Search Results (6,353)

Against the backdrop of global climate warming and rapid urbanization, urban thermal environments exhibit strong spatiotemporal heterogeneity and diurnal contrasts. Based on the high-resolution, seamless land surface temperature dataset (GSHTD), this study systematically evaluates the evolution of extreme urban thermal environments across 107 cities in the Yangtze River Basin (YRB) from 2001 to 2020. Urban cores were delineated using high-density impervious surface area (ISA ≥ 50%), and rural background temperatures were elevation-corrected. To quantify the asynchrony between extreme heat intensification and seasonal background warming, we propose “Risk Amplification Index (Ri)”. The results reveal that: (1) The surface urban heat island intensity (SUHII) intensified across the entire basin, with daytime increases being significantly stronger and more spatially consistent than nighttime ones. (2) The intra-annual SUHII cycle exhibits a unimodal pattern peaking in August, with widening inter-city disparities during the warm season. (3) The intensification of extreme heat is often asynchronous with background warming. Combined with land-use change intensity (ΔISA), our analysis indicates that small and medium-sized cities undergoing rapid expansion (high ΔISA) exhibit a stronger heat-risk amplification effect (higher Ri), whereas mature megacities (high total ISA but low ΔISA) show relatively synchronous thermal evolution. The results suggest that an ISA density of around 70% may act as a threshold beyond which extreme-heat amplification is more likely to intensify. These findings suggest that future heat-risk governance should be time- and region-specific, shifting the focus of climate-adaptive planning from solely megacities to mitigating extreme-heat risk amplification during the rapid urbanization of small and medium-sized cities. Full article

Precise slag addition monitoring in steel continuous casting is critical, yet harsh industrial environments make this task extremely challenging. This research proposes a novel deep learning framework by integrating BiFormer and coordinate-aware squeeze-and-excitation (CoordSE) modules into the YOLOv11 architecture. To efficiently extract features of small slag particles against complex molten steel backgrounds, the BiFormer component employs a dual-level routing attention strategy. Concurrently, the CoordSE module captures spatial and channel-wise feature dependencies by combining direction-aware feature aggregation with multi-branch fully connected layers. Evaluated on a custom dataset of 2847 high-resolution industrial images, the proposed BiFormer-CoordSEBlock-YOLOv11 model achieved 82.5 ± 0.2% precision, 69.1 ± 0.3% recall, and 80.6 ± 0.2% mAP@0.5. Comprehensive ablation studies confirm that the BiFormer and CoordSE modules improved the baseline mAP@0.5 by 23.4% and 12.3%, respectively. Operating at a real-time inference speed of 45.2 FPS on standard hardware, this model offers a highly competitive framework for metallurgical process monitoring. However, the current recall rate of 69.1% and the lack of physical validation on resource-constrained edge devices represent limitations that must be systematically addressed before full-scale industrial deployment. Full article

