Search Results (33,570)

With the increasing complexity of power grid operations, operator training requires timely feedback and objective assessment. Traditional approaches based on lectures and scripted simulations provide limited personalization and weak explainability. This paper presents AI Instructors, an intelligent simulation training system for power-grid control and dispatching. The system is organized into learning, training, assessment, and analysis modules, and is built around two core technical components: (i) parameterized item generation from rule/knowledge bases using a phrase-enhanced transformer (PET), and (ii) solver-grounded, topology-aware grading with hierarchical feedback for both numeric and free-text responses. A voice interaction module is integrated to simulate telephone-based dispatch orders. We validate the system through a pilot deployment with licensed dispatch operators and scenario experiments on benchmark cases. Compared with a conventional scripted DTS workflow, AI Instructors achieves higher stepwise procedure accuracy (68%→90%), a lower topology-violation rate (32%→11%), and shorter response time (120 s→72 s), while increasing the proportion of parameterized questions and accelerating skill acquisition. These results suggest that combining adaptive sequencing with topology-safe, explainable evaluation can improve training effectiveness and operational safety. Full article

This study quantifies the effect of air and fuel filter restriction on fuel consumption, regulated pollutants (CO and HC), and CO₂ greenhouse gas emissions under real driving conditions in a hilly high-altitude environment. Four filter configurations were evaluated: clean air filter–clean fuel filter (CAF–CFF, reference), dirty air filter–clean fuel filter (DAF–CFF), clean air filter–dirty fuel filter (CAF–DFF), and dirty air filter–dirty fuel filter (DAF–DFF). Each test was repeated three times over the same RDE route in Quito (≈2100–2900 m). Fuel consumption was estimated from ECU-based signals, and CO₂ emission factors and regulated pollutant (CO and HC) emission factors were computed from measured exhaust concentrations and distance normalization. Results were analyzed by RDE section (urban, rural, motorway) and expressed as percent changes relative to the reference configuration to directly isolate filter restriction effects. Relative to CAF–CFF, DAF–CFF produced the largest increase in average fuel consumption (+7.2%) and the largest urban CO₂ penalty (+22.7%), indicating a strong efficiency sensitivity to intake restriction under transient operation. CAF–DFF increased average fuel consumption by 6% and produced the strongest motorway penalties for CO (+77.3%) and HC (+44.4%), suggesting that fuel delivery restriction has a stronger influence on incomplete oxidation products under sustained higher load. The combined restriction (DAF–DFF) showed non-additive responses depending on the operating regime. Random Forest models were trained to estimate CO₂, CO, and HC, achieving R² values of 0.8571, 0.8229, and 0.7690, respectively, while multiple linear regression achieved an R² of 0.852 for fuel consumption. The proposed approach supports data-driven monitoring of filter restriction effects under real driving operation, while acknowledging that fuel consumption and CO₂ are obtained through different measurement and conversion paths and may not yield identical percent changes. Full article

Background: Age-related cognitive decline is linked to reduced gait complexity and higher fall risk. Traditional linear gait measures may miss subtle motor-cognitive deficits in older adults with dementia. This study examined whether an 8-week motor-cognitive exercise program could improve gait adaptability in institutionalized older adults with cognitive impairment. Gait complexity, measured using Sample Entropy, was the primary outcome. Methods: Forty-two institutionalized older adults completed follow-up assessments, including 26 with cognitive impairment and 16 controls. Gait was assessed during normal walking (single-task) and while performing cognitive tasks (dual-task), such as naming animals or counting backward. Inertial sensors recorded stride intervals, and Sample Entropy was calculated to evaluate gait regularity and adaptability, (gait complexity). The intervention included 24 structured sessions combining physical and cognitive exercises targeting balance, coordination, and executive function. Non-parametric tests (Wilcoxon) were used, with Bonferroni correction for multiple comparisons. Results: Participants with cognitive impairment showed increased gait complexity, especially during dual-task walking. Significant improvements were found in both limbs under dual-task conditions (left: p = 0.015, effect size = 0.34; right: p = 0.030, effect size = 0.31). During single-task walking, a significant improvement was observed in the left limb (p = 0.006, effect size = 0.39). Conclusions: Motor-cognitive exercise may enhance non-linear gait complexity in institutionalized older adults with cognitive impairment. The use of dual-task training in rehabilitation and highlight the value of entropy-based gait assessment for detecting subtle functional changes. However, the lack of a randomized non-exercising cognitive impairment control group limits definitive conclusions about causality. Full article

