The emissions of exhaust gas and wastewater from non-ferrous metal enterprises (NMEs) pose a significant threat to surrounding agricultural land. Accurately assessing the long-term pollutant emissions from these enterprises and the associated relative risks to nearby agricultural areas present a substantial challenge. This study, based on multi-source data, focuses on the spatial distribution of NMEs in Gejiu City, China, from 1990 to 2020. It systematically analyzes the evolution patterns, pollutant emissions, and pollution extent over three decades, innovatively developing a grid-scale long-term risk assessment framework for agricultural land surrounding these enterprises. Key findings reveal that: (1) Multi-scale geographically weighted regression (MGWR) uncovered scale-dependent effects of driving factors, with metal mineral resource density being the dominant factor (mean coefficient: 0.5766), operating at a much broader spatial scale than secondary factors like topography and traffic. (2) Non-ferrous metal smelting enterprises, particularly Pb-Zn smelters, were the predominant emission sources, contributing to over 90% of the total heavy metal (HM) load. (3) The relative risk assessment identified intensely localized high-risk hotspots in townships with dense NMEs clusters, such as Shadian, Jijie, and Datun. The integrated relative risk from both pathways identified Jijie and Shadian as extreme-risk zones. Notably, atmospheric deposition contributed to a larger area of multi-HM relative risk than wastewater discharge, highlighting it as a priority control pathway. The methodological innovation of this study lies in the integration of dynamic enterprise emissions with grid-scale receptor vulnerability, providing a precise and actionable scientific basis for targeted environmental management and risk mitigation in industrial regions.
Microcirculatory dysfunction is central to the pathophysiology of diseases such as sepsis, diabetes, and cardiovascular disorders. While clinical tools like near-infrared spectroscopy (NIRS) and transcutaneous oxygen monitoring (TcPO₂) offer global oxygenation insights, they lack the spatial resolution for localized microvascular assessment. Photoacoustic computed tomography provides sub-millimeter oxygenation imaging, yet resolving ultra-fine structures for morphology-based functional analysis remains challenging. To address this, we developed the Nine-Grid Segmentation Strategy (NGSS), a significance-analysis framework integrated with a non-invasive vascular occlusion test (VOT). Using the microcirculation-rich thenar muscle as the imaging site, NGSS characterizes each pixel across two physiological dimensions: occlusion response and reperfusion efficiency. The NGSS framework categorizes pixels based on their oxygenation trajectories during occlusion and reperfusion, assigning each to one of three functional states: significant increase, non-significant change, or significant decrease. This two-dimensional classification generates a nine-type map capturing the spatial heterogeneity of tissue microregions. Our results demonstrate that even without discerning specific microvascular morphologies, NGSS evaluates tissue oxygenation on a hundred-micron scale. The analysis reveals distinct oxygenation patterns and perfusion efficiencies across phenotypes, providing quantitative volumetric distributions. NGSS offers a novel, high-resolution, non-invasive approach for microcirculatory assessment, showing significant promise for both clinical monitoring and fundamental vascular research.
This study explores sustainable energy management approaches for a smart distribution network that combines multiple infrastructures, such as electric vehicle charging stations, hydrogen refueling facilities for fuel cell vehicles, and renewable energy systems integrated with hydrogen storage. These components are managed in a coordinated manner to satisfy both operational requirements and security criteria defined by the distribution system operator. A key feature of the hydrogen storage unit is its dual functionality, as it not only stores electrical energy but also supplies hydrogen to end users. The primary objective is to reduce overall energy losses within the distribution system. To accomplish this, the research considers several important factors, including AC power flow modeling, grid voltage operational and security constraints, system flexibility, environmental restrictions, operational characteristics of electric vehicles charging and hydrogen stations, and performance models of renewable energy systems coupled with hydrogen storage. Furthermore, the proposed framework accounts for uncertainties related to load demand, renewable generation, and variations in the number of electric vehicles by applying a scenario-based stochastic optimization technique. The findings demonstrate significant enhancements in both system performance and security. In particular, the proposed method decreases voltage deviations, power losses, and peak load capacity by approximately 24.4%, 32.8%, and 38.3%, respectively, compared to conventional load flow analyses. Moreover, voltage security within the network is improved by nearly 10.2%, confirming the efficiency of the proposed integrated energy management strategy.
