Search Results (1,251)

This study presents an intelligent framework for the analysis of multidimensional geomechanical states in underground mining systems based on numerical simulation and machine learning methods. A three-dimensional geomechanical model of the Zholymbet deposit was developed in the RS3 environment using the generalized Hoek–Brown failure criterion. Numerical simulations were performed for representative mining scenarios characterized by complex excavation interaction and stress redistribution. The modelling results were transformed into a multidimensional geomechanical dataset containing stress, deformation, displacement, and yielding parameters. Principal component analysis (PCA) was applied to investigate the internal structure of the geomechanical state space and identify dominant patterns controlling the rock mass behavior. Clustering analysis revealed several geomechanical regimes corresponding to stable, transitional, and instability-prone conditions. Isolation Forest anomaly detection demonstrated that atypical geomechanical states are not randomly distributed but spatially localized near excavation systems and mining horizons. The obtained results indicate that hazardous geomechanical conditions are governed by complex interactions between stress concentration, deformation intensity, yielding processes, and excavation geometry. The proposed approach provides a basis for intelligent interpretation of large-scale numerical modelling results and may support geomechanical risk assessment in underground mining operations. Full article

Unmanned aerial vehicles (UAVs) are increasingly used in precision pest management, yet their performance in operational forest settings remains underexplored. We evaluated the efficacy of SPLAT^® SM-O mating disruptant applied using a UAV at a dosage of 14.8 for control of the spongy moth, Lymantria dispar dispar L. (Lepidoptera: Erebidae). One treatment plot received 11.4 g AI/ha because of a calibration deviation during application. Both treatments reduced trap catches by >90% for 10 weeks following the application, meeting the efficacy requirement set by the USDA’s National Slow the Spread (STS) Program. One year after the application, trap catches continued to be reduced by 28% and 67% in plots treated with 14.8 and 11.4 g AI/ha, respectively. These levels of trap catch reduction in the year of treatment and one year after the treatment application are comparable to those reported following fixed-wing aerial treatments. These results indicate that UAV-applied SPLAT^® SM-O meets STS requirements for operational use and is suitable for integration into the program for treating small or isolated blocks. These findings also have broader implications for the use of unmanned aerial vehicles to deploy SPLAT^® formulations in forest pest management programs. Full article

Background: Preterm premature rupture of membranes (PPROM) before 32 weeks of gestation with prolonged latency is associated with substantial neonatal morbidity, including Dry Lung Syndrome (DLS), pulmonary hypoplasia (PH), bronchopulmonary dysplasia (BPD), and death. Accurate individualized risk stratification remains elusive, as the interacting contributions of amniotic fluid dynamics, inflammatory status, and microbiological burden are inadequately captured by traditional statistical approaches. Methods: We performed a retrospective, exploratory–predictive analysis of 66 pregnancies complicated by second-trimester PPROM with latency exceeding 14 days. Elastic Net and Random Forest models were trained across six clinically defined predictor domains using a multi-stage block modelling strategy. To address the clinically relevant distinction between antenatal and postnatal information, results are reported separately for Model A—comprising exclusively antenatal predictors available during expectant management (gestational age at PPROM, latency, amniotic fluid trajectory, inflammatory status, vaginal microbiome at admission)—and Model B, which additionally incorporates postnatal variables and characterizes the full mechanistic perinatal risk trajectory. Binary and ordinal outcomes included DLS, PH, BPD, intraventricular hemorrhage (IVH), and neonatal death. Pairwise interaction models were additionally computed to identify cross-domain risk constellations. Results: Distinct predictor architectures emerged per outcome. Pulmonary hypoplasia was most strongly associated with temporal features of oligohydramnios—particularly the persistence and timing of SDP < 1 cm—rather than isolated measurements. For DLS, the antenatal model (Model A) achieved AUC 0.776, driven by gestational maturity and inflammatory status; surfactant administration—a postnatal variable reflecting therapeutic response rather than an antenatal risk factor—dominated only the mechanistic Model B. Neonatal death was driven by a combined profile of respiratory support burden, amniotic fluid persistence, and co-morbidity. IVH showed consistently high ordinal predictability (accuracy 0.863), with amniotic fluid dynamics and microbiological burden as leading contributors. BPD remained the least linearly separable endpoint across all configurations. Conclusions: Multi-domain machine learning reveals outcome-specific, cross-domain risk architectures following second-trimester PPROM that are invisible to conventional statistical models. Longitudinal amniotic fluid trajectory is the dominant antenatal determinant of structural pulmonary morbidity, while microbiological burden independently shapes neurological risk. These findings support prospective validation of integrated ML-based risk stratification tools for individualized antenatal counselling in this high-risk population. Full article