Background: This systematic review examined the relationship between horizontal multi-step jumps and sprint performance, and whether training interventions including these exercises improve sprinting. Methods: A systematic literature search was conducted in SPORTDiscus and PubMed (MEDLINE) and included English-language studies of athletes aged ≥14–15 years that assessed at least one horizontal multi-step jump and reported sprint outcomes over distances up to 100 m. Methodological quality and risk of bias were assessed using design-appropriate critical appraisal tools. Of 316 records identified, 19 studies met the inclusion criteria (10 intervention studies and 9 correlational studies). Results: Across correlational studies, horizontal multi-step jump performance showed associations ranging from weak to very large with sprint performance, with the strongest relationships typically observed during acceleration (≤20–30 m). In trained sprinters, correlations were often large to very large (r ≈ −0.65 to −0.88), whereas team-sport athletes showed more moderate associations, and younger or less specialized populations showed weak or non-significant relationships. Across intervention studies, horizontal multi-step jump training generally improved short-distance sprint performance, with the largest improvements reported for acceleration (up to ~7–12% in some studies), while effects at longer sprint distances and maximal-speed performance were smaller, inconsistent, or not different from comparison training. Conclusions: Overall, the evidence suggests that the association between horizontal multi-step jumps and sprint performance is strongest during the acceleration phase and is influenced by athlete population and training status. Horizontal multi-step jumps appear to be useful for assessing and potentially developing sprint acceleration. However, the findings should be interpreted with caution due to heterogeneity in study design and variable methodological quality, and associations with maximal sprint speed are less consistent across studies. Full article

To support farmers in their transition towards sustainable agriculture, sub-Saharan Africa needs a more effective extension. Thus, effective improvements based on a clear view of current and desired extension capacity are necessary. As in the past, mostly one-sided studies have been conducted. This paper proposes a more integral approach based on both characteristics and viewpoints of both farmers and extension workers. Capacity to provide effective extension and advisory services (EAS), or extension capacity, is defined and analyzed with mixed-research methods using data from 471 Ugandan dairy farmers, from three distinct production systems and 67 extension workers. Extension capacity is determined by farmers’ satisfaction, the frequency of delivering EAS to farmers, and the perceptions of both farmers and extension workers on the use of appropriate methods to deliver EAS. Results revealed moderate satisfaction across production systems, with a pronounced negative effect of long working experiences on the frequency of delivery. Positively influencing factors for delivery frequency are intrinsic motivation and the number of in-service trainings received by extension workers. On-farm demonstrations, individual farm visits, the use of contact farmers, and farmer training are perceived as the four most effective delivery methods among dairy farmers in Uganda. Given the moderate farmer satisfaction, low frequency of delivery, and slight mismatch between the perceived effective delivery methods and those being used, the study concludes that the current extension capacity remains low. However, low-hanging fruits for improvement include increasing in-service training opportunities, employing extension workers on contractual basis and motivating extension workers. Full article

Background/Objectives: This study proposes a comprehensive classification framework for the automatic detection of Alzheimer’s disease using handwriting data. An enriched feature space is constructed by combining 18 baseline features extracted from raw handwriting signals with 30 additional features derived from established handwriting analysis studies, resulting in a total of 48 features. To enhance clinical practicality, a task reduction analysis is conducted by comparing the full dataset containing 25 handwriting tasks with a reduced dataset comprising 14 selected tasks. Methods: The proposed framework employs a two-stage evaluation strategy involving four feature selection methods (Random Forest Feature Importance, Extreme Gradient Boosting Feature Importance, L1 Regularization and Recursive Feature Elimination), three normalization techniques (Unnormalized, Min–Max and Z-Score), and five baseline machine learning classifiers (Random Forest, Logistic Regression, Multilayer Perceptron, XGBoost and Support Vector Machines). In the second stage, a dynamic ensemble learning strategy is introduced, where the most effective classifiers are adaptively selected for each cross-validation fold and integrated using soft and hard voting schemes. Results: The experimental results demonstrate that reducing the number of tasks leads to an improvement in average classification accuracy from 79.47% to 81.03%, while simultaneously decreasing training time and memory consumption by approximately 40% and 35%, respectively. The highest classification performance, achieving an accuracy of 94.20%, is obtained using the Hard Ensemble combined with L1-based feature selection. Conclusions: These findings highlight that the joint use of enriched feature representations, task reduction, and dynamic ensemble learning provides an effective and computationally efficient solution for handwriting-based Alzheimer’s disease detection. Full article

