hyperelastic materials is essential yet challenging due to their nonlinear behavior and the limited availability of experimental data. Traditional methods typically require extensive stress-strain or full-field measurements, which are often difficult to obtain in practical settings. To overcome these challenges, we propose a physics-informed neural network (PINN)-based framework that enables the discovery of constitutive models using only sparse measurement data - such as displacement and reaction force - that can be acquired from a single material test. By integrating PINNs with finite element discretization, the framework reconstructs full-field displacement and identifies the underlying strain energy density from predefined candidates, while ensuring consistency with physical laws. A two-stage training process is employed: the Adam optimizer jointly updates neural network parameters and model coefficients to obtain an initial solution, followed by L-BFGS refinement and sparse regression with l_p regularization to extract a parsimonious constitutive model. Validation on benchmark hyperelastic models demonstrates that the proposed method can accurately recover constitutive laws and displacement fields, even when the input data are limited and noisy. These findings highlight the applicability of the proposed framework to experimental scenarios where measurement data are both scarce and noisy.

physics.comp-ph↗

arXiv2024

Wiedemann-Franz Law and Thermoelectric Inequalities: Effective ZT and Single-leg Efficiency Overestimation

We derive a thermoelectric inequality in thermoelectric conversion between the material figure of merit (ZT) and the module effective ZT using the Constant Seebeck-coefficient Approximation combining with the Wiedemann-Franz law. In a P-N leg-pair module, the effective ZT lies between the individual ZT values of the P- and N-legs. In a single-leg module, however, the effective ZT is less than approximately one-third of the leg's ZT. This reduction results from the need for an external wire to complete the circuit, introducing additional thermal and electrical losses. Multi-dimensional numerical analysis shows that, although structural optimization can mitigate these losses, the system efficiency remains limited to below half of the ideal single-leg material efficiency. Our findings explain the single-leg efficiency overestimation and highlight the importance of optimizing the P-N leg-pair module structure. They also underscore the need for thermoelectric leg-compatibility, particularly with respect to Seebeck coefficients.

physics.app-ph↗