With the widespread adoption of intelligent terminals, edge devices, and distributed information systems in the financial domain, financial security sensing data exhibit multisource heterogeneity, dynamic temporal patterns, and high privacy sensitivity. Traditional centralized anomaly detection methods are no longer able to simultaneously satisfy the requirements of cross-institutional or cross-node collaborative modeling, client data privacy protection, and robust monitoring of transaction and system anomalies. To address this challenge, a data-local federated deep anomaly detection framework has been proposed for distributed financial security sensing systems. Initially, a local deep financial security sensing representation module is constructed to perform temporal encoding and attention-based modeling on multisource financial signals, including terminal operation status, network transaction communication, backend server operation, identity authentication, and anomaly alerts, thereby extracting representations relevant to anomalous behaviors. Subsequently, a data-local federated optimization and personalized aggregation mechanism is developed to enable cross-node knowledge sharing without transmitting raw transaction or client data, while local personalized detection heads are employed to adapt to non-independent and identically distributed (non-IID) financial institution data. Furthermore, an adversarially robust security detection and trust-aware aggregation strategy is introduced to enhance model stability under input noise, feature masking, anomaly camouflage, and potential malicious client updates. Experimental results demonstrate that the proposed method achieves an Accuracy of $92.37\%$, a Precision of $89.41\%$, a Recall of $88.26\%$, an F1-score of $88.83\%$, an AUC of $93.06\%$, and a PR-AUC of $89.15\%$ in the primary financial anomaly detection task, significantly outperforming baseline methods such as Isolation Forest, Autoencoder, LSTM, Transformer, FedAvg, FedProx, SCAFFOLD, and MOON. In robustness experiments, the method attains F1-scores of $87.95\%$, $86.42\%$, $86.88\%$, $84.57\%$, $86.73\%$, and $83.91\%$ under Gaussian noise, feature masking, temporal shift, adversarial perturbation, and $20\%$ and $30\%$ malicious client scenarios, respectively. Ablation studies further confirm the effectiveness of local representation learning, personalized federated optimization, adversarial training, and trust-aware aggregation mechanisms. Overall, the proposed approach provides an efficient intelligent anomaly detection solution for financial AI security monitoring scenarios characterized by data localization requirements, node heterogeneity, and attack perturbations. Full article

Background/Objectives: The trajectory of influenza hospitalization burden from pre-COVID-19 pandemic baseline through post-pandemic recovery remains poorly characterized at the national level. This study characterized phase-stratified burden and seasonal structure, quantified racial and ethnic disparities, and assessed whether post-pandemic seasons represent anomalous departures from pre-pandemic expectations. Methods: Sixteen complete seasons of FluSurv-NET surveillance data (2009–2010 through 2024–2025; 509 observation weeks) were analyzed across pre-pandemic, disruption, and recovery phases using OLS regression with effect-size estimation, bootstrapped age-adjusted rate ratios, seasonal-trend decomposition (STL), Prophet time-series forecasting, and Isolation Forest anomaly detection. Results: Mean peak weekly hospitalization rate nearly doubled from pre-pandemic to recovery (5.1 to 11.1 per 100,000), cumulative seasonal burden increased from 46.3 to 87.0 per 100,000, and median peak timing advanced from MMWR week 9 to week 50. STL decomposition revealed a marked shift from weak pre-pandemic seasonality (Fs = 0.14) to substantially stronger annual regularity (Fs = 0.98) across three recovery seasons, with threefold amplitude increase. Non-Hispanic Black persons had rate ratios of 1.72, 2.16, and 1.99 relative to White persons across phases; American Indian and Alaska Native persons showed the highest disruption-phase ratio (2.24, 95% CI 1.90–3.53), based on two contributing seasons. A flat-growth Prophet model detected first exceedance in February 2020, outperforming a linear-growth specification on held-out validation. Isolation Forest identified 2017–2018, 2023–2024, and 2024–2025 as robust anomalies across all contamination thresholds. Conclusions: Post-COVID-19 pandemic influenza recovery is characterized by intensified and restructured seasonality, persistent racial and ethnic disparities, and anomalous burden exceeding pre-pandemic projections, identified independently by time-series forecasting and unsupervised anomaly detection. Full article

