Search Results (372)

The semiconductor industry faces converging pressures from energy-constrained edge electronics and energy-bottlenecked high-performance computing, motivating heterogeneous monolithic three-dimensional (M3D) integration as a system-level response. M3D imposes a strict back-end-of-line (BEOL) thermal budget on upper-tier devices, restricting the channel materials and contact processes available and degrading conventional thin-film transistor performance. The source-gated transistor (SGT), in which drain saturation is set by gate-modulated injection across an engineered source barrier rather than by drain-side channel pinch-off, provides a device-level response: low saturation voltage, high output impedance, large intrinsic gain, and tolerance to channel-length variation, all achieved with moderate-mobility and nonideal-contact channel materials. This review organizes reported SGTs by source-barrier architecture and channel-material platform, develops a spatial characterization framework that complements electrical measurements for unambiguous identification of source-controlled operation, and surveys applications across standalone edge electronics and BEOL-compatible upper tiers in M3D stacks. Integrating non-volatile memory mechanisms into the source barrier further extends SGTs into a compute-in-memory and neuromorphic upper-tier role in which the voltage-invariant saturation current itself functions as a programmable, read-bias-robust state variable. Together, these considerations position SGTs as a flexible architectural primitive for heterogeneous M3D platforms that address the energy demands of both edge and high-performance computing. Full article

Long-term displacement in ultra-high rockfill dams is spatially heterogeneous, yet point-based monitoring and single-track InSAR provide only a limited account of its three-dimensional evolution. To address this limitation, we develop a workflow combining multi-track three-dimensional InSAR reconstruction, temporal feature clustering and modified HST/HTT modeling. In the modified models, the conventional linear time term is replaced by a predominant displacement component, allowing nonlinear long-term behavior to be represented alongside hydraulic and thermal responses. Three Sentinel-1 tracks acquired between 2014 and 2023 were used to analyze the Pubugou Dam. Maximum vertical and eastward displacement rates reached 17.70 and 19.41 mm/yr, respectively, while cumulative vertical displacement reached approximately 178 mm. Three coherent displacement zones were resolved: the upper dam was dominated by long-term accumulation, the central section displayed the strongest periodic response, and the lower dam and abutments remained comparatively stable. The modified models reproduced the observed series more closely in sample than the conventional formulations, while the fitted coefficients indicated a stronger and more spatially variable response to reservoir level than to air temperature. This integrated analysis provides a spatially resolved account of long-term displacement and environmental response across an ultra-high rockfill dam. Full article

Environmental monitoring of freshwater ecosystems faces significant challenges in detecting illicit wastewater discharges, which often remain unrecognized due to their intermittent nature and limited spatial footprint. This study presents a novel integrated strategy combining satellite remote sensing, Unmanned Aerial Vehicle (UAV)-based proximal sensing, and in situ measurements to enhance pollution detection in vulnerable aquatic environments. The methodology was applied to Lake Avernus (Italy), a volcanic lake historically affected by eutrophication and toxic cyanobacterial blooms. Landsat 8–9 thermal analysis revealed no detectable anomalies, reflecting the limitations of its coarse spatial resolution. Sentinel-2 multispectral imagery was then analyzed through spectral indices, band ratios, and reflectance signatures, revealing localized variations in surface reflectance and spatial heterogeneity in water optical properties. These satellite-derived anomalies guided targeted high-resolution UAV surveys. UAV-based thermal imaging revealed an elevated-temperature zone along the adjacent shoreline. In situ field screening flagged a candidate chemical anomaly at this location. The hierarchical framework demonstrates that satellite screening effectively identifies areas of concern, while UAV thermal imaging enables high-resolution localization of features invisible to satellite sensors, and in situ measurements provide essential ground-truth validation. This replicable, low-cost methodology offers a powerful tool for early warning, surveillance, and sustainable management of sensitive freshwater ecosystems. Full article

