GIST Scholar
Not a member yet
30271 research outputs found
Sort by
Integrated MEA for polymer electrolyte membrane fuel cells enabled by freeze–casting and direct membrane deposition
No full textPolymer electrolyte membrane fuel cells require electrode architectures that promote efficient gas and proton transport, while minimizing platinum usage. However, conventional catalyst-coated membranes, which employ thin and planar catalyst layers, form a two-dimensional triple-phase boundary (TPB), limiting further improvements in the power output and mass transport. To overcome these limitations, herein, an integrated membrane electrode assembly (MEA) was fabricated by combining freeze-casting and direct membrane deposition (FC + DMD). The freeze-cast catalyst layer was 30.9 μm thick, with high porosity and an open-pore structure, facilitating oxygen diffusion. Simultaneously, DMD formed a membrane that infiltrated the porous structure, establishing an interdigitated interface that enhanced proton conduction. Structural analysis confirmed 49 % porosity and 0.273 mL/g of pore volume, which was 2.6 times higher than GDE + DMD MEA. Simultaneously, the integrated FC + DMD MEA exhibited up to 50 % reductions in proton diffusion resistance (5.6 mΩ·cm2) and charge transfer resistance (526 mΩ·cm2) compared with Nafion laminated (GDE + N211) MEA. Finally, the electrochemical performance of the integrated FC + DMD MEA revealed a peak power density of 1.62 W/cm2 in oxygen. This is higher than that of the GDE + DMD (1.424 W/cm2) and at least threefold higher than FC + N211 (< 0.5 W/cm2) MEAs, evidencing the synergistic formation of 3D TPB and reduced ionic/ohmic losses. © 2025 Elsevier B.V.FALSEsciescopu
A Review of Electric Vehicle Charging Demand Forecasting Models by Prediction Horizon: A Multi-Criteria Decision Analysis Approach
No full textThe rapid increase in the number of electric vehicles (EVs) has raised significant concerns regarding the stability of power grid operations and the effective management of charging infrastructure. In this review paper, we categorize EV charging demand forecasting into three perspectives: short-term, medium-term, and long-term. Subsequently, we compare and analyze the input characteristics essential for enhancing forecasting accuracy. Additionally, we comprehensively examine the features, advantages, and disadvantages of various methodologies employed in recent studies. Specifically, a systematic review was conducted encompassing contemporary research trends, including time-series models, deep learning (DL) approaches, growth-curve models, and scenario-based simulation models. Furthermore, we propose a model selection framework that comprehensively considers forecasting accuracy alongside realistic operational scenarios, such as the data-collection period and the optimal scale of EV charging stations. This framework employs the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), a well-known multi-criteria decision making (MCDM) method technique, to evaluate the relative merits of different forecasting models. Ultimately, this study aims to provide researchers and practitioners engaged in EV charging demand forecasting with clear analytical criteria and practical guidelines for model selection.FALSEsciekc
Flood inundation mapping with CYGNSS over CONUS: a two-step machine-learning-based framework
No full textAccurate and timely flood inundation mapping is vital for early warning, disaster response, and mitigation planning. Global Navigation Satellite System-Reflectometry (GNSS-R) at L-band shows promise in detecting inundation extent, especially over land and lightly vegetated areas. However, the complex interaction between land surface features and the bistatic configuration of GNSS-R presents challenges for reliable flood mapping. In this work, a machine learning (ML) framework was developed to retrieve fractional inundation as the area proportionally covered by water using bistatic reflectance observations acquired from Cyclone GNSS (CYGNSS) and ancillary variables to characterize land surface conditions. Active C-band synthetic aperture radar (SAR) based high-resolution flood maps derived from Sentinel-1 were used as the reference for ML model training. The Random Forest (RF) model was used to retrieve surface water fraction in two steps through inundated pixel classification and water fraction regression. The sequential two-step (STS) structure was compared with the parallel two-step (PTS) model. Results show that the STS model outperforms both the PTS and the single regressor in deriving daily CYGNSS