Search Results (117)

The proliferation of intelligent edge devices demands compact, low-power hardware capable of dynamically switching between sensing, logic, and learning tasks—a versatility that traditional multi-chip solutions fundamentally lack. Here, we demonstrate a reconfigurable spin–orbit torque (SOT) device based on an FeTb/Ru/Co synthetic antiferromagnetic (SAF) heterostructure. By modulating the input current amplitude, the device dynamically switches between two distinct operating modes: saturation and activation. In the saturation regime (>80 mA), deterministic magnetization reversal enables Boolean logic operations (AND, NOR). In the activation regime (<80 mA), gradual, non-volatile conductance modulation emulates synaptic plasticity. Benefiting from the strong antiferromagnetic coupling and near-zero net magnetization of the SAF structure, all operations are achieved without external magnetic fields. This single-device, dual-mode reconfigurable architecture establishes a new paradigm for high-density, low-power, multifunctional in-memory computing units, with promise for advancing adaptive edge computing chips. Full article

Conventional VAR-based volatility spillover measures rely on homoskedasticity and single-Gaussian assumptions, limiting their ability to capture structural breaks and heterogeneous shocks during crises. This study develops a flexible framework to analyze volatility transmission in U.S. commodity markets under multiple crisis regimes. We propose a Bayesian Structural Vector Autoregressive Mixture Normal (BSVAR-MIX) model that embeds finite normal mixtures within a mixture-based heteroskedastic structural VAR framework. The model combines generalized forecast error variance decomposition with posterior-probability weighting. Daily data for eight U.S. benchmark commodities across food, energy, and precious metals markets are examined over the 2008–2016 global financial crisis and the 2017–2025 multi-crisis period, including COVID-19 and the Russia–Ukraine conflict. The BSVAR-MIX framework provides a flexible descriptive setting for capturing multimodal shocks, heteroskedastic volatility states, and regime-dependent spillover patterns in commodity markets. Empirically, Gold and oil dominate systemic volatility transmission, soybeans amplify food–energy spillovers, while coal and wheat exhibit rising fragility under policy and geopolitical shocks. Assets commonly viewed as safe havens may contribute to systemic stress during extreme events. Overall, the framework offers a robust tool for structural shock identification and cross-commodity risk monitoring relevant to U.S. macroprudential policy. Full article

Lightweight day-ahead photovoltaic (PV) forecasting models encounter a significant technical challenge: under resource-constrained deployment conditions, it is difficult to simultaneously address weather-regime heterogeneity, maintain model interpretability, and preserve adaptability as operating conditions evolve. To address this issue, we propose a multimodal short-term photovoltaic (PV) forecasting method that integrates weather-mode partitioning using the Clustering Using Representatives (CURE) algorithm with a self-updating Batch-Lasso model. First, the meteorological-PV dataset is partitioned along two dimensions by combining seasonal grouping with CURE clustering within each season, producing representative weather modes and enhancing the fidelity of weather pattern classification. Second, to extract informative predictors from high-dimensional meteorological inputs while maintaining interpretability, we formulate per-mode Lasso regression and adopt the Fast Iterative Shrinkage-Thresholding Algorithm (FISTA) to efficiently solve for the sparse regression coefficients. Third, we introduce a batch-based self-update and correction mechanism with rollback verification, enabling the mode-specific models to be refreshed as new historical data become available while preventing performance degradation. Compared with representative machine learning baselines, the proposed method maintains competitive accuracy with substantially lower computational and storage overhead, enabling high-frequency and energy-efficient inference on resource-constrained terminals, thereby reducing operational burdens and computational energy costs and better meeting the deployment needs of sustainable energy systems under heterogeneous weather conditions. Full article