Amid the low-carbon transition, university teaching buildings feature high occupancy and energy use, making the synergistic enhancement of their daylighting and thermal environments crucial for “dual carbon” goals. However, traditional “north–south homogenization” designs in cold regions fail to address the spatial heterogeneity of solar radiation and climate resources, intensifying the trade-off between natural daylighting and Heating Energy Use Intensity (E_h) while restricting space performance optimization. Focusing on a typical cold-region teaching building, this study proposes a “parametric modeling–multi-objective optimization–machine learning” integrated framework. Targeting spatial daylight autonomy (sDA), useful daylight illuminance (UDI), and E_h, we compared the homogeneous baseline model with the Pareto-optimal solution set, demarcated key design parameter boundaries, and developed an ensemble-based rapid prediction model. Based on the parametric simulation analysis of this representative case building in a cold region, results indicate that: (1) Compared to the baseline, the overall optimal scheme reduced E_h by 17.43% while increasing UDI and sDA by 12.0% and 10.5%, respectively. (2) The Pareto set strictly converges toward a due-south orientation and a “deep-south, shallow-north” layout (depth ratio: 0.66–0.77); thermal configurations exhibit “enhanced northern insulation and southern heat gain,” confirming heterogeneous design matches cold climates better. (3) The four constructed machine learning models (MLP, LightGBM, XGBoost, and Random Forest) uniformly achieved test recall rates exceeding 99%, enabling highly precise, rapid classification of top-performing design scenarios during early-stage design. This study overcomes climate-matching blindness in traditional design, providing a multi-objective synergistic optimization path balancing low energy and high-quality daylighting with substantial engineering and theoretical value. Full article

Polystyrene (PS) thermal degradation involves localized molecular bond-cleavage events that are not directly equivalent to the apparent kinetic parameters obtained from bulk thermal analysis. In this study, a finite hydrogen-terminated PS oligomeric model was examined using density functional theory at the M06-2X/LANL2DZ level, whereas the non-isothermal degradation behavior of a PS sample was independently evaluated by thermogravimetric analysis under nitrogen. The computational analysis considered frontier molecular orbital distributions and site-specific thermodynamic descriptors associated with homolytic C–C cleavage and radical-mediated β-scission reactions. The calculated HOMO–LUMO gap of 742.62 kJ mol⁻¹ indicated a comparatively large orbital-energy separation within the selected oligomeric model, while the localization of the frontier orbitals over aromatic and benzylic regions revealed a spatially heterogeneous electronic distribution. Homolytic C–C cleavage exhibited bond dissociation energies ranging from 414.09 to 481.24 kJ mol⁻¹, demonstrating that the thermodynamic requirement for radical generation depends on the local molecular environment of the evaluated structure. The Gibbs free-energy changes calculated for the selected radical β-scission reactions ranged from 55.44 to 189.41 kJ mol⁻¹. These quantities represent model-dependent reaction thermodynamics and should not be interpreted as activation barriers because transition states were not calculated. Thermogravimetric analysis showed systematic increases in Tonset and Tmax with increasing heating rate, consistent with kinetic delay and thermal-lag effects under non-isothermal conditions. The Kissinger method yielded a global apparent activation energy of 186.61 kJ mol⁻¹, whereas the residual-mass-corrected Flynn–Wall–Ozawa and Kissinger–Akahira–Sunose methods produced average apparent activation energies of 180.81 and 178.49 kJ mol⁻¹, respectively, over α = 0.05–0.95. Across the same conversion interval, the FWO apparent activation energy increased from 143.10 to 221.71 kJ mol⁻¹, while the KAS values increased from 140.10 to 220.23 kJ mol⁻¹, indicating an evolving macroscopic degradation response with greater uncertainty toward high conversion. The computational and experimental datasets were therefore interpreted as complementary but non-equivalent scale-dependent descriptions: DFT compares the relative thermodynamics of selected molecular reactions within a finite isolated oligomer, whereas TGA characterizes the global apparent kinetic behavior of the condensed polymer sample. No direct numerical correspondence was established between the molecular reaction energies and the TGA-derived apparent activation energies, and no individual cleavage reaction was assigned to a specific conversion interval. Extrapolation of these results to high-molecular-weight, polydisperse, additive-containing, cross-linked, or environmentally aged PS microplastics should therefore be made with caution. Full article