In the semi-arid Batna Belezma region of northeastern Algeria, groundwater is a vital resource for agriculture and drinking water. However, the climate leads to intense evaporation, which affects its quality. This study aims to identify the key hydrogeochemical processes that control groundwater composition in the Merouana Basin and to evaluate the predictive performance of machine learning (ML) models. A total of 30 groundwater samples were analyzed using multivariate statistical techniques, including Principal Component Analysis (PCA), and were modeled using PHREEQC to assess mineral saturation states. Additionally, ML-based regression models, including K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Random Forest (RF), and Extreme Gradient Boosting (XGB),were employed to predict groundwater chemistry. The results indicate that the dominant ion distribution follows the following trend: Ca²⁺ > Mg²⁺ > Na⁺ and HCO₃⁻ > SO₄²⁻ > Cl⁻. Alkaline earth metals (Ca²⁺ and Mg²⁺) constitute the major fraction of total dissolved cations, reflecting carbonate equilibrium and dolomite dissolution processes. In contrast, Na⁺ represents a smaller proportion of the cationic load; however, its hydro-agronomic significance is substantial due to its influence on sodium adsorption ratio (SAR) and soil permeability. The PHREEQC modeling showed that calcite and dolomite precipitation promote evaporite dissolution, while most samples remain undersaturated with respect to gypsum. The PCA results reveal high positive loadings of Mg²⁺, Cl⁻, SO₄²⁻, HCO₃⁻, and EC, suggesting that ion exchange and seawater mixing are the primary controlling processes, with carbonate weathering playing a secondary role. To enhance predictive assessment, several supervised machine learning models were tested. Among them, the Random Forest model achieved the highest predictive performance (R² = 0.96) with low RMSE and MAE values, confirming its robustness and reliability. The results indicate that silicate weathering and mineral dissolution are the primary mechanisms governing groundwater chemistry. The integration of multivariate statistics and machine learning provides a comprehensive understanding of groundwater evolution and offers a reliable predictive framework for sustainable water resource management in semi-arid environments. Geochemical model performance showed a high global accuracy (GPI = 0.91), confirming a strong agreement between observed and simulated chemical data. However, the HH value (0.81) indicates some discrepancies, particularly for specific ions or extreme conditions. Full article

Atmospheric corrosion of carbon and low-alloy steels causes direct economic losses that are estimated at around 3.4% of the global GDP, and its accurate multi-year prediction is essential for protective coating selection, service-life estimation, and infrastructure maintenance scheduling. In this study, machine learning (ML) algorithms, including gradient boosting regressor (GBR), eXtreme gradient boosting (XGBoost), random forest (RF), support vector regression (SVR), and ridge regression, were trained on a 600-sample physics-grounded dataset to predict the cumulative atmospheric corrosion loss (µm) of low-alloy steels over 1–10 years of exposure. The dataset was constructed using the exact ISO 9223:2012 dose–response function (DRF) for a first-year corrosion rate and the ISO 9224:2012 power-law multi-year kinetic model (C(t) = C₁·t0.5), spanning ISO 9223 corrosivity categories C2–CX across 11 environmental and material input features. All models were evaluated on the original (untransformed) corrosion scale under an 80/20 train/test split and five-fold cross-validation. Gradient boosting achieved the best overall performance with test set R² = 0.968, CV-R² = 0.969, RMSE = 10.58 µm, MAE = 5.99 µm, and MAPE = 12.6%. XGBoost was a close second (R² = 0.958, CV-R² = 0.960). RF achieved an R² of 0.944. SHAP (SHapley Additive exPlanations) analysis identified SO₂ deposition rate, exposure time, relative humidity, Cl⁻ deposition rate, and temperature as the five most influential predictors. The dominance of the SO₂ deposition rate (mean |SHAP| = 26.37 µm) and the high second-place ranking of exposure time (13.67 µm) are fully consistent with the ISO 9223:2012 dose–response function and ISO 9224:2012 power-law kinetics, respectively, while among the material features, Cu and Cr contents showed the strongest negative SHAP contributions, confirming their corrosion-inhibiting roles in weathering steels. These results establish a physics-consistent, interpretable ML benchmark exceeding R² = 0.90 for multi-year cumulative corrosion loss prediction and provide a quantitative tool for alloy screening, coating selection in aggressive atmospheric environments, and service-life planning. Full article

