Search Results (1,420)

Predicting road crash severity remains a major challenge in transportation safety research, requiring models that combine predictive accuracy, interpretability, and computational efficiency. This study introduces a Multi-Risk Index based on Copula Integration (MRI-Copula)—a hybrid framework that integrates Categorical Boosting (CatBoost) with SHapley Additive exPlanations (SHAP) and Vine Copula dependence modeling to assess and predict crash severity. The approach leverages CatBoost–SHAP to quantify the marginal contribution of each risk factor while maintaining model transparency and employs copula-based tail dependence to capture the joint escalation of risk under extreme crash conditions. Using a dataset of 877 police-reported crashes from Jeddah, Saudi Arabia, the framework constructs three interpretable sub-indices—Environmental Risk Index (ERI), Behavioural Risk Index (BRI), and Systemic Risk Index (SRI)—representing distinct domains of crash causation. These indices are combined through a convex weighting parameter (α), optimized via cross-validation (optimal α = 0.80), ensuring a balanced integration of predictive and dependence-based information. Comparative evaluation across multiple classifiers—CatBoost, Light Gradient Boosting Machine (LightGBM), Histogram-based Gradient Boosting (HistGB), and Logistic Regression—demonstrated the robustness of the framework. The CatBoost + MRI-Copula configuration achieved the highest predictive performance (AUC = 0.986; F1 = 0.904), while LightGBM and HistGB offered comparable accuracy (AUC ≈ 0.958; F1 ≈ 0.89) at a fraction of the computational time (≤1 s versus 32 s for CatBoost), highlighting a trade-off between analytical precision and scalability. Consequently, the MRI-Copula framework provides a transparent and theoretically grounded foundation for data-driven road safety management. It bridges predictive analytics and decision support offering a scalable, interpretable, and policy-relevant tool for proactive crash risk mitigation. Full article

Against the backdrop of rapid population aging and a high prevalence of cognitive impairment in China, identifying modifiable environmental risk factors is a public health priority. Although environmental noise is widely recognized as a significant stressor, its effects on cognitive health remain underexplored within the Chinese context. Drawing on balanced panel data from three waves of the China Health and Retirement Longitudinal Study (CHARLS), we examined 3459 individuals aged 45 and above to assess the association between noise pollution and cognitive function using a two-way fixed-effects model. Additionally, we employed a chained mediation approach to investigate whether sleep disturbances and depressive symptoms serve as intermediary mechanisms. The findings indicated a significant inverse relationship: each unit increase in the noise pollution index corresponded to a 0.41-point reduction in overall cognitive scores. These results were robust across various noise exposure measures. Sensitivity analyses using alternative noise metrics also supported this finding. Sleep duration and depression were identified as significant mediators in the relationship between noise pollution and cognitive decline. This longitudinal analysis offers compelling evidence that environmental noise constitutes a substantial risk factor for declining cognitive function in middle-aged and older adults in China. Full article