Electric vehicle(EV) diffusion exhibits nonlinear, path-dependent dynamics shaped by interacting economic, technological, and social constraints. This paper develops a unified hybrid systems framework that captures these complexities by integrating microfounded household choice, capacity-constrained firm behavior, local network spillovers, and multi-level policy intervention within a Filippov differential-inclusion structure. Households face heterogeneous preferences, liquidity limits, and network-mediated moral and informational influences; firms invest irreversibly under learning-by-doing and profitability thresholds; and national and local governments implement distinct financial and infrastructure policies subject to budget constraints. The resulting aggregate adoption dynamics feature endogenous switching, sliding modes at economic bottlenecks, network-amplified tipping, and hysteresis arising from irreversible investment. We establish conditions for the existence of Filippov solutions, derive network-dependent tipping thresholds, characterize sliding regimes at capacity and liquidity constraints, and show how network structure magnifies hysteresis and shapes the effectiveness of local versus national policy. Optimal-control analysis further demonstrates that national subsidies follow bang–bang patterns and that network-targeted local interventions minimize the fiscal cost of achieving regional tipping. Beyond theoretical characterization, the framework is structurally calibrated to match the order-of-magnitude effects reported in leading empirical and simulation-based studies, including network diffusion models, agent-based simulations, bass-type specifications, and fuel-price shock analyses. The hybrid formulation reproduces short-run percentage-point subsidy effects, long-run forecast dispersion under alternative network assumptions, and policy-induced equilibrium shifts observed in the applied literature while providing a unified geometric interpretation of these heterogeneous results through explicit basin boundaries and regime switching. The framework provides a complex systems perspective on sustainable mobility transitions and clarifies why identical national policies can generate asynchronous regional outcomes. These results offer theoretical foundations for designing coordinated, cost-effective, and network-aware EV transition strategies. Full article

With the growing penetration of photovoltaic (PV) generation, robust uncertainty characterization is essential for secure operation, economic dispatch, and flexibility planning. This review surveys PV scenario generation from three perspectives: (i) explicit probabilistic approaches (distribution fitting, Copula-based dependence modeling, autoregressive moving average (ARMA)-type time-series methods, and clustering/dimensionality reduction), (ii) deep generative models (GANs, VAEs, and diffusion models), and (iii) hybrid Statistical Relational AI (SRAI) frameworks. We discuss the strengths of explicit models in interpretability and tractability, and their limitations in representing high-dimensional nonlinear, multimodal, and multiscale spatiotemporal dependencies. We also examine the ability of deep generative methods to synthesize diverse scenarios across meteorological regimes and multiple sites, while noting persistent challenges in interpretability, physical consistency, and deployment. To bridge these gaps, we outline an SRAI-oriented integration pathway that embeds statistical structure, meteorology–power relations, spatiotemporal coupling, and operational constraints into generative architectures. Finally, we highlight directions for future research, including unified evaluation protocols, cross-regional data collaboration, controllable extreme-scenario generation, and computationally efficient generative designs. Full article

This study evaluates the effectiveness of laser cleaning techniques for the non-contact removal of unwanted deposits from the surface of contemporary urban mural paintings. Two sets of mock-up samples, painted with popular graffiti spray paints on lime-based plaster, and artificially contaminated, were subjected to various cleaning procedures using Nd:YAG lasers operated in Q-switched (QS), long Q-switched (LQS) or short free-running mode (SFR). A multi-analytical approach—including X-ray fluorescence spectroscopy (XRF), Fourier-transform infrared spectroscopy (FTIR), colorimetry, and hyperspectral imaging (HSI)—was used to identify pigments and binders, and to evaluate cleaning efficiency and pigment preservation. XRF and FTIR were useful in understanding the composition of the sprays, while colorimetric ΔE values quantified cleaning efficiency and potential damage, and hyperspectral reflectance and LSU (linear spectral unmixing) abundance maps provided spatial distribution insights into contaminant removal and pigment preservation. The results demonstrate that laser cleaning effectiveness and selectivity are strongly dependent on the operational regime and fluence. In particular, long Q-switched laser irradiation at moderate fluence levels achieved effective contaminant removal with minimal chromatic and chemical alteration of the original paint layers. These findings support the development of tailored, sustainable, and non-contact laser cleaning protocols for the conservation of contemporary urban murals and contribute to the establishment of objective, multi-parameter criteria for evaluating cleaning outcomes in street art conservation. Full article

Dynamic mechanical thermal analysis (DMTA) is widely used to assess the effects of process- and environment-induced changes in polymer matrix composites, with the glass transition temperature (T_g) often reported from the tan d peak at a single excitation frequency. However, such an approach neglects the inherently kinetic nature of the glass transition and may obscure thermally induced changes in relaxation response. Multi-frequency DMTA was employed to investigate the evolution of glass transition response of a wet-layup unidirectional carbon/epoxy composite subjected to thermal aging at temperatures ranging from 66 °C to 260 °C for periods up to 72 h, using unexposed (23 °C) results as an ambient baseline reference. Tests were conducted using a single cantilever mode in both longitudinal and transverse configurations over a range of excitation frequencies from 0.3 to 30 Hz. Results demonstrate that thermal exposure affects not only the absolute value of the glass transition temperature, but also its frequency sensitivity and directional dependence. A frequency sensitivity parameter and a directional amplification factor are introduced to quantify frequency–direction coupling. While post-cure-dominated aging regimes exhibit relatively stable coupling behavior, degradation-dominated conditions at elevated temperatures and longer periods of thermal exposure lead to pronounced increases in transverse frequency sensitivity, which reflects early evolution of matrix- and interphase-level deterioration. These findings highlight the value of multi-frequency DMTA with tests in both primary directions for the mechanistic assessment of effects of thermo-oxidative response in polymer matrix composites. Full article