Anomaly detection models trained exclusively on healthy data are widely used in wind turbine condition monitoring because failure data are scarce, heterogeneous, and often unavailable. However, these models produce anomaly indicators that are sensitive not only to fault-related degradation but also to normal operational variability, transient disturbances, and changes in loading conditions. As a result, the practical behavior of an alarm system depends not only on the anomaly detection model but also on the decision rule used to activate and maintain alarm states. This study presents a decision-oriented evaluation of persistence-based alarm confirmation in wind turbine anomaly detection. Four representative techniques are analyzed within a unified framework: Isolation Forest, One-Class Support Vector Machine, Referenced Moving Window Principal Component Analysis using Q-statistic and percentage component weight indicators, and Autoencoder-based reconstruction error. The evaluation combines controlled OpenFAST simulations of rotor unbalance under different severity and noise conditions with an industrial SCADA case study involving a documented main bearing fault. Results show that temporal persistence strongly shapes alarm outcomes across methods and datasets. Low persistence values favor early detection but promote alarms from isolated threshold exceedances, whereas moderate persistence substantially reduces false positives while preserving detection capability in severe and well-observable faults. Excessive persistence increases detection latency and missed detections, particularly for weak, intermittent, or slowly evolving fault signatures. These findings indicate that persistence-based alarm confirmation should be treated as an explicit decision-level configuration variable, rather than as a fixed post-processing or alarm-state heuristic, when designing anomaly detection systems for wind turbine condition monitoring. Full article

Soil erosion (SE) is a constant, complex land degradation process, a common natural disaster that occurs all over the world and severely impacts soil fertility, food security, and environmental balance. Soil erosion depends on many factors, including soil properties, slope, vegetation, rainfall amount and intensity, and anthropogenic activities. There are two main natural erosive forces by which soil is eroded and transported—water and wind. Water erosion refers to the detachment, transportation, and deposition of soil particles (solid runoff) into river networks. These particles, varying in size and composition, are the main products of soil erosion and most strongly affect river ecosystems. Solid runoff, or sediment-laden runoff, affects water quality, destroying habitats, carrying pollutants, reducing reservoir storage, and causing flooding. Erosion control activities also influence river ecosystems in different ways. Hydrotechnical facilities, a major erosion control practice, can alter the composition of aquatic biota by disrupting longitudinal connectivity and isolating populations. Reforestation and afforestation are other erosion control practices that have a strong impact on ecosystems. Stormwater retention systems in urban and forest areas are also important measures addressed in this review. This review examines complex environmental interactions and the roles of erosion and erosion control activities in river ecosystems. During the research, several key points were established: erosion and erosion control activities significantly affect river ecosystems. There is a lack of quantitative analysis of erosion intensity and its influence on ecosystems. This is probably due to the exceptional complexity and diversity of river ecosystems, but such a study would provide important information about complex relationships in nature. Full article

The Gaizi River Basin, an alpine region in China crossed by the Karakoram Highway, is highly prone to glacier-related debris flows (GDF). Accurate debris flow susceptibility assessment in this high-altitude area remains challenging due to complex terrain, active tectonics, and dynamic glacial processes. This study develops a hybrid model integrating statistical methods and machine learning-based anomaly detection for debris flow susceptibility mapping. To address data noise, certainty factor (CF) distributions of debris flow predisposing factors (DFPFs) were derived via Locally Weighted Scatterplot Smoothing (LOWESS). The strength of the association between DFPFs and GDF susceptibility was evaluated using the mean residual between the raw and LOWESS-smoothed CF values. Multiple anomaly detection algorithms, including distance-based (L2 Norm), density-based (One-Class SVM), ensemble (Isolation Forest, RandNet), and GAN-based (WBiGAN-GP) methods, were tested on raw and CF-transformed data, using only the GDF inventory as the label. The CF-WBiGAN-GP model delivers the most balanced performance, excelling at identifying both high- and low-susceptibility zones. Results show that distance to stream, slope, and the topographic roughness and wetness indices are strongly associated with GDF susceptibility. Distance to glacier and precipitation appear less informative for direct susceptibility inference under our specific dataset and analytical setup. Full article