Carbon dioxide (CO₂) has been widely applied in gas flooding for reservoir development due to its remarkable oil recovery potential. However, because its viscosity is lower than that of water and most crude oils, severe channeling often occurs during the flooding process, resulting in a significant reduction in the sweep efficiency. To address this issue, foam flooding has attracted considerable attention as an effective method for controlling CO₂ mobility. In this study, a compound foam system was developed with alpha-olefin sulfonate (AOS) as the primary foaming agent, alcohol ethoxylate (AEO) and cetyltrimethylammonium bromide (CTAB) as co-surfactants, and partially hydrolyzed polyacrylamide (HPAM) as the stabilizer. The optimal system was screened through evaluations of comprehensive foam index, salt tolerance, oil resistance, and shear resistance. Results indicate that the AOS+AEO formulation exhibits superior foaming ability, salt tolerance, and foam stability compared with the AOS+CTAB system, with the best performance achieved at a mass ratio of 2:1 (AOS:AEO), balancing both adaptability and economic feasibility. A heterogeneous reservoir model was constructed using parallel core flooding to investigate the displacement performance and blocking capability of the system. Nuclear magnetic resonance (NMR) imaging was employed to monitor in situ oil phase migration and clarify the recovery mechanisms. Experimental results show that the compound foam system demonstrates excellent conformance control performance, achieving a blocking efficiency of 84.5% and improving the overall oil recovery by 4.6%. NMR imaging further reveals that the system effectively mobilizes low-permeability zones, with T₂ spectrum analysis indicating a 4.5% incremental recovery in low-permeability layers. Moreover, in reservoirs with larger permeability ratio, the system exhibits enhanced blocking efficiency (up to 86.5%), though the incremental recovery is not strictly proportional to the blocking effect. Compared with previous AOS-based CO₂ foam studies that primarily relied on pressure drop and effluent analyses, this work introduces NMR imaging and T₂ spectrum diagnostics to directly visualize pore-scale fluid redistribution and quantify sweep efficiency within heterogeneous cores. The NMR data provide mechanistic evidence that the enhanced recovery originates from selective foam propagation and the mobilization of residual oil in low-permeability channels, rather than merely from increased flow resistance. This integration of advanced pore-scale imaging with macroscopic displacement analysis represents a mechanistic advancement over conventional CO₂ foam evaluations, offering new insights into the conformance control behavior of AOS-based foam systems in heterogeneous reservoirs. Full article

In promoting the “dual-carbon goals” and sustainable development strategy, analyzing the spatio-temporal response mechanism of landscape patterns to carbon emissions is a critical foundation for achieving carbon emission reductions. However, existing research primarily targets urbanized zones or individual ecosystem types, often overlooking how landscape pattern affects carbon emissions across entire watersheds. This research examines spatial–temporal characteristics of carbon emissions and landscape patterns in China’s Yellow River Basin, utilizing Kernel Density Estimation, Moran’s I, and landscape indices. The Geographically and Temporally Weighted Regression model is used to analyze the impact of landscape patterns and their spatial–temporal changes, and recommendations for sustainable low-carbon development planning are made accordingly. The findings indicate the following: (1) The overall carbon emissions show a spatial pattern of “low upstream, high midstream and medium downstream”, with obvious spatial clustering characteristics. (2) The degree of fragmentation in the upstream area decreases, and the aggregation and heterogeneity increase; the landscape fragmentation in the midstream area increases, the aggregation decreases, and the diversity increases; the landscape pattern in the downstream area is generally stable, and the diversity increases. (3) The number of patches, staggered adjacency index, separation index, connectivity index and modified Simpson’s evenness index are positively correlated with carbon emissions; landscape area, patch density, maximum number of patches, and average shape index are negatively correlated with carbon emissions; the distribution of areas positively or negatively correlated with average patch area is more balanced, while the spread index shows a nonlinear relationship. (4) The effects of landscape pattern indices on carbon emissions exhibit substantial spatial heterogeneity. For example, the negative impact of landscape area expands upstream, patch density maintains a strengthened negative effect downstream, and the diversity index shifts from negative to positive in the upper reaches but remains stable downstream. This study offers scientific foundation and data support for optimizing landscape patterns and promoting low-carbon sustainable development in the basin, aiding in the establishment of carbon reduction strategies. Full article

This study develops a coupled hydropower–ecological optimization model to balance energy production and ecosystem sustainability. The ecological objective is quantified by a composite Eco-Index, derived via Principal Component Analysis from seven key parameters of 32 Indicators of Hydrologic Alteration, enhancing representativeness while reducing computational complexity. Hydrological years are classified into wet, normal, and dry types using the Standardized Runoff Index and runoff quantiles, showing that wet years exhibit the strongest hydropower–ecology coupling, followed by normal and dry years. The optimized average annual hydropower revenues are 3.75 billion USD in wet years, 3.10 billion USD in normal years, and 2.70 billion USD in dry years, with average EI values being 0.35, 0.27 and 0.26, respectively. Spatial analysis identifies Xiaowan and Nuozhadu reservoirs as critical control points sensitive to hydrological variability. Moreover, optimization substantially enhances system resilience and reduces vulnerability. These results demonstrate that coordinated cascade reservoir operation can improve system robustness while signaling a caveat for careful trade-offs between economic and ecological objectives. Full article