Understanding spatial patterns in small, elusive species is critical for behavioral ecology and conservation, yet traditional tracking methods often face logistical constraints. We evaluated the utility of an Internet of Things (IoT)-based proximity biologging system as an automated tool for determining the activity patterns and utilization distributions (UDs) of small, free-ranging animals. We monitored female Bechstein's bats (Myotis bechsteinii) using a grid of 65 logging stations over two breeding seasons. Individual UDs were estimated using proximity data via autocorrelated kernel density estimation. We analyzed kinship-driven space-sharing and site fidelity through spatial overlap metrics. To validate system accuracy, we conducted a human-mediated field simulation for comparing proximity against GPS-derived UDs. Female bats exhibited individualized home ranges; however, mother-daughter pairs showed significantly higher overlap than non-related pairs. Repeatedly tagged individuals showed site fidelity. These results converge with previous patterns reported using VHF data. Field simulations demonstrated 88% (95% AKDE-home range) and 78% (50% AKDE-core area) of spatial congruence between proximity-based and GPS-derived UDs, confirming high locational accuracy within the detection grid. While the grid-based approach limits UD estimations to the monitored area, the IoT proximity system provides a reliable and automated alternative for studying fine-scale activity patterns. Our findings highlight the potential for this technology to identify activity hotspots and spatial dynamics in conservation-priority habitats of species with relatively small home ranges and high site fidelity, such as the Bechstein's bat.
Electroluminescence (EL) imaging is widely used for photovoltaic (PV) module inspection, yet EL defect detection remains challenging due to the need for high-resolution inputs, low-contrast defects, and strong structured background patterns. To address these issues, we propose MSW-Mamba-Det, an end-to-end defect detection framework built on RT-DETR, comprising three components. (1) MSW-Mamba, a multi-scale windowed state-space module, adopts a Local/Stripe/Grid architecture to jointly model fine details and long-range dependencies; the Stripe branch strengthens directional continuity for elongated defects, while the Grid branch introduces coarse global context to improve cross-region consistency. Saliency- and gradient-guided gating is further used to suppress background-induced false responses. (2) DetailAware compensates for detail attenuation by restoring high-frequency textures and edges through multi-scale local enhancement, and applies pixel-wise adaptive gating to integrate global semantics and mitigate smoothing effects in deep representations. (3) PAFB (Pyramid Attention Fusion Block) aligns adjacent-scale features and improves multi-scale fusion, enhancing localization stability across defect sizes. Experiments on two public EL datasets show that MSW-Mamba-Det achieves AP50:95 of 60.4% on PV-Multi-Defect-main and 68.0% on PVEL-AD, improving over RT-DETR by 2.5 points (from 57.9% to 60.4%) and 2.2 points (from 65.8% to 68.0%), respectively. MSW-Mamba-Det also outperforms 12 representative baselines, including CNN-, Transformer-, and recent YOLO-based models, in AP50:95 on both datasets, with particularly strong performance on medium and large defects. These results demonstrate the effectiveness of the proposed modules for robust PV EL defect inspection under low-contrast and structured-background conditions.
After publication of this article [1], minor production error is identified, an incorrect link was displayed under Heading 4. This error has now been corrected. The original article can be found online at: https://www.benthamscience.com/article/152908 Details of the errors and the corrections are provided here. Original: When calculating properties related to this function, such as the interaction energy between ligand and receptor in docking, instead of using the entire Cartesian space, the computational effort is substantially reduced since only the grid points are used for these calculations [14]. (https://doi.org/10.1002/jcc.21256). Corrected: When calculating properties related to this function, such as the interaction energy between ligand and receptor in docking, instead of using the entire Cartesian space, the computational effort is substantially reduced since only the grid points are used for these calculations [14].