Accurate environmental perception is fundamental to safe autonomous driving; however, most existing multimodal systems rely on fixed or heuristic sensor fusion strategies that cannot adapt to scene-dependent variations in sensor reliability. This paper proposes Cross-Modal Adaptive Attention (CMAA), a unified end-to-end Bird’s-Eye-View (BEV) perception framework that dynamically fuses camera, LiDAR, and RADAR information through learnable, context-aware modality gating. Unlike static fusion approaches, CMAA adaptively reweights sensor contributions based on global scene descriptors, enabling the robust integration of semantic, geometric, and motion cues without manual tuning. The proposed architecture jointly performs 3D object detection, multi-object tracking, and motion forecasting within a shared BEV representation, preserving spatial alignment across tasks and supporting efficient real-time deployment. Experiments conducted on the official nuScenes validation split demonstrate that CMAA achieves 0.528 mAP and 0.691 NDS, outperforming fixed-weight fusion baselines while maintaining a compact model size and efficient inference. Additional tracking evaluation using the official nuScenes tracking devkit reports improved tracking performance, while motion forecasting experiments show reduced trajectory displacement errors (minADE and minFDE). Ablation studies further confirm the complementary contributions of adaptive modality gating and bidirectional cross-modal refinement, and a stratified dynamic analysis reveals consistent reductions in velocity estimation error across object classes, motion regimes, and environmental conditions. These results demonstrate that adaptive multimodal fusion improves robustness, motion reasoning, and perception reliability in complex traffic environments while remaining computationally efficient for deployment in safety-critical autonomous driving systems. Full article

A multiple strong coupling system comprising monolayer MoS₂ and Ag nanodisk (Ag-ND) arrays is investigated using transient absorption (TA) spectroscopy. By tuning the diameter and period of the Ag-NDs arrays, the surface plasmon polariton (SPP) resonances are made to simultaneously overlap with the A (~660 nm) and B (~608 nm) excitons of monolayer MoS₂. As a result, three distinct negative ground-state bleaching (GSB) peaks, corresponding to the upper (UP), middle (MP), and lower (LP) hybrid polariton states, were observed in the TA spectra. This confirms that a multiple strong coupling regime was achieved with both the A and B excitons of monolayer MoS₂ and SPPs modes, which was also highlighted by the anti-crossing behavior across varied Ag-NDs arrays parameters. Finally, by adding an insulating spacer layer of Al₂O₃ film, the coupling strength can be modulated from a strong coupling regime to a weak coupling regime. These results reveal a multi-exciton–plasmon strong coupling system and establish a versatile platform for ultrathin polaritonic devices, including polariton lasers and all-optical switches. Full article

Ergonomic load in human–autonomy teams is commonly treated as a static score or a post-hoc audit, even though modern sensing and communication enable real-time regulation of operator effort. We model ergonomic load as a dissipative dynamical state inferred online from multimodal effort proxies and task context, and couple it to autonomy through load-dependent gain moderation and compliance shaping. The method is evaluated on public human–swarm and human–robot interaction traces together with effort-proximal wearable and myographic datasets using a unified, windowed pipeline and controlled stress tests that emulate latency, downsampling, packet loss, and channel dropouts. On a large human–swarm benchmark, the estimator achieves strong discrimination and calibration for rare high-load events (up to AUROC $0.87$, AUPRC $0.41$, ECE $0.031$ at $q=0.90$) and degrades predictably under delay, with a knee around 300–$400\mathrm{ms}$ (AUROC $0.87\to 0.80$, ECE $0.031\to 0.061$ at $500\mathrm{ms}$). Embedding the estimate in the adaptation schedule reduces overload incidence and oscillatory redistribution while preserving coordination proxies in surrogate closed-loop simulation: overload time drops from $7.8\%$ to $4.1\%$ (relative reduction $\approx 47\%$) with throughput maintained near baseline ($1.00\to 0.97$) and oscillation power reduced ($0.26\to 0.14$) under nominal timing. These results provide a reproducible pathway for making ergonomics a control-relevant feedback signal, together with explicit operational constraints on estimator calibration (target ECE $\le 0.05$) and end-to-end latency (effective $\tau \le 300\mathrm{ms}$) required to avoid regime switching and maintain stable, interpretable adaptation. Full article