Water demand has increased due to population growth and rapid socioeconomic development, creating conflicts between human activities and water resources and having a substantial impact on the balance between blue and green water supplies. Existing study lacks a spatial perspective to examine the inherent relationship between blue and green water supply and demand, particularly in terms of geographical differentiation characteristics and rational allocation of blue and green water supply–demand balance in inland river basins. Using the Taolai River Basin as a case study, this research uses the distributed hydrological model SWAT from a blue–green water resources viewpoint to simulate the spatiotemporal distribution features of blue and green water resources at the sub-basin scale from 2002 to 2021. The supply and demand balance relationship of blue and green water resources within the basin was investigated, an assessment index system for water resource security was developed, and the realizable potential of blue water resources was quantified using various indicators. The findings show that during the study period, the average annual green water resources in the Taolai River Basin were 1.95 times greater than blue water resources, making green water the most abundant component of regional water resources. Spatially, both blue and green water resources showed considerable latitudinal zonality, with a declining tendency from south to north and very consistent distribution patterns. Blue water resources showed high geographic variability, with a safety index more than one, suggesting that supply–demand imbalances were most concentrated in the upper and intermediate ranges of the irrigated region, as well as the desert zone, where safety levels were relatively low. In contrast, green water resources had a safety score ranging from 0.7 to 1.0, indicating great overall safety and negligible regional variability. During the research period, the average annual theoretical transferable blue water resources were 4.06 × 108 m³, based on cross-regional water resource allocation potential analysis. This reveals tremendous potential for enhancing regional water resource allocation, hence providing substantial support for effective water consumption within the Taolai River Basin and regional economic growth. In conclusion, the assessment method developed in this work provides a solid foundation for improving water resource allocation and sustainable management in river basins. It provides technical assistance in the construction of water network systems in inland river basins, which is critical in establishing reasonable water resource distribution across various areas within these basins. Full article

In the process of shield tunneling in clayey strata, the fine-grained clay mineral components in the soil easily adhere to the cutter plate. The clay adhering to the cutterhead and the soil compartment then solidifies and hardens, which results in the production of mud cake and clogging. At present, research on cutter plates in clayey ground is limited and has focused mostly on static tests or simplified models. There is a lack of in-depth studies on the effect of temperature on clay adhesion, which is crucial for understanding the clogging risks. In this study, we independently researched and developed a rotary adhesion tester to investigate the adhesion effect and adhesion force change in a shield cutter plate under the influence of different temperatures, water contents (ω), and clay types, revealing the change rule of the adhesion effect under the joint influence of the temperature and the consistency index (I_c). This study provides experimental evidence and an empirical model for assessing the clogging risk in shield tunneling through clay strata, offering valuable insights that support the efficient operation of earth pressure balance (EPB) shields. Full article

Balancing the development of urban road infrastructure (URI) with the pace of urbanization is crucial to supporting high-quality urban growth. This study constructed a comprehensive evaluation framework of URI and urbanization using data from 101 Chinese cities between 2002 and 2021. The spatio-temporal characteristics of URI and urbanization were assessed using the entropy weighting method and the relative development index (RDI). Key variables were identified through the obstacle degree model and further refined via relative importance analysis. To investigate the nonlinear interactions among the most influential factors, a random forest model was employed in combination with SHapley Additive exPlanations (SHAP). The results revealed three key findings: (1) both URI and urbanization levels exhibited overall upward trends during the study period, although notable disparities were observed across cities; (2) URI development generally outpaced urbanization, indicating a lack of synergy between the two systems; and (3) key determinants of this mismatch included road density, total road area, the number of streetlights per unit road length, resident population size, and educational human capital. By integrating multidimensional URI and urbanization metrics in a comprehensive evaluation framework, this study provides new insights into the spatial synergy mechanisms and supports the formulation of tier-specific urban planning strategies. Full article