Unmanned aerial vehicle (UAV)-based multispectral remote sensing has emerged as a promising tool for monitoring crop physiological status and supporting precision disease management. However, the capacity of multispectral vegetation indices to detect foliar diseases under commercial field conditions remains insufficiently understood, particularly in perennial crops grown in humid tropical environments. This study evaluated the potential of UAV-derived multispectral vegetation indices to assess the physiological response of avocado (Persea americana Mill.) canopies affected by anthracnose caused by Colletotrichum fructicola in Amazonas, Peru. Disease incidence and severity were assessed through field evaluations, while multispectral imagery was acquired using a UAV equipped with a MicaSense RedEdge-MX Dual sensor. Vegetation indices including the Normalized Difference Vegetation Index (NDVI), Green Normalized Difference Vegetation Index (GNDVI), Soil-Adjusted Vegetation Index (SAVI), Normalized Difference Red Edge Index (NDRE), Red Edge Chlorophyll Index (CIred-edge), Plant Senescence Reflectance Index (PSRI), and Visible Atmospherically Resistant Index (VARI) were calculated, and their relationships with anthracnose incidence were analyzed using Spearman’s rank correlation. Field observations confirmed a high incidence of foliar anthracnose with a heterogeneous spatial distribution across the orchard. Multispectral imagery successfully characterized spatial variability in the canopy physiological condition, revealing differences in vegetation vigor, chlorophyll-related reflectance, and senescence among experimental blocks. Nevertheless, none of the evaluated vegetation indices showed statistically significant correlations with anthracnose incidence (p > 0.05), indicating that spectral variability primarily reflected the general canopy physiological status rather than a disease-specific spectral response. These findings demonstrate that UAV-derived multispectral vegetation indices are valuable for monitoring spatial variability in the avocado canopy condition but have limited capability for independently detecting anthracnose under humid tropical field conditions. Future studies integrating multi-temporal UAV acquisitions, hyperspectral and thermal imagery, LiDAR, environmental variables, and machine-learning approaches are expected to improve the early detection and spatial prediction of anthracnose in avocado production systems. Full article

Rapid urbanization in tropical Asia has fundamentally transformed land use–land cover while intensifying urban thermal stress, yet the relationship between land cover change and thermal conditions is frequently assumed to be spatially uniform. This study challenges that assumption by demonstrating that land cover–thermal relationships in Colombo District, Sri Lanka, are highly spatially and temporally heterogeneous, with statistically significant associations detected in only 17–47% of the study area in any given year, underscoring that context, not land cover type alone, governs thermal outcomes. Using multi-temporal Landsat satellite imagery, LULC maps were derived, and the urban heat island effect (UHIE) and urban thermal field variance index (UTFVI) were calculated for seven time periods (1989, 1996, 2002, 2009, 2014, 2019, 2024). Geographically weighted regression (GWR) was applied to model local relationships between LULC classes, namely wetland vegetation, water bodies, impervious surfaces, and other pervious surfaces, and thermal indices across a 500 m spatial grid, revealing a 74% loss in wetland vegetation and a 326% increase in impervious surfaces over the study period. Water bodies exhibited spatially variable cooling effects relative to wetland vegetation, most pronounced in eastern regions during earlier periods, while impervious surfaces showed consistent, spatially persistent warming effects concentrated in western and southern urban cores. By coupling GWR with a 35-year multi-sensor time series, this study provides a spatially explicit, longitudinal account of how land cover–thermal relationships evolve as tropical urbanization intensifies, offering an evidence base for spatially targeted rather than uniform climate adaptation planning in rapidly urbanizing tropical cities. Full article

Strong heterogeneity in unconventional reservoirs leads to complex fracture propagation and challenges in quantitative stimulation evaluation. This study integrates true triaxial fracturing experiments, three-dimensional CT reconstruction, multi-field coupled numerical simulation, and multiple linear regression to investigate the fracture behavior of Gulong shale (Q1, Q9) and tight sandstone under supercritical carbon dioxide (SC-CO₂) fracturing. A fracture complexity index (FCI) that incorporates fractal dimension, spatial uniformity, and aperture distribution is proposed as a quantitative metric. The results show that SC-CO₂ significantly reduces breakdown pressure and increases fracture complexity compared to water. For Q9 shale, SC-CO₂ gives a breakdown pressure of 32.91 MPa (10.46% lower than water), a fractal dimension of 2.41, and an FCI of 8.92. In tight sandstone, the SC-CO₂ breakdown pressure is 34.12 MPa, whereas water increases it to 44.50 MPa; the fractal dimension and FCI are only 2.05 and 3.40, respectively, lower than those of shale fractured with water. Multiple linear regression quantifies contribution weights: lithological weak-plane development dominates fracture complexity (41.6%), far exceeding the brittleness index. The injection rate mainly controls stimulation scale: the damage area ratio rises from 1.79% to 2.90% when the rate increases from 10 to 40 mL/min. The horizontal stress difference is key to complexity enhancement: the fractal dimension increases from 1.9230 to 1.9901 as the stress difference rises from 0 to 4 MPa. The numerical simulations further reveal the coupled thermal-hydraulic-mechanical effects. The proposed FCI-based evaluation and regression models provide a quantitative framework for optimizing SC-CO₂ fracturing design in heterogeneous unconventional reservoirs. Full article