A water-rich muscle-equivalent tissue-mimicking phantom within a polymeric matrix was experimentally evaluated through a multimodal characterization methodology to determine whether it reproduces the coupled dielectric–thermal behavior of hydrated biological tissue under exposure to electromagnetic waves. The material was analyzed using thermogravimetric analysis, microwave dielectric spectroscopy from 1.5 to 4.0 GHz, VIS–NIR spectroscopy between 350 and 1200 nm, and terahertz time-domain reflection. The thermogravimetric results confirmed dominant water content, with primary mass loss below 200 °C, establishing hydration as the governing factor of its thermal response. Next, the microwave dielectric measurements show that the phantom exhibits a relative permittivity of 37.4 and an electrical conductivity of 2.4 S/m. On the other hand, the VIS–NIR spectra show smooth broadband absorption with limited spatial variability, and principal component analysis reveals macroscopic optical homogeneity without structural spectral distortion. In the THz regime, strong broadband attenuation characteristic of water-rich matrices is observed, and reflection-mode measurements enable robust assessment of temporal stability through time- and frequency-domain signatures. Finally, a microwave thermal validation demonstrates stable behavior under low-power excitation, whereas under hyperthermia-level irradiation, a significant thermal drift of −3.985 °C/h was reached under non-adiabatic conditions, identifying hydration-mediated moisture redistribution as the principal limitation during prolonged high-power exposure. Collectively, these results demonstrate cross-regime dielectric–thermal consistency while explicitly defining the hydration-driven constraints governing long-term stability, providing a validated reference material for broadband electromagnetic and thermal biomedical experimentation. Full article

Low-intensity pulsed ultrasound (LIPUS) has emerged as a versatile, non-invasive physical modality with growing potential in regenerative medicine and neural repair. Advances in ultrasound physics and biomedical engineering have enabled precise spatiotemporal control of acoustic stimulation, positioning therapeutic ultrasound as an alternative to conventional pharmacological and surgical interventions that often suffer from limited targeting and substantial side effects. Unlike high-intensity focused ultrasound, which relies primarily on thermal ablation, LIPUS operates within a low-energy, non-thermal regime and modulates cellular behavior through mechanical cues, mechano-transduction, and downstream biological responses. Accumulating evidence demonstrates that LIPUS regulates calcium dynamics, cytoskeletal remodeling, neurotrophic factor expression, inflammation, myelination, and local vascular remodeling, thereby promoting functional recovery in both peripheral and central nerve injury models. Moreover, the integration of LIPUS with biomaterials, including piezoelectric scaffolds and acoustically responsive drug delivery systems, has expanded its functionality from direct stimulation to on-demand electrical signaling and controlled therapeutic release. Despite these advances, challenges remain regarding parameter standardization, mechanistic consistency, and clinical translation. In this review, we summarize the systems, parameters, and biological mechanisms underlying LIPUS, discuss its applications in peripheral and central nerve injury repair, and highlight emerging strategies and translational barriers toward intelligent, multimodal, and personalized ultrasound-based therapies. Full article

Flatland metasurfaces provide a fundamentally distinct approach to optical gas sensing by confining light–matter interaction to planar, subwavelength interfaces, where resonant energy storage and near-field enhancement replace extended optical path lengths. This review presents a physics-driven perspective on metasurface-enabled gas sensing, focusing on how gaseous analytes perturb the complex eigenmodes of engineered planar resonators. Diverse sensing modalities, including enhanced molecular absorption, refractive index-induced resonance shifts, loss modulation, polarization conversion, and chemo-optical transduction, are unified within a common perturbative framework that links sensitivity to mode confinement, quality factor, and analyte overlap. The analysis highlights fundamental trade-offs imposed by material dispersion, intrinsic loss, and radiation balance across plasmonic, dielectric, polaritonic, and hybrid metasurface platforms operating from the visible to the terahertz regime. Attention is given to the limits of chemical selectivity in flatland architectures and to the role of functional materials, multimodal transduction, and computational inference in addressing these constraints. System-level considerations, including thermal stability, fabrication tolerance, and integration with detectors and electronics, are identified as critical determinants of real-world performance. By consolidating disparate approaches within a unified flatland framework, this review provides physical insight and design guidance for the development of compact, integrable, and application-specific optical gas sensing systems. Full article