Background/Objectives: The primary outcome of this 8-week randomized, controlled, parallel trial was to assess longitudinal shifts in gut microbiota structure and predicted metabolic potential in 45 elite football players following protein supplementation. Methods: Participants combined resistance training with daily intake (30 g) of whey protein concentrate (WPC), pea protein isolate (PPI), rice protein isolate (RPI), or a plant-protein blend (MIX). For the acquisition of prokaryotic metataxonomic data, the V3–V8 region of the 16S rRNA gene was sequenced using Oxford Nanopore Technology (ONT). Functional potential was inferred through the MACADAM database and STAMP software. Strict dietary monitoring and gravimetric adherence checks were performed to isolate the intervention effect. Results: While microbial alpha-diversity indices (Chao1, Shannon, Simpson) remained stable across all groups, significant source-specific shifts in taxonomic structure and predicted metabolic activity were identified. Whey protein concentrate (WPC) was associated with an increase in Bacteroidetes abundance and greater balance within the microbial community structure, whereas pea protein isolate (PPI) and the MIX correlated with reduced fermentative bacteria and elevated taxa potentially involved in cadaverine biosynthesis. Rice protein isolate (RPI) supplementation was associated with a higher predicted representation of taxa involved in succinate-to-butyrate fermentation pathways. These functional markers and differential responses of selected bacterial groups to particular protein types were observed. Conclusions: The data indicate complex interactions between supplement type, exposure duration, and microbiome response, underscoring the necessity for individualized dietary recommendations and supplementation strategies to optimize gut health and training adaptation in professional football players. Full article

Pediatric gait abnormalities are closely intertwined with musculoskeletal dysfunctions and heightened injury risk, underscoring the urgency of early and accessible screening tools. Here, we develop and validate a video-based semi-supervised Abnormal Gait Recognition Module (AGRM) to address unmet needs in pediatric gait assessment, with a focus on diagnostic performance and clinical interpretability. The AGRM is built on a 3D ResNet backbone, synergistically integrated with a Mean Teacher Module (MTM) to mitigate the limitations of limited labeled clinical data, and a Spatial Hierarchical Pooling Module (SHPM) for robust multiscale spatiotemporal feature extraction—two core innovations tailored to gait dynamics. We trained and validated the model on a hybrid dataset combining self-collected pediatric gait videos and the public CASIA-B dataset, evaluating its performance in binary (normal vs. abnormal) and three-class (normal, genu varum, genu valgum) classification tasks using accuracy, macro-precision, macro-recall, and macro-F1 score. Ablation studies quantified the incremental contributions of MTM and SHPM, while Grad-CAM visualization was employed to enhance model interpretability. In the three-class classification task, the AGRM achieved a 70.5% accuracy, 72.1% macro-precision, 71.5% macro-recall, and a macro-F1 score of 0.718; in the binary task, it yielded a 80.3% precision and 79.2% recall. SHPM significantly augmented spatiotemporal feature aggregation, capturing fine-grained gait dynamics, whereas MTM improved model generalization under constrained labeled data scenarios—findings corroborated by ablation experiments. Grad-CAM visualization confirmed the model’s targeted attention to lower extremity regions, particularly the knee joints, aligning with the pathological loci of gait abnormalities. Collectively, our AGRM demonstrates robust performance and generalization in identifying pediatric gait abnormalities, while effectively capturing key pathological gait characteristics. This video-based intelligent approach offers a promising tool for early gait screening in both clinical and community settings, addressing barriers to accessible pediatric musculoskeletal assessment. Full article