In recent years, UAV-based visible and thermal (RGB-T) multimodal object detection has demonstrated significant potential for monitoring forest fires in complex environments. However, restricted by the intrinsic discrepancies between modalities, existing RGB-T models still struggle to achieve optimal detection accuracy. To address the aforementioned issues, this paper proposes FireRGBTNet, a lightweight and efficient RGB-T fusion model for UAV-based forest fire detection. First, a heterogeneous dual-stream backbone is designed to precisely capture modality-specific information via customized feature extraction modules, utilizing Target-Enhanced Downsampling to adaptively preserve the fragile features of small targets. Second, a multi-scale semantic alignment enhancement branch is proposed to explicitly bridge the semantic gap through the dual constraints of statistical distribution and semantic direction. Finally, a multimodal spatial gated fusion module is constructed, which utilizes spatial gated interactions to effectively suppress multimodal noise and incorporates Mish Gated Linear Units to compensate for global semantics, thereby achieving high-quality heterogeneous feature fusion. Extensive experiments on the RGBT-3M dataset demonstrate that with a mere 4.13M parameters, FireRGBTNet achieves accuracies of 95.9% and 62.8% in terms of mAP@0.5 and mAP@0.5:0.95, respectively. The proposed model achieves an optimal trade-off between detection accuracy and computational efficiency, providing a highly effective solution for UAV-based forest fire monitoring in complex environments. Full article

Accurate energy efficiency prediction is fundamental to green computing and low-carbon 6G infrastructure. Data centers represent a challenging testbed due to their high energy density and complex device interactions. Existing methods either ignore physical laws or fail to capture spatial dependencies among heterogeneous devices. To address these limitations, this paper proposes a hybrid framework integrating Physics-Informed Neural Networks (PINNs) with Graph Neural Networks (GNNs) for Power Usage Effectiveness (PUE) prediction. Two algorithmic variants are developed: Physically Decomposed PINN-GNN (PDPG) and Unified End-to-End PINN-GNN (UEPG). Physical regularization constraints derived from practical energy and thermal principles are embedded into model training to improve the physical rationality of the prediction results. Validated on real-world data center datasets, the proposed method achieves superior accuracy and robustness over mainstream baselines, providing reliable support for cooling and resource management. Full article

The Upper Yellow River (UYR) basin is the predominant runoff-yielding area of the entire watershed. Accurately simulating annual runoff is crucial for water resources management. While the Budyko framework effectively captures long-term hydro-thermal equilibrium, it struggles to represent nonlinear dynamics, flow channel routing, and spatial interconnections. Here, we propose a hybrid physics- and data-driven approach by coupling the Budyko framework (Fu’s equation) with an eXtreme Gradient Boosting (XGBoost) model, integrating upstream channel routing and antecedent storage-lag features using long-term hydrologic observations for nonlinear runoff simulation and driver attribution. To resolve the feature multicollinearity on machine learning attributions, input variables were consolidated into three groups: precipitation driven, evaporation limitation, and flow storage lag. The results demonstrate that the Budyko–XGBoost coupled model enhances annual runoff prediction accuracy compared to the standalone Fu equation and pure XGBoost, raising the coefficient of determination (R²) to 0.63–0.86 (mean R² = 0.75) and capturing both nonlinear dynamics and turning points, alongside reductions of 6.2% in the mean RMSE (18.77 mm) and 11.0% in the MAE (13.66 mm) compared to the pure XGBoost model (mean R² = 0.70, RMSE = 20.01 mm, and MAE = 15.35 mm). Group-level SHAP attributions reveal that flow storage-lag drivers (R_lag and $R_{\mathit{lag}}^{\prime }$) exert a primary control on runoff evolution across all the stations. Spatially, secondary drivers exhibit heterogeneity: in relatively humid, energy-limited regions (Maqu), high precipitation promotes positive runoff deviations, whereas in arid/semi-arid, water-limited reaches (e.g., Guide, Xunhua, Xiaochuan, and Lanzhou stations), high precipitation is absorbed by severe soil moisture deficits and reservoir interception, exerting a negative effect on runoff deviation. Full article

Functionally Graded Materials (FGMs) vary in composition and properties for tailored structural performance. Additive Manufacturing (AM), particularly Wire Arc Additive Manufacturing (WAAM), offers a scalable route to metallic FGMs, but manufacture and mechanical qualification remain disconnected. Scopus and Web of Science Core Collection were searched for eligible publications up to and including 31 July 2026. The review includes 176 studies, and every conclusion is graded by a review-specific certainty scheme and bounded to this corpus. The review connects FGM manufacture with full-field inverse identification of spatially varying properties. Evidence indicates interface-defect risk depends on metallurgical compatibility and the composition path. Intermetallic-forming or thermally mismatched pairs remain vulnerable despite process optimisation. Stainless-steel–Ni-superalloy combinations are more frequently reported as sound but only within the composition intervals validated in the cited builds: this apparent advantage reflects unequal study numbers and cracking in specific composition windows. A four-stage sequence—screening phase stability and thermal-expansion mismatch before optimising deposition parameters, then validating the complete composition path—is proposed as an evidence-informed framework rather than a validated decision map. Conventional tests generally provide averaged or location-specific properties rather than a continuous constitutive gradient. Digital image correlation coupled with inverse identification methods can enable the determination of spatially varying properties from a heterogeneous test. However, no experimental study identified a spatial constitutive law across a deliberate WAAM compositional gradient. Full article