Video-based monitoring technologies enable continuous and contactless monitoring of vital signs. This study evaluates the clinical concordance and determinants of performance of contactless video-based heart and respiratory rate monitoring compared with reference standards in a heterogeneous population of critically ill patients. In this prospective observational study, 35 intensive care unit patients were continuously monitored for 24 h. Video-based heart rate and respiratory rate were compared with the clinical reference standard. Agreement was assessed using Bland-Altman plots, intraclass correlation coefficients (ICC), and error grid analyses. Generalized estimating equations (GEE) identified factors affecting agreement. For heart rate, bias was 2.1 bpm (limits - 33.6 to 37.7), with 81.9% within ± 5 bpm and 99.3% in error grid zones A/B. ICC was 0.43. For respiratory rate, bias was - 2.4 breaths/min (limits - 14.3 to 9.5), with 63.5% within ± 3 breaths and 87.5% in zones A/B. ICC was 0.41. High heart rate, atrial fibrillation, norepinephrine administration, and movement reduced agreement for heart rate; movement reduced agreement for respiratory rate. Video-based monitoring shows promise for detecting abnormal vital signs in critically ill patients, but improved robustness to motion is needed for reliable clinical implementation.
Integrating renewable energy (RE) into power generation systems enhances sustainability by reducing greenhouse gas emissions, strengthening energy security, lowering operational costs, and promoting sustainable development, particularly in remote or underserved areas. This paper investigates the integration of RE into Mosul's power infrastructure through a hybrid renewable energy system (HRES) comprising the electrical grid, photovoltaic (PV) panels, pumped hydro energy storage (PHES), and an electrolyzer. Using HOMER Pro software, three system configurations were evaluated to optimize component sizing and assess techno-economic and environmental performance under the operating conditions of a hot semi-arid climate in northern Iraq. Among these configurations, the PV/grid/PHES/electrolyzer system demonstrated the best performance, achieving a renewable energy penetration of 254%. The proposed system results in a net present cost of $9.75 million and a levelized cost of energy of $0.06673/kWh. Despite modest reductions in operation and maintenance costs, the system demonstrates significant long-term economic efficiency when evaluated over its lifetime and projected revenues. From an environmental perspective, the proposed design achieves an annual reduction of approximately 18,089.31 tons of CO₂, corresponding to an estimated carbon credit value of $271.34 K, thus contributing to both sustainability and economic resilience. The findings confirm that the proposed HRES is a viable, cost-effective, and sustainable energy solution for Mosul and other regions with similar climatic and energy characteristics.
NASA's Goddard Earth Observing System (GEOS) infrastructure was used to couple a cloud-admitting (7-km grid, 72 levels) configuration of the GEOS atmospheric model with a mesoscale-resolving (2-4-km grid, 90 levels) Estimating the Circulation and Climate of the Ocean (ECCO) configuration of the Massachusetts Institute of Technology general circulation model (MITgcm), and to conduct a 14-month "nature" simulation initialized with January 20, 2020, 21Z conditions. The output of this simulation is contained in the dataset described here. The NASA GEOS/ECCO Coupled Nature Run includes astronomical tidal forcing in the ocean component of the simulation, an interactive aerosol component and aerosol-cloud interactions in the atmosphere, and the storage of copious amounts of model output. The inclusion of tidal forcing permits a more realistic representation of vertical mixing in the ocean and of high-frequency variability that is aliased in satellite observations. All of the above make this simulation well suited as a nature run in Observing System Simulation Experiments (OSSEs) and for the study of high frequency/wavenumber coupled processes in weather and climate.
Electric energy metering cabinets serve as critical nodes in power grid operations, providing essential protection for key components in distribution networks. Under environmental stressors, the non-metallic casings of electric energy metering cabinets are susceptible to aging-induced performance degradation, which may result in electrical safety hazards. However, rapid and precise methods for evaluating the performance of these non-metallic casings are still lacking. Laser-Induced Breakdown Spectroscopy (LIBS), capable of rapid multi-element detection with non-contact analytical advantages, was employed in this study. Thermal aging experiments were conducted to investigate the performance degradation mechanisms of sheet molding compound (SMC)-a representative non-metallic cabinet material. The research analyzed time-dependent trends in material performance and microstructural evolution during aging. By integrating LIBS with multi-analytical techniques, this study further explored the feasibility of quantitatively evaluating the bending strength of thermally aged SMC, which has rarely been reported in previous studies. Based on LIBS spectral data, bending strength characterization revealed its attenuation patterns with aging duration. The relationships between bending strength and plasma temperature, as well as the characteristic line intensity ratios of K, Al, and Ca, were systematically examined. A multivariate linear regression model incorporating these key variables was subsequently developed, yielding a high coefficient of determination (R2 = 0.9657) between the predicted and measured bending strength values. This model represents a promising initial step, but further validation with a larger dataset is necessary to enhance its reliability and generalizability.