The prevention and control of forest fires are of vital importance for ecological security. The efficiency and environmental friendliness of fire-extinguishing agents remain the core focus of current research. This paper reviews the research progress and fire extinguishing mechanisms of three types of forest-fire-extinguishing agents, namely, foam extinguishing agents, gel extinguishing agents, and fire-resistant barrier materials. These three types of extinguishing agents work together to extinguish fires through three principles: isolating combustibles, reducing the oxygen concentration, and lowering the temperature. This paper systematically summarizes the performance evaluation methods, covering the cooling rate, fire extinguishing time, and re-ignition rate, and combines numerical simulation and field experiments to build a multi-scale verification system. The environmental assessment focuses on biodegradability, the ecological toxicity to soil and water systems, and the impact on plant germination and biodiversity. It clearly indicates that degradability, low toxicity, and low residue are key development directions. The current research still needs to further deepen in aspects such as long-term stability, adaptability to complex terrains, and ecological risk assessment during the life cycle. In the future, priority should be given to promoting green, multi-functional, and precise application technologies to provide solid support for scientific forest fire prevention and ecological protection. Full article

This paper proposes a data-driven framework for simulating turbocharger (TC) failure scenarios and modelling specific fuel oil consumption (SFOC) degradation in two-stroke low-speed marine diesel engines. A GaussianCopula model was fitted to the joint distribution of fifteen variables, using approximately eleven months of operational sensor data (n = 480 clean records, 4 h interval, January–December 2014) taken from a container ship. Three physically motivated failure scenarios were produced: turbine blade fouling, bearing wear and compressor surge. Predictive models trained on the real dataset achieved R² = 0.9998 for TC RPM and R² = 0.984 for fuel flow when using Gradient Boosting with 5-fold cross-validation. Feature importance analysis showed that the dominant determinants of TC speed were scavenging air intake pressure (35.3%) and engine power (MCR, 31.3%). Shaft power (45.5%) and TC RPM (19.3%) together explained most of the fuel consumption variance. Simulated failure scenarios produced SFOC increases of +6.6% (fouling), +9.6% (surge), and +13.3% (bearing wear) when compared to a normal operating baseline of 202 g/kWh, which is in line with published empirical data from MAN B&W engine performance curves. An IsolationForest anomaly detector trained only on normal operating samples flagged failure scenario records at a rate of 17.5–23.7%, which demonstrates that moderate-sensitivity early warning detection is feasible from routine sensor streams. The results show that TC condition monitoring could serve as a leading indicator of fuel-efficiency degradation. This has significant implications for condition-based maintenance planning and CII (Carbon Intensity Indicator) compliance. Full article

Bioprospecting for extremozymes from unique ecological niches is crucial for developing robust biocatalysts for green chemistry. Here, we report the de novo hybrid genome assembly of Penicillium crustosum CTM10622, isolated from the humid montane forest of El Feïdja National Park, Tunisia. Using Illumina NextSeq™ 500 and Nanopore PromethION 2 Solo, a highly contiguous 31.38 Mb assembly (N50 = 1.94 Mb; 98.3% BUSCOs) was achieved. This robust genomic foundation enabled the identification of an extensive hydrolase repertoire, leading to the discovery of a novel alkaline serine lipase, PCLIP, subsequently heterologously expressed in Pichia pastoris. Recombinant rPCLIP exhibited a high specific activity (15,000 U/mg at pH 10, 65 °C) and exceptional thermostability, with half-lives of 14 and 8 h at 80 and 90 °C, respectively. The enzyme’s identity as a serine lipase was confirmed by its complete inhibition by Orlistat or tetrahydrolipstatin (THL) (51 µM), PMSF (5 mM), and diisopropylfluorophosphate (DIFP) (2 mM). To determine its substrate specificity, advanced computational approaches, including convolutional neural network-based docking and explicitly solvated molecular dynamics, were employed to compare rPCLIP with its homologue PCrL, a recombinant serine alkaline lipase from Penicillium crustosum Thom P22. While rPCLIP showed optimal experimental activity toward short-chain glyceryl tributyrate, simulations revealed that long-chain trioctanoin acts as a ‘thermodynamic trap’ due to over-stabilization. Conversely, the rigid rPCrL favors tricaprylin, driven by a ‘hydrophobic engine’ effect where the solvated environment forces chain burial with minimal entropic penalty. The findings demonstrate that rPCLIP specificity is driven by a delicate interplay of geometric complementarity, Van der Waals enthalpy, and conformational entropy. Full article