The Integrated Fixed-film Activated Sludge (IFAS) partial nitritation/anammox (PN/A) process offers robust nitrogen removal, yet startup using pre-colonized carriers incurs high logistical costs. This study investigated the mechanism of inoculating a pilot-scale IFAS system with granular anammox sludge to treat anaerobic digestion supernatant. The treatment train integrated coagulation, pre-aeration, and an IFAS-PN/A unit. The granular-inoculated IFAS-PN/A unit achieved stable biofilm formation and a nitrogen removal rate of 0.36 kg N m⁻³ d⁻¹, benefiting from the effective interception of excessive organic carbon by the preceding coagulation and pre-aeration steps. Microbial analysis identified Candidatus brocadia as the dominant anammox genus, revealing a distinct migration pathway: bacteria transferred from disintegrating granules to the suspended sludge—acting as a transitional vector—before ultimately colonizing the carriers. While granular biomass diminished, anammox abundance in the biofilm increased to 12.0% by day 166. Furthermore, distinct spatial niches were observed: ammonium-oxidizing bacteria (AOB) dominated the suspended sludge, while nitrite-oxidizing bacteria (NOB) were effectively suppressed. These findings demonstrate the feasibility of granular inoculation for cost-effective IFAS startup and provide critical insights into the bacterial migration dynamics required for stable operation. Full article

Teacher training should promote a social understanding of disability to support effective inclusive practices and reduce barriers. This study surveyed 150 pre-service primary and secondary teachers at one German university using a mixed-methods design to examine their attitudes toward inclusive education and their agreement with the social model of disability. It was found that participants held neutral to slightly positive attitudes toward inclusion and partially agreed with the social model of disability. Structural equation modeling revealed that social contact, even if limited, influenced the agreement with the social model of disability, but not the attitudes toward the inclusion of pupils with special educational needs in mainstream schools (CMIN/df = 1.50, CFI = 0.96, IFI = 0.96, TLI = 0.94, RMSEA = 0.06). Interview data showed that participants have concerns about working with these pupils, citing a lack of training in special education and inclusive practices. Full article

Doxorubicin (DOX) is a highly effective chemotherapeutic agent whose clinical use is limited by dose-dependent cardiotoxicity associated with oxidative stress, inflammation, and cellular stress responses. Here, we investigated whether preventive aerobic training could protect against DOX-induced cardiac injury in Wistar rats. Animals were assigned to sedentary control (C), sedentary DOX (D), trained control (CT), and trained DOX (DT) groups. The moderate-intensity (~50–80% maximal exercise test) treadmill protocol (40 min/day, 4 days/week for 4 weeks) was performed before intraperitoneal administration of DOX (4 mg/kg, weekly for 4 weeks) or saline. Preventive training markedly improved exercise capacity (p < 0.001) and attenuated oxidative damage, maintaining antioxidant enzyme activity (GR, SOD) at control levels (p > 0.05). DOX significantly upregulated cardiac IL-6 and IL-1β expression (p < 0.01), while trained animals preserved IL-1β expression similar to controls (p > 0.99). In parallel, DOX increased cardiac HIF-1 expression (p < 0.05), indicating activation of hypoxia- and stress-related signaling pathways, an effect that was attenuated by preventive training (p > 0.99). DOX-induced cardiac atrophy was evidenced by reduced left ventricular mass (p < 0.001), which was partially prevented by training (p < 0.05). Although hematological toxicity persisted, preventive aerobic exercise effectively counteracted DOX cardiotoxicity by restoring redox homeostasis, dampening inflammation, and limiting apoptotic signaling. Collectively, these findings highlight exercise preconditioning as a promising non-pharmacological strategy in cardio-oncology to mitigate chemotherapy-associated cardiac injury. Full article

Pedestrian and cyclist safety at urban intersections remains a critical challenge for transportation agencies, as vulnerable road users are significantly exposed to crash risks in complex traffic environments. Identifying high-risk locations and factors that contribute to crashes is essential for improving road safety. This study developed an explainable machine learning framework to predict motor vehicle-involved pedestrian and cyclist crash occurrence at urban intersections using five years of crash, geometric, operational, and socioeconomic data from a large set of urban intersections. Five supervised machine learning algorithms were trained and evaluated, including Binary Logistic Regression, K-Nearest Neighbors, Support Vector Machine, Decision Tree, and Random Forest. The evaluated models demonstrated strong predictive performance overall, with accuracies approaching 91% and high discriminative capability. In particular, the Binary Logistic Regression and Random Forest models achieved the highest area under the receiver operating characteristic curve (AUC) values of 0.961 and 0.964, respectively. To enhance transparency, SHAP values were used to quantify the contribution of predictors and examine feature effects at both the global and local levels. The results indicate that roadway hierarchy, intersection markings, and total entering volume are among the most influential determinants of crash likelihood, while socioeconomic variables exhibit weaker but interpretable effects. Full article