For electric vehicles, accurate real-time estimation of the road friction coefficient is critical for maintaining stability, as the millisecond-level response of electric motors and the integration of regenerative braking demand higher perception fidelity than traditional internal combustion vehicles. This paper proposes a methodological framework for road friction estimation specifically designed for intelligent E-Chassis based on micro-signal features of intelligent tires and deep learning. An intelligent tire system, integrated with tri-axial accelerometers and strain gauges, was installed on the front-left wheel of a test vehicle to capture raw dynamic signals during transitions from cement to snow-covered surfaces across a velocity gradient of 10–50 km/h. The Savitzky–Golay convolutional smoothing algorithm was applied to reconstruct the high-frequency raw signals, enabling the extraction of a five-dimensional feature vector comprising vehicle velocity, peak strain, contact patch width, peak-to-peak acceleration, and signal standard deviation. The study revealed a natural filtering effect originating from the porous elastic properties of snow, resulting in a 60–70% reduction in signal standard deviation compared to cement, accompanied by a cliff-like feature collapse at the moment of snow entry. A BP neural network model with a 5-7-1 architecture achieved an identification accuracy of 96.2% on the test set, facilitating a rapid real-time prediction of the friction coefficient transitioning from 0.75 to 0.23. Unlike traditional methods, the proposed approach does not rely on high slip ratios and can complete identification within the first physical rotation cycle. This provides a robust physical criterion for the torque vectoring and regenerative braking stability of intelligent electric vehicles in extreme environments. Full article

Early-age shrinkage is a critical factor governing the dimensional stability and cracking susceptibility of ultra-high-performance concrete (UHPC). However, accurate prediction of UHPC shrinkage remains challenging due to the strong nonlinear interactions among mixture parameters, curing conditions, and hydration-induced internal moisture evolution, particularly when only limited experimental data are available. In this study, a systematic experimental program was conducted to investigate the influence of the binder-to-sand ratio, water-to-binder ratio, polypropylene fiber dosage, and curing environment on both early drying shrinkage and autogenous shrinkage of UHPC. Based on the experimental results, a structured dataset covering all shrinkage test data was constructed to support data-driven modeling. To improve prediction reliability under small-sample conditions, a Bayesian-optimized Extreme Gradient Boosting (BO-XGBoost) framework was developed and benchmarked against several conventional machine learning models, including Backpropagation Neural Networks (BPNNs), Random Forest (RF), and Support Vector Machines (SVMs). Shrinkage test data from other literature validated the prediction accuracy of this model, demonstrating its rationality and practicality. In addition, the Shapley Additive Explanations (SHAP) method was employed to quantitatively interpret the contribution and interaction mechanisms of key variables affecting shrinkage behavior. The results show that the BO-XGBoost model achieves the highest prediction accuracy and stability among the evaluated algorithms. SHAP analysis further reveals that curing age and curing environment dominate drying shrinkage, whereas autogenous shrinkage is primarily governed by the curing age and water-to-binder ratio. The interaction analysis also identifies the coupled effects between low water-to-binder ratio and extended curing age. The proposed framework not only improves prediction robustness for UHPC shrinkage under limited data conditions but also provides interpretable insights into the mechanisms governing early-age deformation. These findings offer a data-driven basis for optimizing UHPC mixture design and mitigating early-age cracking risks in engineering applications. Full article

In insurance portfolios, classifying customers without a prior history at a given company is particularly challenging due to the absence of historical behavior, extreme class imbalance, heavy-tailed loss distributions, and strict operational constraints. Traditional machine learning approaches, including the baseline methodology proposed in previous studies, typically optimize global predictive accuracy and therefore fail to capture business-critical outcomes, especially the identification of high-risk clients. This study extends the existing approach by evaluating two complementary business-aware classification strategies: (i) a balanced bagging ensemble specifically designed to handle class imbalance and maximize expected profit under explicit customer-omission constraints, and (ii) a lightweight Transformer-based architecture capable of learning richer feature representations. Both approaches incorporate the asymmetric financial cost structure of insurance and operate under operational selection limits. The empirical analysis is conducted on a proprietary large-scale auto insurance dataset comprising 51,618 customers and is complemented by validation on nine synthetic datasets to assess robustness. Model performance is evaluated using statistical tests (ANOVA, Friedman, and pair-wise comparisons) together with business-oriented metrics. The results show that both proposed approaches consistently outperform the baseline methodology (p < 0.001) in terms of profit, with the ensemble offering a better balance of performance and efficiency, while the Transformer shows stronger robustness and generalization under data perturbations. The balanced ensemble provides the most favourable trade-off between predictive performance, robustness, interpretability, and computational efficiency, making it suitable for deployment in regulated insurance environments, while the Transformer achieves competitive results and exhibits stronger generalization under data perturbations. The proposed approach aligns machine learning with actuarial portfolio optimization by explicitly integrating profit-driven objectives and operational constraints, offering two practical and scalable solutions for risk-based decision-making in real-world insurance settings. Full article