Objectives: When a Candida species is identified in an ICU patient, susceptibility results are typically available in 24–72 h. In this study, we built a machine learning model using four variables available at identification to estimate resistance probability in real time. Methods: We analysed 747 fungal isolates from 725 ICU patients (January 2021–March 2026). We trained and compared a Random Forest and a Logistic Regression model, evaluating both with temporal cross-validation, permutation feature importance, three-category (S/I/R) prediction, and calibration analysis. Results: Multidrug resistance doubled from 24.5% (2021) to 51.1% (2025), and Candida auris grew eight-fold in three years. Random Forest reached AUC 0.885 on the held-out test set and 0.848 on prospective 2024–2025 data (Brier score 0.093). Species identity and drug choice together explained 87% of predictive signal. Local C. albicans fluconazole resistance (~16%) far exceeded the ECMM European figure of 0%, and C. krusei was four times more prevalent than the continental average. Conclusions: A four-variable model may provide calibrated resistance estimates during the critical gap before susceptibility results return, though performance reflects predominantly deterministic species–drug patterns rather than complex learned biology. Overall performance was comparable to a rule-based lookup table, confirming that the majority of predictive signal derives from established species–drug susceptibility patterns. Meaningful added value is limited to temporal trend tracking and improved prediction where resistance is acquired rather than intrinsic (C. albicans, C. tropicalis hard-subset AUC 0.929 vs. rule-based 0.899). The model complements a local antifungal testing; it does not replace one. Full article

The Easterlin paradox and recent distributional reassessments suggest that average effects obscure how subjective disadvantage is generated and reproduced over time. We propose the Social Comparison System (SCS), a framework that represents subjective well-being (SWB) as an internal state and relative income rank as an external conditioning variable within a feedback structure, with three structural properties: threshold activation, state-dependent gain, and rank-conditioned attractor displacement. The properties are recovered through a sample-isolated three-stage framework integrating tree-based machine learning, forest-based heterogeneity estimation, panel-data estimation, and hierarchical Bayesian Markov modeling on a balanced four-wave panel of the China Family Panel Studies (CFPS; 8099 individuals; 32,396 person-wave observations). Stage 1 locates a discrete predictive discontinuity in relative income rank between rank 2 and rank 3 (SHAP jump = 0.383, permutation p < 0.001). Stage 2 carries this boundary into a disjoint validation panel and recovers a negative rank-by-prior-SWB interaction (β = −0.036) and a 2.30-fold larger conditional effect in low- than in high-prior-SWB strata. Stage 3 recovers a 22.6-percentage-point gap in the rank-conditioned occupancy of the lowest within-wave SWB quartile between low- and high-rank subsystems, which under a first-order Markov approximation corresponds to a long-run stationary gap, robust to alternative state-space discretizations. Throughout this paper, relative income rank is treated as a conditioning variable, and the rank-conditioned patterns are interpreted as associational; the long-run quantities are reported under a first-order dynamical approximation rather than as identified causal or fully validated long-run effects. Persistent subjective disadvantage is therefore characterized by unequal dynamics of activation, amplification, and escape, rather than by unequal resources alone. This reframing provides a methodological template for identifying rank-conditioned feedback structures in social-systems data. Full article

