1663 papers
This paper proposes PI-TransUNet, a physics-informed deep learning framework that integrates a TransUNet architecture with a differentiable forward surrogate model and a physics-consistency objective to achieve accurate, robust, and physically reliable steel plate defect detection in electromagnetic tomography.
This study introduces the Conditional Diffusion Downscaling Model (CDDM), a novel generative approach that leverages a modified Brownian bridge process to effectively downscale coarse GCM precipitation forecasts to fine resolutions, consistently outperforming traditional and state-of-the-art methods in probabilistic and deterministic metrics across diverse climate conditions.
This study utilizes CPFD simulations to evaluate a Tesla valve-type reactor for chemical looping processes, revealing that while forward and reverse modes generally exhibit fluidization dead zones that can be mitigated by optimizing gas velocity, bed-to-particle ratios, and purge configurations, a reverse operating mode at 20 m/s uniquely eliminates stagnant regions by directing bed material along the longest flow path.
This study proposes a systematic framework for generating physically consistent training datasets through non-stationarity screening and Maximum Likelihood Estimation-based label verification, demonstrating that such data quality improvements significantly enhance the accuracy and generalization of Convolutional Neural Networks in estimating the scale of fluctuation for spatially variable ground.
This paper presents FusionTouch, a glove-free spherical haptic interface that utilizes a cross-domain transparent architecture and bio-inspired textile fabrication to synergistically integrate multimodal sensing and bidirectional haptic feedback on a single flexible surface, enabling high-fidelity hand pose estimation, deformation reconstruction, and force estimation for diverse human-robot interaction applications.
This paper introduces CorroSense-Net, an interpretable hybrid CNN-Transformer framework featuring a novel Cross-Attention Fusion module and uncertainty quantification that achieves state-of-the-art accuracy in distinguishing between pristine, erosion, and corrosion states in industrial pipelines, thereby addressing the limitations of existing single-model approaches in safety-critical inspection.
This study proposes a novel battery-free autonomous mobility architecture that leverages spatiotemporal resonance, reversible computing, and evanescent wave coupling to achieve constant-latency control, high-efficiency wireless power transfer, and zero hacking risk by eliminating chemical batteries and traditional memory constraints.
The paper introduces LIBRA, an open, interactive web-based framework that integrates techno-economic and environmental assessments to evaluate the viability of diverse lithium-ion battery recycling pathways under evolving chemistries, demonstrating that chemistry-specific mechanical-hydrometallurgical routes remain profitable even with high lithium iron phosphate penetration.
This paper reviews the convergence of additive manufacturing, architected metamaterials, and AI-assisted design for multifunctional composites, while demonstrating through a case study on 420 fibre-reinforced polymers that ensemble machine learning models can effectively predict mechanical properties from processing and constituent descriptors to accelerate inverse design and 4D printing.
This paper presents a robust, reproducible feasible-domain modeling workflow for spaceborne heterogeneous high-performance computing that integrates multi-constraint screening (including timing, memory, power, thermal, and reliability) to identify optimal task mappings, demonstrating through synthetic case studies that robust selection significantly reduces makespan and eliminates deadline violations under perturbation with only a marginal energy trade-off compared to nominal energy-optimized approaches.