In low-latency edge-intelligence scenarios such as autonomous driving and industrial edge analytics, the processing of large-scale sensor data imposes extremely stringent requirements on communication latency. However, the high overhead of the traditional TCP protocol makes it difficult to satisfy such demands, while the semantic gap between the high-performance RoCE protocol and the standard Socket API prevents existing applications from directly exploiting its advantages. To address this problem, this paper proposes TransBridge, a lightweight user-space communication middleware that transparently bridges TCP and RoCE. Its design is realized through three key innovations: a transparent user-space compatibility architecture that enables unmodified Socket-based applications to benefit from RoCE performance; a microsecond-level low-latency transmission engine that bypasses kernel and protocol stack overhead; and a lightweight lock-free resource management mechanism based on a decentralized peer-to-peer architecture and deferred buffer updates. Experiments on a real RoCE network show that TransBridge significantly outperforms mainstream schemes: it achieves an average round-trip latency of 5.926 $\mathsf{\mu }$s for 16 B messages and a throughput of 20.254 Gbps for 16 KB messages; in the Fast DDS application-level evaluation, it achieves a throughput of 188 Mbps and an average round-trip latency of about 150 $\mathsf{\mu }$s. The results indicate that TransBridge can provide transparent and effective RoCE acceleration for existing Socket-based applications in resource-constrained edge environments. Full article

Tailings management presents a critical challenge for the mining industry, particularly in mountainous regions with high seismicity and steep slopes. This article presents the development and design criteria for dry stacking of filtered tailings as a sustainable and safe alternative to conventional slurry tailings storage facilities (TSFs). The study focuses on the extreme conditions of a mountainous location characterized by complex topography with 10% slopes, space constraints, and significant seismic activity defined by a peak ground acceleration (PGA) of 0.3 g. The design methodology, which incorporates layered compaction of the filtered tailings to achieve a geotechnically stable structure, is detailed for a filtered TSF consisting of 7 terraces, each 10 m high, reaching a total height of 70 m. This approach minimizes the risk of liquefaction and prepares the filtered tailings surface for progressive closure, with unit operating costs (OPEX) of 2.5 USD/t. The results of the physical stability analysis confirm the viability of this solution: pseudo-static stability analysis yielded a safety factor of 1.22, demonstrating a significant reduction in water consumption and potential environmental impact. It is concluded that the dry disposal of filtered tailings is a technically robust option for tailings management in extreme mountainous environments, offering greater long-term safety guarantees and facilitating landscape integration, thus setting a precedent for mining projects in similar geographies. Full article

To address the limitations of conventional metamaterials in thermo-mechanical coupling environments, this study proposes a multifunctional metamaterial structure through material selection and structural optimization, demonstrating stable vibration isolation performance under thermal fluctuations. The thermal deformation mechanisms and zero thermal expansion (ZTE) behavior of curved-beam unit cell are systematically examined through the chained beam constraint model (CBCM). A novel dual-zero metamaterial featuring both quasi-zero-stiffness (QZS) and ZTE characteristics is developed using curved-beam unit cell design. A parametric analysis, through finite element modeling, systematically investigated the effects of geometric parameters and material properties on the thermal expansion deformation and mechanical responses in the curved-beam unit cell structure. Furthermore, cylindrical metamaterials featuring dual-zero properties were engineered, and their deformation control mechanisms and vibration characteristic evolution across a broad temperature range were systematically studied. The simulation results indicate that while the Al–Al structure exhibits a significant resonance peak shift of up to 64.32% at 200 °C, the Al–Steel zero-stiffness design restricts this shift to only 7.72%. Furthermore, the Steel–Invar configuration demonstrates exceptional vibrational stability, with its center frequency shifting marginally from 5.58 Hz to 5.61 Hz at 200 °C. This methodology presents a viable solution for engineering metamaterials in extreme-temperature environments. Full article