This paper will introduce a radio frequency system to track the location of a stent designed to work inside a human artery. The stent is designed as a hemostasis aid tool for emergency situations where common surgical equipment, such as fluoroscopy systems, is not available, such as on the battlefield. In the application of interest, the stent must be guided to the correct location to achieve effective hemostasis and prevent complications. The locating approach uses the radiation pattern from the transmitter as the reference. When the transmitting frequency changes over a certain range, the measurement amplitude from a receiver depends on its relative location with respect to the transmitter. However, when the input frequency is unequal to the resonance frequency, the radiation pattern varies in an unpredictable way. To solve this problem, a deep learning model was trained to recognize variations in the radiation pattern and predict the receiver's location as one of the classes in the reference grid. The deep learning model also reduces the impact of noise and disturbing signals, which effectively improves the system's robustness.
Accurate long-term electrical load forecasting is required for reliable smart grid operation, yet it remains difficult due to multi-scale periodic patterns and non-stationary temporal variations across different prediction horizons. This paper presents MoE-Transformer, a dual-domain forecasting framework that learns to route representations in both the time and frequency domains through reinforcement learning. To mitigate spectral misalignment in multi-step forecasting, we introduce an Extended Discrete Fourier Transform (Extended DFT) that aligns the input spectrum with the frequency grid of the full prediction window. The proposed model incorporates parallel Mixture-of-Experts modules in the time and frequency domains (T-MoE and F-MoE), where domain-specific experts capture complementary temporal dynamics and spectral structures. Expert routing in each domain is modeled as an independent Markov Decision Process and optimized using reinforcement learning to jointly consider forecasting accuracy, routing consistency, and balanced expert utilization. Experiments on five benchmark datasets, including ETTh1, Electricity, and Traffic, across four forecasting horizons show that MoE-Transformer achieves MSE reductions of 50.9-56.9% relative to state-of-the-art baselines under matched training protocols. Relative to a same-capacity dense Transformer baseline on NVIDIA RTX 4090, sparse top-1 expert activation reduces peak GPU memory by [Formula: see text] and single-sample inference latency by [Formula: see text] (mean ± std over 5 runs), with measured absolute batched latency of [Formula: see text] ms per sample, supporting real-time forecasting deployment. Ablation results confirm the individual effects of Extended DFT, dual-domain modeling, and reinforcement-based routing, yielding performance gains of 5.8%, 4.6%, and up to 47.2%, respectively.
Climate change is expanding the suitable habitat range for mosquito species into previously uninhabitable alpine regions, necessitating tools for projecting vector establishment risk. This study presents the development of an agent-based cellular automaton simulation platform for modeling the establishment and spread of Aedes aegypti and Aedes albopictus mosquito populations in Tyrol, Austria. We developed a Java-based simulation platform integrating high-resolution weather data from meteorological stations for the year range of 2019-2024 with climate projections under RCP 2.6 and RCP 4.5 scenarios for the year range 2036-2040 and 2076-2080. The simulation uses a 100×100 cell grid (33×33 meters per cell) overlaying a region in Innsbruck, Austria, modeling species as agents with temperature-dependent development rates, stage-specific mortality, and dispersal patterns. The platform enables comparison of vector population dynamics under current and projected climate conditions. Preliminary simulations demonstrate expected biological patterns, including seasonal population dynamics, species-specific cold tolerance differences, and spatial spread patterns. The platform provides a scalable computational foundation for evidence-based public health adaptation strategies, enabling proactive risk assessment for vector-borne diseases in alpine regions.