The growing penetration of distributed photovoltaic (PV) systems presents significant operational challenges for power grids, driven by the scarcity of historical data and the high spatiotemporal variability of PV generation. To address these challenges, we propose Generative Reconstruction and Adaptive Identification via Latents (GRAIL), a unified, end-to-end framework that integrates generative modeling with adaptive clustering to discover latent structures and representative scenarios in PV datasets. GRAIL operates through a closed-loop mechanism where clustering feedback guides a cluster-aware data generation process, and the resulting generative augmentation strengthens partitioning in the latent space. Evaluated on a real-world, multi-site PV dataset with a high missing data rate of $45.4\%$, GRAIL consistently outperforms both classical clustering algorithms and deep embedding-based methods. Specifically, GRAIL achieves a Silhouette Score of 0.969, a Calinski–Harabasz index exceeding $4.132\times {10}^6$, and a Davies–Bouldin index of $0.042$, demonstrating superior intra-cluster compactness and inter-cluster separation. The framework also yields a normalized entropy of $0.994$, which indicates highly balanced partitioning. These results underscore that coupling data generation with clustering is a powerful strategy for expressive and robust structure learning in data-sparse environments. Notably, GRAIL achieves significant performance gains over the strongest deep learning baseline that lacks a generative component, securing the highest composite score among all evaluated methods. The framework is also computationally efficient. Its alternating optimization converges rapidly, and clustering and reconstruction metrics stabilize within approximately six iterations. Beyond quantitative performance, GRAIL produces physically interpretable clusters that correspond to distinct weather-driven regimes and capture cross-site dependencies. These clusters serve as compact and robust state descriptors, valuable for downstream applications such as PV forecasting, dispatch optimization, and intelligent energy management in modern power systems. Full article

As electric vehicles (EVs) continue to advance, the impact of their charging on the power grid is receiving increasing attention. This study evaluates the efficiency of EV charging piles in performing peak shaving and valley filling for power grids, a critical function for integrating Renewable Energy Sources (RESs). Utilising a high-resolution dataset of over 240,000 charging transactions in China, the research classifies charging volumes into “inputs” (charging during peak grid load periods) and “outputs” (charging during off-peak, low-price periods). The Vector Autoregression (VAR) model is used to analyse interrelationships between charging periods. The methodology employs a Slack-Based Measure (SBM) Data Envelopment Analysis (DEA) model to calculate overall efficiency, incorporating charging variance as an undesirable output. A Malmquist index is also used to analyse temporal changes between charging periods. Key findings indicate that efficiency varies significantly by charging pile type. Bus Stations (BS) and Expressway Service Districts (ESD) demonstrated the highest efficiency, often achieving optimal performance. In contrast, piles at Government Agencies (GA), Parks (P), and Shopping Malls (SM) showed lower efficiency and were identified as key targets for optimisation due to input redundancy and output shortfall. Scenario analysis revealed that increasing off-peak charging volume could significantly improve efficiency, particularly for Industrial Parks (IP) and Tourist Attractions (TA). The study concludes that a categorised approach to the deployment and management of charging infrastructure is essential to fully leverage electric vehicles for grid balancing and renewable energy integration. Full article