Rainfall is a significant input for several engineering designs such as hydraulic structures, culverts, bridges and ducts, rainfall water sewer, and highway drainage system. The detailed statistical analysis of extreme daily rainfall of each arid environment’s region is essential to estimate the relevant input value for designing and analyzing engineering structures and agricultural planning. This paper aims to assess the best-fitting distribution to estimate the design of rainfall depth (XT) and maximum rainfall values for different return periods (2, 10, 25, 50, 100, and 150). This study used extreme daily rainfall historical data collected in period of 1970–2020, collected from four rainfall gauge stations nearby the Wadi Al-Aqiq that are selected for analysis; they are Al Faqir (J109), Umm Al Birak (J112), Madinah Munawara (M001), and Bir Al Mashi (M103). The methodology approved in this paper examined four frequency distributions, namely: GEV (Generalised Extreme Value), Gumbel, Weibull, and Pearson type III to identify the most suitable and extreme storm design depth corresponding to different return periods. The results demonstrate that GEV and Pearson Type 3 produce higher extremes values, while the Weibull method is commonly suggested in the HYFRAN-PLUS MODEL (DSS) for criterion suitability. The findings for the 100-year storm design demonstrate that extreme values generated by the Hyfran-Plus model are higher than the decision support system (DSS). All (DSS) comparative values are less than the maximum historical data from 1970–2020, except the Al Faqir station (DSS), which has a value of 79.6 mm that exceeds the historical maximum of 71 mm. This study will provide advantageous information about the study area for water resources planners, farmers, and urban engineers to assess water availability and create storage. Full article

Accurately characterizing the spatiotemporal variations in precipitation in China’s drylands is important for solving water scarcity in the region, guaranteeing security in the ecological environment, and conducting precise drought disaster management. To reduce the uncertainty in the existing precipitation products, we developed a two-stage machine-learning framework combining extreme gradient boosting (XGBoost) and random forest (RF) residual corrections. Based on the ground-based observation data from 1030 meteorological stations and numerous high-precision precipitation products (GPM IMERG Final V6, MSWEP V2, CMFD 2.0, TerraClimate), a monthly fused precipitation dataset (XGB-RF) for China’s drylands was produced during the 2001–2020 period at the 0.1° resolution. The validation results showed that the XGB-RF had a monthly Kling–Gupta Efficiency (KGE) of 0.941, and it improved 20.6–62.2% relatively with that of input individual products. For the dataset as a whole, we found very consistent, reliable performance in all seasons and topography, in particular in winter time and data-scarce western areas where individual products have large biases. More importantly, the XGB-RF was employed for drought monitoring based on the 1-month Standardized Precipitation Index that calculated the median KGE of 0.888, which made good drought trend tracking and drought features possible. Notably, the KGE for the mean drought intensity was 0.757, which was higher than that of independent original products. This study provides a high-resolution precipitation forcing dataset and demonstrates the effectiveness of two-stage machine learning strategies in enhancing hydroclimatic monitoring and drought risk assessment in arid and semi-arid regions. Full article

Ka-band carrier-frequency-agility (CFA) synthetic aperture radar (SAR) employs pulse-to-pulse random wide-range frequency hopping to enhance anti-interference capability. However, the random hopping disrupts the azimuth phase continuity, and the millimeter-wave wavelength of the Ka band makes the imaging quality extremely sensitive to motion errors. To address these challenges, this paper proposes an auto-focusing imaging framework and performs a performance analysis for Ka-band CFA SAR. First, a back-projection (BP)-based imaging model is derived to restore the coherent phase history from the hopped echoes. Second, to compensate for the residual phase errors inevitable in high-resolution millimeter-wave imaging, an auto-focusing framework is developed. This framework incorporates a dynamic sub-aperture strategy and an adaptive spectral notching mechanism to ensure precise phase error estimation in complex scattering environments. Furthermore, the imaging performance under different frequency-selection modes is analyzed to provide a guideline for the parameter selection of the Ka-band CFA SAR. Experiments with a vehicle-mounted Ka-band SAR system demonstrate that the proposed method achieves well-focused images with 5 cm resolution. Full article