inundation retrievals at a 3-km resolution across the contiguous United States (CONUS). Cross-validation using a leave-one-year-out approach yields a correlation coefficient of 0.762 [-1 and root-mean-square-error of 0.039 [-1 between the CYGNSS inundation retrievals and the reference SAR-based water fractions. Consistent spatial variations are found between CYGNSS and Sentinel-1 inundated regions, suggesting satisfactory performance of the proposed ML model. In addition, the CYGNSS inundation are compared against several other inundation products, including the official CYGNSS water mask product, one semi-empirical method-based CYGNSS product, and a microwave remote sensing inundation product. Our CYGNSS product shows comparable performance in characterizing flood inundation at a 3-km resolution and daily temporal frequency.FALSEsciescopu
STAT3 SH2 Domain Aspartic Acid 661 Mutations Activate Immune Gene Programs
No full textThe conserved aspartic acid residue D661 within the STAT3 SH2 domain is a recurrent mutational hotspot in hematologic malignancies, including T-cell large granular lymphocytic leukaemia, myelodysplastic syndromes and acute lymphoblastic leukaemia. To define the functional consequences of distinct STAT3(D661) variants, we integrated computational, structural and in vitro and in vivo genetic approaches. AlphaMissense and PolyPhen-2 classified all four STAT3(D661) variants (D661Y, D661V, D661N and D661H) as pathogenic. ClinVar classified D661Y and D661V as variants of uncertain significance. AlphaFold 3-based modelling predicted that D661Y and D661V strongly promoted SH2-TAD-mediated dimerization, while D661N and D661H exerted weaker structural effects. Functional in vitro assays in Stat3-deficient T cells demonstrated a gain-of-function (GOF) hierarchy of the STAT3 variants (D661Y approximate to V > H > N) resulting in activation of canonical STAT3 target genes and immune transcriptional programs. In vivo, only STAT3(D661H) mice were viable, displaying reduced CD4(+) T cells, expansion of memory CD8(+) T cells and enhanced immune gene expression. Collectively, our findings define a gradient of STAT3 D661 GOF variants, consistent with in vitro and in vivo experiments. D661Y and D661V mutants exhibited stronger transcriptional activity in T cells with impaired viability of mice carrying these variants.TRUEsciescopu
DG-DETR: Toward domain generalized detection transformer
No full textEnd-to-end Transformer-based detectors (DETRs) have demonstrated strong detection performance. However, domain generalization (DG) research has primarily focused on convolutional neural network (CNN)-based detectors, while paying little attention to enhancing the robustness of DETRs. In this letter, we introduce a Domain Generalized DEtection TRansformer (DG-DETR), a simple, effective, and plug-and-play method that improves out-of-distribution (OOD) robustness for DETRs. Specifically, we propose a novel domain-agnostic query selection strategy that removes domain-induced biases from object queries via orthogonal projection onto the instance-specific style space. Additionally, we leverage a wavelet decomposition to disentangle features into domain-invariant and domain-specific components, enabling synthesis of diverse latent styles while preserving the semantic features of objects. Experimental results validate the effectiveness of DG-DETR. Our code is available at https://github.com/smin-hwang/DG-DETR . © © 2025. Published by Elsevier B.V.FALSEsciescopu
Multiscale topology optimization with explicit de-homogenization for graded pin-fin heat sink design
No full textThis paper is focused on a homogenization-based multiscale topology optimization (TO) framework with explicit de-homogenization for the design of graded pin-fin heat sinks subject to conjugate heat transfer. The proposed approach enables the design of diverse pin-fin geometries, and facilitates practical implementation through direct computer-aided design (CAD) model generation using explicit geometry feature representation. The proposed framework consists of three steps. Surrogate models for effective thermal conductivity and fluid permeability are first constructed using numerical homogenization and artificial neural networks. These models are then incorporated with cell rotations and macroscopic design variables to build the material property model, which is subsequently used to define and solve the TO problem. Finally, explicit de-homogenization is employed to restore optimized pin-fin microstructures using CAD feature representation, allowing direct generation of