Extreme heatwaves intensify urban heat islands and pose increasing risks to urban sustainability and human health. However, how urban morphology is associated with daytime and nighttime land surface temperature (LST) across urban functional zones (UFZs), particularly under heatwave conditions, remains insufficiently understood. To address this gap, this study integrates daytime and nighttime LST data derived from SDGSAT-1, multi-dimensional urban morphology indicators, and two interpretable ensemble models (XGBoost and GWRF) to investigate overall sample-level nonlinear model-based associations between urban morphology and LST and to explore spatial variation in local predictor importance within Beijing and Nanjing, China. Because the daytime and nighttime scenes were not always paired within the same heatwave episode, the analysis focuses on selected heatwave-condition observations. The results show marked contrasts between the selected daytime and nighttime observations in UFZ-level thermal patterns. Industrial zones generally exhibited the highest daytime LST, whereas residential zones showed the highest nighttime LST. Building density was identified as the primary model-based predictor of daytime LST in both cities, although its association with LST was nonlinear and varied across density ranges. In contrast, nighttime LST was characterized by more heterogeneous predictor associations, involving vegetation structure, sky openness, building form, anthropogenic indicators, and material-related variables, with their relative importance differing across cities and UFZ types. Local predictor-importance patterns also varied across neighborhoods, cities, and observation times, indicating that model-identified locally important predictors were not spatially uniform within each city. These findings highlight the potential of SDGSAT-1 daytime and nighttime thermal observations and interpretable machine learning for screening candidate local thermal priority areas and key morphology-related factors under heatwave conditions. Full article

The semi-underground granary is a new type of energy-saving grain storage facility that can use shallow geothermal energy to reduce energy consumption during grain storage. However, unclear temperature fields and the lack of grain pile temperature control methods are not conducive to the design and application of semi-underground granaries. In this study, the temperature fields and temperature control methods for grain piles in a semi-underground granary were numerically investigated by using an experimentally verified COMSOL model and a collaborative simulation method combining steady-state heat transfer and dynamic heat transfer. Multiple grain storage temperature control methods for the semi-underground granary were presented to improve grain storage safety, including an intermediate floor slab, an embedded-pipe wall, floor burial depth, and envelope insulation. The results showed that there was significant spatial heterogeneity in the temperature field distribution of the grain pile in the semi-underground granary. The large thermal inertia of the soil and the stable low-temperature soil environment reduced the influence of outdoor air temperature variations on the grain pile temperature field. Installing an intermediate floor slab could achieve natural low-temperature grain storage in the underground section of the semi-underground granary. An embedded-pipe wall could effectively solve the problem of local temperature increases in grain piles caused by heat transfer through the granary walls. The floor burial depth of the semi-underground granary was a key influencing factor of heat transfer through the granary wall. Granary wall thickness had a significant impact on the thermal performance of the walls and the grain pile temperature field due to changes in wall insulation. These results can provide beneficial suggestions for guiding the design of grain storage temperature control methods in semi-underground granaries. Full article

Martian landslide segmentation is challenging because annotated samples are limited and landslide deposits can have weak boundaries, heterogeneous textures, and visual similarity to crater rims and canyon walls. This study evaluates a terrain-residual and boundary-assisted network (TRB-Net) on the locally available MMLSv2 train/validation/test split. TRB-Net combines RGB texture with digital elevation model (DEM), slope, thermal inertia, and grayscale information through terrain-residual fusion, an atrous spatial pyramid pooling decoder, and auxiliary boundary supervision. The compact evaluation checkpoint, using a validation-selected threshold of 0.55, achieves an mIoU of 0.8060, foreground IoU of 0.7502, F1-score of 0.8573, precision of 0.8473, and recall of 0.8676 with 5.255 million parameters. In same-split comparisons, DeepLabV3+ obtains the highest overlap scores, while TRB-Net provides competitive segmentation and an explicit architecture for tracing how terrain and boundary cues enter the prediction. These results apply only to the local MMLSv2 split; geographically isolated and large-area Martian mapping performance were not evaluated. Full article