Systematic comparisons of how plants with contrasting ecological strategies respond to extremely wide nutrient availability gradients remain limited. We investigated the physiological, photosynthetic, and growth adaptations of four plant species representing distinct ecological strategies: Triticum aestivum L. (C3 annual crop), Zea mays L. (C4 annual crop), Ipomoea aquatica Forssk. (C3 annual/perennial aquatic vegetable), and Canna glauca L. (C3 perennial wetland ornamental). Plants were grown hydroponically under nitrogen (N), phosphorus (P), and potassium (K) gradients ranging from 0% to 500% of standard Hoagland nutrient solution. The study results showed that all measured plant traits exhibited characteristic unimodal dose–response patterns. Optimal performance mostly occurred at 100–150% nutrient availability gradients. Severe inhibition or mortality occurred at extreme gradients. Simultaneously, different plant species displayed markedly varying response amplitudes and nutrient-specific sensitivities. Z. mays showed the highest nutrient use efficiency and broadest optimal ranges, particularly for N and K. C. glauca exhibited extraordinary N responsiveness (32-fold increase in photosynthetic rate) but narrow optimal ranges (e.g., 1.01 ± 0.15 μmol CO₂/(m²·s) at the 1% N treatment vs. 32.52 ± 3.33 μmol CO₂/(m²·s) at the 150% N treatment). I. aquatica showed pronounced P limitation with broad tolerance to supra-optimal N and K. T. aestivum displayed moderate responses with clear sensitivity to N limitation. Root–shoot ratios declined systematically with increasing nutrient availability across all plant species, following negative exponential functions. The results of data analyses revealed significant effects of N, P, and K availability on all the determined plant traits. Correlation analyses demonstrated tight coupling effects among physiological, photosynthetic, and growth traits, indicating integrated whole-plant responses to nutrient variations. These findings reveal that plant ecological strategy systematically modulates nutrient response patterns and provide a quantitative framework for species-specific nutrient management. This study provides a theoretical basis for precision fertilization of aquatic vegetables and wetland plants, and more broadly support species-specific nutrient management in controlled-environment agriculture. Full article

Portfolio optimization is one of the most difficult sequential decision problems, as uncertainty and the non-stationary nature of financial markets hinder the development of robust strategies. Reinforcement learning is an attractive framework for addressing this problem, as it allows agents to learn market-adaptive strategies through data-driven interactions. However, existing risk-neutral reinforcement learning solutions for portfolio management are oblivious to downside risk and are mainly concerned with maximizing returns. To address this limitation, this paper proposes a novel risk-sensitive reinforcement learning framework for risk-aware portfolio optimization based on a conditional value-at-risk-based learning objective that explicitly controls extreme loss events. It formulates the portfolio optimization problem as a Markov decision process and solves it using a linearized actor–critic architecture. It also develops theoretical results to analyze important aspects of the learning process, specifically proving that the convexity of the conditional value-at-risk-based formulation and convergence of learning hold under standard assumptions. The proposed algorithm is applied in a realistic investment setting using NIFTY 50 market data. Quantitative results from a rolling window backtesting methodology show that the proposed model achieves the best risk-adjusted portfolio performance, i.e., a Sharpe ratio (0.610), while significantly reducing tail risk, as measured by the conditional value-at-risk (−0.121) and maximum drawdown (−0.198), compared to classical strategies and risk-neutral reinforcement learning solutions. Overall, the results demonstrate that integrating coherent risk measures into reinforcement learning provides an effective approach for developing robust and risk-aware portfolio optimization strategies in dynamic financial environments. Full article