Visual foraging lies at the intersection of visual perception, decision-making and action planning. An attractive feature of this paradigm is that it generates a rich stream of sequential decision data. However, this presents a number of challenges for analysis. To this end, we have developed FoMo, a robust and flexible generative model for spatial-sequential data that allows prediction of participants' selection behaviour on a target-by-target basis. Building upon our initial work, we present an updated version of FoMo (Clarke AD, Hunt AR, Hughes AE. Foraging as sampling without replacement: A Bayesian statistical model for estimating biases in target selection. PLOS Computational Biology. 2022;18(1):e1009813.)‌‌, which incorporates spatial structure allowing us to model organised spatial behaviours. FoMo provides estimates of a range of interpretable parameters, meaning we can use it to understand the causes of behavioural differences: for example, incorporating spatial-structure parameters improves model prediction accuracy for a number of visual foraging datasets, predominantly due to improvements for a subset of participants who use grid-following strategies. Our approach can also account for individual differences across the wide range of descriptive statistics that have previously been used to explore human and non-human animal behaviour, providing a unified framework for analysing these data.
This study presents a systematic decision-making approach for tuning hyperparameters of machine learning (ML) models that employ the cross-validation technique in their learning process. It provides a more efficient and precise alternative to conventional hyperparameter optimization methods (i.e., grid search, random search, and Bayesian optimization) and demonstrates that bi-level data-driven optimization enhances this task, especially when non-linear loss functions are used in the training process. Hyperparameters are external adjustable parameters that configure and control the learning process of ML models and cannot be estimated directly from the training data. Their tuning plays a crucial role in constructing accurate and generalized ML models. However, this process is often treated as a trial-and-error search, where numerous hyperparameter combinations are evaluated through training and validation, and the best-performing one is selected after an exhaustive search. An alternative is to pose hyperparameter tuning as a bi-level optimization problem, which explicitly captures the interdependence between hyperparameter selection and model evaluation. This formulation enables a more structured optimization strategy but introduces significant algorithmic and computational challenges. To address this, we use the Data-driven Optimization of bi-level Mixed-Integer non-linear problems (DOMINO) framework to approximate the bi-level formulation of the cross-validated hyperparameter optimization problem as a single-level problem. This transformation enables the use of data-driven methodologies to solve the otherwise intractable bi-level problem more efficiently, while still capturing the unknown interactions between hyperparameter choices and model performance. We evaluate 17 different data-driven optimization algorithms, including heuristic vs. deterministic methods, local vs. global approaches, and sample-based vs. model-based algorithms, on six hyperparameter tuning problems for regression and classification tasks. Our results show that the data-driven bi-level approach outperforms conventional tuning algorithms in predictive accuracy on blind test datasets and yields ML models that exhibit excellent generalization across all case studies. We further observe that local optimization algorithms integrated into DOMINO are more computationally efficient when tuning models with a single hyperparameter, whereas global algorithms are more effective for models involving multiple hyperparameters or non-linear characteristics.
To investigate the inevitable aging of composite insulators under the coupled effects of electrical, thermal, ice, and fog stresses, as well as to explore their aging mechanisms and residual strength prediction methods, this study collected operational insulator samples from four environmental regions: Tibet, Yunnan, Hunan Xuefeng Mountain, and Anhui/Chongqing. Mechanical properties, including tensile strength, elongation at break, and shear resistance, were tested. The results indicate that the degradation of mechanical performance in composite insulation components can be attributed to the synergistic interaction of operational environments and material characteristics, with the aging behavior of high-temperature vulcanized (HTV) silicone rubber exhibiting significant non-linearity. Based on existing research, molecular dynamics simulations were employed to construct microstructural models at different aging stages, and it was verified that main chain scission, reduced system density, and changes in the elemental chemical environment during aging are closely related to the degradation of material mechanical properties. Based on hyper-elastic constitutive theory and fracture mechanics, a quantitative method for assessing the comprehensive aging degree was proposed, with "service years" and "operational altitude" as the core dimensions. A negative exponential model was established to describe the strength degradation of silicone rubber materials. This model enables the non-destructive estimation of the residual mechanical strength of in-service insulators in complex regions without power interruption, providing a decision-making framework for grid operation and maintenance.