manufacturable CAD models. The effectiveness of the proposed framework is validated through a quantitative comparison of three benchmark heat sink designs using full three-dimensional (3D) numerical simulations. This comparison confirms that a multiscale heat sink design outperforms conventional (uniform fin, size-optimized fin, and macro-topology optimized fin) approaches by achieving a balanced configuration through effective flow redistribution. In addition, the practical applicability of the framework is established through additional examples including a Pareto front investigation for trade-off analysis and the direct generation of a 3D CAD model for additive manufacturing (AM). © 2026 Elsevier Ltd.FALSEsciescopu
Supercapacitors beyond energy storage: Multi-functional devices for sensing, actuation, and smart systems
No full textSupercapacitors (SCs) are evolving from passive high-power energy storage units into active, multifunctional elements that simultaneously store energy, sense, actuate, harvest, and communicate. This review critically examines how electrodes, electrolytes, separators, and current collectors can be engineered to couple charge storage with secondary functions such as mechanical, chemical, and thermal sensing, electrochemical actuation, electrochromism, self-healing, and self-charging. Rather than cataloguing demonstrations, we compare material families (carbonaceous materials, conducting polymers, transition metal oxides, MXenes, MOFs) and device architectures (flexible, stretchable, micro-, fiber/yarn and structural SCs) using common figures of merit: energy/ power density, sensitivity, response time, durability, safety, and integration complexity. Particular attention is given to trade-offs between capacitance and transduction sensitivity, energy density and mechanical robustness, and multifunctionality and long-term stability under coupled electro-chemo-mechanical loading. We highlight cross-cutting design strategies such as hierarchical porosity, interfacial/spacing engineering, healable solid and gel electrolytes, and 3D or textile-integrated formats, and assess their practicality for wearable systems, soft robotics, e-skin, smart windows, and IoT nodes. Finally, we identify key gaps, including limited energy density, inadequate standards for benchmarking multifunctional performance, and immature system-level integration, and outline research directions towards manufacturable, safe, and truly smart SC-based power–sensing–actuation platforms.FALSEsciescopu
Uncertainty quantification and optimization of precipitating hydrometeor parameters for winter precipitation in a cloud microphysics scheme
No full textThe precipitating hydrometeor parameters used in cloud microphysics schemes carry inherent uncertainties. The quantification of these uncertainties, together with parameter optimization, can significantly improve precipitation forecasts. This study investigates the effects of 13 parameters in the Weather Research and Forecasting (WRF) Double-Moment 6-class (WDM6) microphysics scheme, which define the hydrometeor characteristics such as fall velocity–diameter and mass–diameter relationships, as well as the shape parameter of the drop size distribution for precipitating particles such as rain, snow, and graupel on simulated winter precipitation. A comparison between the model's pre-defined parameters and observations from the International Collaborative Experiments for the PyeongChang 2018 Olympic and Paralympic winter games (ICE-POP 2018) field campaign reveals that the fall velocity–diameter relationship for rain, the mass–diameter relationships for snow and graupel, and the shape parameters for all precipitating particles in the WDM6 scheme deviate from the median values observed by the two-dimensional video disdrometer (2DVD). To quantify parameter sensitivities, a perturbed parameter ensemble (PPE) of 256 simulations was conducted within parameter ranges constrained by 2DVD observations for three winter precipitation cases. Bayesian optimization was then applied to identify parameter sets that minimized the root mean square error (RMSE) for each case, achieving reductions of up to 30.2 %. These results demonstrate that ensemble-based uncertainty quantification and parameter optimization can help identify key parameters and provide a pathway to improving precipitation simulation performance. In addition, measurement sites can be strategically selected based on regions that show high sensitivity to variations in hydrometeor characteristic parameters. © 2025 Elsevier B.V., All rights reserved.TRUEsciescopu