Despite decades of studies on symmetric impinging-jet atomization, the combined role of controlled pre-impingement asymmetry and viscosity in setting the instability pathways and droplet statistics of laminar microjets remains insufficiently quantified. The effects of pre-impingement jet-length difference and liquid viscosity on the flow morphologies, instability dynamics, and atomization behavior of laminar impinging microjets are investigated experimentally using high-speed imaging. By systematically varying the jet-length asymmetry and viscosity over a range of Weber numbers, the evolution of liquid-sheet motion and breakup is resolved from synchronized front- and side-view observations. Specifically, the scientific objective of this work is to elucidate how pre-impingement jet-length asymmetry and liquid viscosity jointly regulate the dynamical behavior of laminar impinging microjets, with particular emphasis on regime transitions of liquid-sheet morphologies, the coupling between upper-sheet oscillations and rim instabilities revealed by synchronized multi-view imaging and POD-based frequency analysis and the resulting droplet-size statistics. These aspects address physical questions that have not been systematically resolved in classical impinging-jet studies, which predominantly focus on symmetric configurations or performance-oriented atomization. With increasing Weber number, the flow undergoes a sequence of regime transitions, including merged-jet, liquid-chain, wavy-rim, fishbone, closed-rim, open-rim, and arc-shaped atomization states. The presence and extent of the closed-rim regime depend sensitively on both jet-length asymmetry and liquid viscosity. Increasing jet-length difference accelerates transitions between these regimes, whereas increasing liquid viscosity stabilizes the liquid sheet and shifts the onset of unsteady breakup to higher Weber numbers. Proper orthogonal decomposition is applied to time-resolved image sequences to extract dominant oscillatory modes and their characteristic frequencies. Within the fishbone regime, the oscillation frequency of rim deformation either coincides with that of the upper region of the liquid sheet or appears as its subharmonic, indicating period-doubling behavior under specific combinations of Weber number and jet-length asymmetry. These frequency characteristics govern the spatiotemporal organization of ligament formation and detachment along the sheet rim. In the arc-shaped atomization regime, droplet-size distributions follow a log-normal form, and at sufficiently high Weber numbers, the mean droplet diameter shows only a weak dependence on jet-length asymmetry. These findings provide microscale-regime guidance for tunable droplet formation in open microfluidic jetting and related small-scale multiphase flows. The innovation of this study lies in the systematic use of synchronized multi-view imaging combined with POD-based frequency analysis and droplet statistics to directly connect liquid-sheet oscillations, rim instability dynamics, and breakup organization under controlled geometric asymmetry and viscosity variations. This approach enables a unified physical interpretation of regime transitions and instability mechanisms that cannot be resolved from single-view observations or morphology-based classification alone. Full article

Semi-supervised joint classification of Hyperspectral Images (HSIs) and LiDAR-derived Digital Surface Models (DSMs) remains challenging due to scarcity of labeled pixels, strong intra-class variability, and the heterogeneous nature of spectral and elevation features. In this work, we propose a Hybrid Mamba–Graph Fusion Network (HMGF-Net) with Multi-Stage Pseudo-Label Refinement (MS-PLR) for semi-supervised hyperspectral–LiDAR classification. The framework employs a spectral–spatial HSI backbone combining 3D–2D convolutions, a compact LiDAR CNN encoder, Mamba-style state-space sequence blocks for long-range spectral and cross-modal dependency modeling, and a graph fusion module that propagates information over a heterogeneous pixel graph. Semi-supervised learning is realized via a three-stage pseudolabeling pipeline that progressively filters, smooths, and re-weights pseudolabels based on prediction confidence, spatial–spectral consistency, and graph neighborhood agreement. We validate HMGF-Net on three benchmark hyperspectral–LiDAR datasets. Compared with a set of eight state-of-the-art (SOTA) baselines, including 3D-CNNs, SSRN, HybridSN, transformer-based models such as SpectralFormer, multimodal CNN–GCN fusion networks, and recent semi-supervised methods, the proposed approach delivers consistent gains in overall accuracy, average accuracy, and Cohen’s kappa, especially in low-label regimes (10% labeled pixels). The results highlight that the synergy between sequence modeling and graph reasoning in combination with carefully designed pseudolabel refinement is essential to maximizing the benefit of abundant unlabeled samples in multimodal remote sensing scenarios. Full article