Background/Objectives: Preclinical chest computed tomography (CT) imaging in small animals is often limited by low resolution due to scan time and dose constraints, which hinders accurate detection of subtle lesions. Traditional super-resolution (SR) metrics, such as peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM), may not adequately reflect clinical interpretability. We aimed to evaluate whether deep learning-based SR models could enhance image quality and lesion detectability in rat chest CT, balancing quantitative metrics with radiologist assessment. Methods: We retrospectively analyzed 222 chest CT scans acquired from polyhexamethylene guanidine phosphate (PHMG-p) exposure studies in Sprague Dawley rats. Three SR models were implemented and compared: single-image SR (SinSR), segmentation-guided SinSR with lung cropping (SinSR3), and omni-super-resolution (OmniSR). Models were trained on rat CT data and evaluated using PSNR and SSIM. Two board-certified thoracic radiologists independently performed blinded evaluations of lesion margin clarity, nodule detectability, image noise, artifacts, and overall image quality. Results: SinSR1 achieved the highest PSNR (33.64 ± 1.30 dB), while SinSR3 showed the highest SSIM (0.72 ± 0.08). Despite lower PSNR (29.21 ± 1.46 dB), OmniSR received the highest radiologist ratings for lesion margin clarity, nodule detectability, and overall image quality (mean score 4.32 ± 0.41, κ = 0.74). Reader assessments diverged from PSNR and SSIM, highlighting the limited correlation between conventional metrics and clinical interpretability. Conclusions: Deep learning-based SR improved visualization of rat chest CT images, with OmniSR providing the most clinically interpretable results despite modest numerical scores. These findings underscore the need for reader-centered evaluation when applying SR techniques to preclinical imaging. Full article

The increasing use of AI systems for face, object, action, scene, and emotion recognition raises significant privacy risks, particularly when processing Personally Identifiable Information (PII). Current privacy-preserving methods lack adaptability to users’ preferences and contextual requirements, and obfuscate user faces uniformly. This research proposes a user-centric, context-aware, and ontology-driven privacy protection framework that dynamically adjusts privacy decisions based on user-defined preferences, entity sensitivity, and contextual information. The framework integrates state-of-the-art recognition models for recognising faces, objects, scenes, actions, and emotions in real time on data acquired from vision sensors (e.g., cameras). Privacy decisions are directed by a contextual ontology based in Contextual Integrity theory, which classifies entities into private, semi-private, or public categories. Adaptive privacy levels are enforced through obfuscation techniques and a multi-level privacy model that supports user-defined red lines (e.g., “always hide logos”). The framework also proposes a Re-Identifiability Index (RII) using soft biometric features such as gait, hairstyle, clothing, skin tone, age, and gender, to mitigate identity leakage and to support fallback protection when face recognition fails. The experimental evaluation relied on sensor-captured datasets, which replicate real-world image sensors such as surveillance cameras. User studies confirmed that the framework was effective, with over 85.2% of participants rating the obfuscation operations as highly effective, and the other 14.8% stating that obfuscation was adequately effective. Amongst these, 71.4% considered the balance between privacy protection and usability very satisfactory and 28% found it satisfactory. GPU acceleration was deployed to enable real-time performance of these models by reducing frame processing time from 1200 ms (CPU) to 198 ms. This ontology-driven framework employs user-defined red lines, contextual reasoning, and dual metrics (RII/IVI) to dynamically balance privacy protection with scene intelligibility. Unlike current anonymisation methods, the framework provides a real-time, user-centric, and GDPR-compliant method that operationalises privacy-by-design while preserving scene intelligibility. These features make the framework appropriate to a variety of real-world applications including healthcare, surveillance, and social media. Full article

This paper introduces the Gesture-Based Physical Stability Classification and Rehabilitation System (GPSCRS), a low-cost, non-invasive solution for evaluating physical stability using an Arduino microcontroller and the DFRobot Gesture and Touch sensor. The system quantifies movement smoothness, consistency, and speed by analyzing “up” and “down” hand gestures over a fixed period, generating a Physical Stability Index (PSI) as a single metric to represent an individual’s stability. The system focuses on a temporal analysis of gesture patterns while incorporating placeholders for speed scores to demonstrate its potential for a comprehensive stability assessment. The performance of various machine learning and deep learning models for gesture-based classification is evaluated, with neural network architectures such as Transformer, CNN, and KAN achieving perfect scores in recall, accuracy, precision, and F1-score. Traditional machine learning models such as XGBoost show strong results, offering a balance between computational efficiency and accuracy. The choice of model depends on specific application requirements, including real-time constraints and available resources. The preliminary experimental results indicate that the proposed GPSCRS can effectively detect changes in stability under real-time conditions, highlighting its potential for use in remote health monitoring, fall prevention, and rehabilitation scenarios. By providing a quantitative measure of stability, the system enables early risk identification and supports tailored interventions for improved mobility and quality of life. Full article