Massive Internet of Things (IoT) and sensor-network services in 5G/6G systems increasingly rely on network slicing to support large-scale sensing, monitoring, and mission-critical applications. In such sliced infrastructures, Proportional Fair (PF) allocation assigns resources according to slice-reported demands. This reliance on trusted demand reporting makes coexisting slices, including mMTC-based IoT sensor slices, vulnerable to resource exhaustion attacks, where a malicious slice inflates its demand to monopolize shared resources and induce Service Level Agreement (SLA) violations. Existing unsupervised defenses mainly focus on anomaly detection, while the translation of detection results into resource-level mitigation remains insufficiently addressed. To bridge this gap, this paper proposes AutoGuard-Hybrid, an unsupervised detection-to-mitigation framework that combines complementary anomaly detectors with allocation-aware mitigation policies to preserve slice-level service availability. Unlike prior detection-only approaches, AutoGuard-Hybrid converts unsupervised anomaly evidence into allocation-aware demand purification before PF scheduling. Its key design is a closed-loop integration of Isolation Forest (IF) and Long Short-Term Memory Autoencoder (LSTM-AE) as spatial and temporal front-end detectors with Adaptive Clipping and a Safety Cap, which translate anomaly scores into demand purification actions. Experiments show that AutoGuard-Hybrid remains comparable to Isolation Forest under Continuous attacks and improves the mean system-wide SLA violation rate by 27.6% under Adaptive Probing attacks. Stage activation analysis further shows that LSTM-AE activations increase from 9.3 under Continuous attacks to 29.4 under Adaptive Probing attacks. Ablation results show that Adaptive Clipping alone reduces the system-wide SLA violation rate by 75.0%, while the full mitigation pipeline achieves an 84.6% total reduction. AutoGuard-Hybrid operates within the 1 ms Transmission Time Interval (TTI) constraint and provides a practical defense framework for next-generation network slicing-enabled IoT and sensor-network services. Full article

In rocky mountainous regions characterized by shallow, barren soils and water scarcity, non-rainfall water, such as condensation, plays a crucial ecological role in mitigating seasonal drought in forest trees. To enhance the water-use capacity of vegetation, this study utilized a previously developed eco-friendly PVA–CS/SA–Ca²⁺ hydrogel. The primary objective was to elucidate the synergistic mechanisms by which the hydrogel optimizes condensed water utilization and drives the ecophysiological recovery of Pinus tabuliformis and Platycladus orientalis, two keystone afforestation species in northern China. Utilizing a controlled environmental chamber to simulate the condensation and humidification process, the experiment established three treatments: a control group (CK), a pot-sealed group (PS, to isolate soil water absorption), and a hydrogel-amended group (Hydrogel-Root Wrapping, HRW). To comprehensively evaluate the water utilization mechanisms, the amount of condensed water captured by the system was quantified, and hydrogen isotope tracing techniques were employed to precisely track water transport pathways and contribution rates. Concurrently, key physiological parameters were systematically determined, including leaf water potential, stomatal conductance, leaf water content, net photosynthetic rate, and transpiration rate. The results demonstrated the following: (1) the hydrogel significantly enhanced the condensation water capture capacity of the system. The net mass gains of the Pinus tabuliformis and Platycladus orientalis systems under the HRW treatment reached 26.3 g and 32.9 g, respectively, which represented 1.17 and 1.30 times those of the CK treatment, and 1.52 and 1.54 times those of the PS treatment. (2) Isotope tracing confirmed that both tree species possess significant Foliar Water Uptake (FWU) capacity. Following condensation, the δ²H values in the leaves of Platycladus orientalis and Pinus tabuliformis surged to 113.5‰ and 85.3‰, respectively, with stem δ²H values increasing by 31‰ and 22‰ compared to their initial baseline. (3) The introduction of the hydrogel in the HRW treatment provided 11.2% and 10.9% of the stem water supply for Platycladus orientalis and Pinus tabuliformis, respectively, thereby reducing their dependence on soil water by 8.3% and 13.1%. In contrast, there was no significant difference in the fractional contribution of condensation water to stem water between the PS and CK treatments. (4) Regarding physiological responses, the application of the hydrogel material effectively improved the physiological status of the plants. The leaf water potentials of Pinus tabuliformis and Platycladus orientalis increased to −0.15 MPa and −1.32 MPa, respectively. Concurrently, stomatal conductance (3.25 and 3.64 mm·s⁻¹) and leaf water content (58.4% and 67.4%) were significantly higher than those in the other treatments. In summary, the hydrogel can significantly enhance the capture, conversion, and utilization efficiency of condensation water by vegetation, effectively optimizing the water supply dynamics of the system. This provides key theoretical and technical support for ecological afforestation in difficult sites within rocky mountainous areas. Full article