The focus of this work was to investigate the mechanical properties of additively manufactured (AM) discontinuous carbon fibre-reinforced polymer (DCFRP) composites. Towards the specimen's fabrication, the Fused Filament Fabrication (FFF) additive manufacturing technique was employed. A number of input printing parameters were varied, such as the infill pattern, infill density, layer height, shell configuration, and raster orientation, in a systematic way. The role of these paraments on the mechanical properties, such as tensile, flexural, and impact strength were investigated. The data was analysed in-depth and the "main effect method" was employed for their comparative ranking. The results of this study showed that tensile and bending strengths were strongly correlated with material content and structural reinforcement. The specimens attained up to 76.7 MPa of tensile strength, while the flexural strength was up to 159.4 MPa, with a deflection of up to 8 mm and 16 mm, respectively. Solid infills, higher densities, finer layer heights, and added shells significantly improved the strength and stiffness. Grid-patterned and low-density specimens caused poor load-bearing capacities, while hexagonal and gyroid infills offered a more balanced performance.
Knowledge about the condition of electrical equipment in energy networks is of great importance to network operators. Partial discharges are a key parameter for evaluating the health of the insulation. While a quantifiable PD measurement for offline tests is state of the art, it is costly and labour-intensive. It, therefore, makes sense to carry out permanent monitoring during operation. At the medium-voltage level in the European interconnected grid, comprehensive monitoring of PD is not implemented. This study presents a novel sensor concept that is used to detect PD in medium-voltage switchgear and cables: the so-called Magnetic Flux Concentrator Sensor (MFCS). It is an inductive sensor concept with high sensitivity in the frequency range of a few MHz, like well-established High-Frequency Current Transformers (HFCTs) but with better magnetic saturation properties in specific use cases. The highly permeable ferrite core of the MFCS is unconventionally shaped, resulting in a higher-saturation field strength. Therefore, this sensor is not driven into saturation by the operating currents of typical MV power cables. Using the MFCS and conventional HFCT in a suitable combination enables direction-selective PD detection. This work presents the sensor concept and the method for directional detection of the PD location, as analysed and evaluated theoretically and practically with laboratory experiments.
Sodium-ion batteries (SIBs) represent a compelling alternative to lithium-ion batteries for grid-scale energy storage, owing to the high natural abundance and low cost of sodium resources, as well as their strategic alignment with national energy security priorities. Nevertheless, the sluggish Na+ diffusion kinetics and limited specific capacity of anode materials continue to impede practical deployment. Herein, nitrogen-doped carbon-coated TiO2 nanofibers (TiO2/C-N) were rationally engineered through a facile electrospinning route integrated with synergistic defect and coating engineering. The in situ-formed N-doped carbon shell establishes a continuous, high-conductivity electron-transport network while simultaneously buffering volumetric strain during repeated (de)sodiation, thereby preserving long-term structural integrity. Electrochemical assessments demonstrate that the TiO2/C-N electrode delivers a reversible specific capacity of 233.64 mAh g-1 at 0.1 A g-1 (initial Coulombic efficiency 54.13%). Quantitative kinetic analysis reveals a pronounced pseudocapacitive contribution of 41.4% at 1.2 mV s-1, confirming a surface-controlled Na+ storage pathway that markedly enhances rate capability. Moreover, the electrode retains 245.5 mAh g-1 after 150 cycles at 1 A g-1, underscoring exceptional cycling stability. This work elucidates the synergistic regulation of N-doped carbon coating and pseudocapacitive kinetics in TiO2-based anodes, offering a robust design strategy for high-rate, long-cycle-life SIB anodes.
This paper presents a statistical framework for early warning change-point detection in electrical grid frequency time series. Frequency deviations outside the tolerance band of 49.85-50.15 Hz are treated as error events. A high-volatility (HV) measure is computed using a rolling-window approach and compared against a Hoeffding-bound threshold to identify significant transitions that may precede hazardous excursions. A dataset of 1250 error-event sequences collected over six months is divided into training (34%), validation (33%), and testing (33%) subsets. To improve efficiency, k-means clustering and dynamic time warping (DTW) are used to select representative training sequences, and a mapping-with-regression procedure is applied to generate warning signals. Experimental results show that the proposed method achieves 98.04% accuracy and an F1-score of 98.06%, while maintaining a false-negative rate of 1.1%. Lead-time evaluation confirms consistent early detection, and baseline comparison against deep learning approaches, demonstrates competitive performance with low computational cost.