This study presents a comprehensive assessment of the energy systems in eight Eastern European countries—Bulgaria, Croatia, the Czech Republic, Hungary, Poland, Romania, Slovakia, and Slovenia—focusing on their energy transition, security of supply, decarbonisation, and energy efficiency. Using principal component analysis (PCA) and clustering techniques, we identify three different energy profiles: countries dependent on fossil fuels (e.g., Poland, Bulgaria), countries with a balanced mix of nuclear and fossil fuels (e.g., the Czech Republic, Slovakia, Hungary), and countries focusing mainly on renewables (e.g., Slovenia, Croatia). The sectoral analysis shows that industry and transport are the main drivers of energy consumption and CO₂ emissions, and the challenges and policy priorities of decarbonisation are determined. Regression modelling shows that dependence on fossil fuels strongly influences the use of renewable energy and electricity consumption patterns, while national differences in per capita electricity consumption are influenced by socio-economic and political factors that go beyond the energy structure. The Decarbonisation Level Index (DLI) indicator shows that Bulgaria and the Czech Republic achieve a high degree of self-sufficiency in domestic energy, while Hungary and Slovakia are the most dependent on imports. A typology based on energy intensity and import dependency categorises Romania as resilient, several countries as balanced, and Hungary, Slovakia, and Croatia as vulnerable. The projected investments up to 2030 indicate an annual increase in clean energy production of around 123–138 TWh through the expansion of nuclear energy, the development of renewable energy, the phasing out of coal, and the improvement of energy efficiency, which could reduce CO₂ emissions across the region by around 119–143 million tons per year. The policy recommendations emphasise the accelerated phase-out of coal, supported by just transition measures, the use of nuclear energy as a stable backbone, the expansion of renewables and energy storage, and a focus on the electrification of transport and industry. The study emphasises the significant influence of European Union (EU) policies—such as the “Clean Energy for All Europeans” and “Fit for 55” packages—on the design of national strategies through regulatory frameworks, financing, and market mechanisms. This analysis provides important insights into the heterogeneity of Eastern European energy systems and supports the design of customised, coordinated policy measures to achieve a sustainable, secure, and climate-resilient energy transition in the region. Full article

Background/Objectives: Paspalum nicorae Parodi is a native subtropical grass species with promising agronomic attributes, such as persistence, drought and cold tolerance, and rapid establishment. However, the species remains underutilized in breeding programs due to the absence of well-characterized germplasm and limited studies on its genetic variability and agronomic potential. This study aimed to estimate genetic parameters, predict genotypic values, and identify superior ecotypes with desirable forage traits, integrating stability and adaptability analyses. Methods: A total of 84 ecotypes were evaluated over three consecutive years for twelve morphological and forage-related traits. Genetic parameters, genotypic values, and selection gains were estimated using mixed models (REML/BLUP). Stability was assessed through harmonic means of genotypic performance, and the multi-trait genotype–ideotype distance index (MGIDI) was applied to identify ecotypes with balanced performance across traits. Results: Substantial genetic variability was detected for most traits, particularly those related to biomass accumulation, such as total dry matter, the number of tillers, fresh matter, and leaf dry matter. These traits exhibited medium to high heritability and strong potential for selection. Ecotype N3.10 consistently showed superior performance across productivity traits while other ecotypes, such as N4.14 and N1.09, stood out for quality-related attributes and cold tolerance, respectively. The application of the MGIDI index enabled the identification of 17 ecotypes with balanced multi-trait performance, supporting the simultaneous selection for productivity, quality, and adaptability. Comparisons with P. notatum suggest that P. nicorae harbors competitive genetic potential, despite its lower level of domestication. Conclusions: The integration of REML/BLUP analyses, stability parameters, and ideotype-based multi-trait selection provided a robust framework for identifying elite P. nicorae ecotypes. These findings reinforce the strategic importance of this species as a valuable genetic resource for the development of adapted and productive forage cultivars in subtropical environments. Full article