Understanding the coupling between hydrological dynamics and carbon sequestration is critical for predicting hemiboreal forest resilience to climate extremes. This study investigates water–carbon interactions in the Järvselja forest (Estonia) through a multi-objective hybrid modeling framework. We integrated long-term (2014–2025) eddy covariance flux measurements and daily meteorological data with a coupled architecture combining the process-based GR4J-Cemaneige model and a Long Short-Term Memory (LSTM) network. To validate the physical consistency of the deep learning component, we employed Support Vector Regression (SVR) diagnostic probes to map LSTM internal cell states against ERA5 soil moisture reanalysis data and in situ water table measurements. The combined LSTM + GR4J-Cemaneige model outperformed standalone approaches in the calibrated Reola catchment (NSE = 0.887), so by assuming hydrological similarity the hybrid model was regionalized to the streamflow ungauged Kalli basin. An in silico interpretability probe validated that the LSTM implicitly encoded physically meaningful soil moisture dynamics ($r>0.9$) without explicit training data. The analysis revealed that the 2018 heatwave triggered a synchronous collapse in water availability and carbon uptake, shifting the ecosystem from a robust sink to a net source. A significant legacy effect was observed, with carbon sequestration capacity lagging behind hydrological recovery for two years. The results of this paper substantiate the influence of climate warming on hemiboreal forests, demonstrating its implications for soil hydrology and the availability of water to sustain photosynthesis. Full article

Accurate power prediction is essential for assessing wind turbine performance under real-world operating conditions and for supporting condition monitoring and maintenance planning using SCADA data. Most existing approaches rely directly on raw SCADA signals, which may limit their ability to capture complex spatiotemporal dependencies among operational variables. To address this limitation, this paper proposes a novel SCADA-driven power prediction framework that transforms selected SCADA variables into multi-channel grayscale images and leverages an optimized LeNet-5–LSTM hybrid neural network for active and reactive power prediction. First, the SCADA dataset is analyzed to identify the most influential variables affecting power output. Six key variables are then selected, segmented, and encoded as 2D grayscale images, enabling the model to learn richer feature representations compared to conventional raw SCADA data-based methods. The proposed network combines convolutional layers for spatial feature extraction from SCADA data-based grayscale images with LSTM layers to capture temporal dependencies. Model training incorporates a customized loss function that integrates both data-driven supervision and physics-based constraints. The model is trained using 70% of the image-based dataset, with five independent runs to ensure robustness and reproducibility, while the remaining 30% is used for testing. The proposed approach is validated using SCADA data from three real-world cases: (i) a 2 MW Siemens wind turbine in Poland, (ii) a Vestas V52 wind turbine in Ireland, and (iii) the La Haute Borne wind farm in France, consisting of four wind turbines. The results demonstrate that the SCADA-based image representation enables the proposed LeNet-5–LSTM model to effectively learn discriminative feature patterns and achieve accurate active and reactive power predictions across different turbine types and operating conditions. Full article

Accurate identification of hydrodynamic parameters is essential for high-fidelity modeling and control of unmanned underwater vehicles (UUVs). Compared with towing tank experiments and computational fluid dynamics simulations, system identification based on free-running trial data offers a cost-effective and scalable alternative. However, in real ocean environments, unmodeled lumped disturbances—such as shear currents, stratification-induced buoyancy variations, and wave-induced drift forces—strongly couple with the vehicle’s intrinsic dynamics. Conventional least-squares estimators and physics-informed neural networks tend to absorb environmental effects into the physical parameters, leading to physically inconsistent estimates. To address this challenge, this paper proposes a physics-guided hybrid network (PG-HyNet) with input-domain structural decoupling. The architecture explicitly separates the intrinsic rigid-body dynamics from spatially varying environmental disturbances by assigning dynamics-related states to a physics-constrained branch and position-dependent variables to a residual disturbance branch. A staged training strategy is introduced to stabilize identification and suppress parameter drift during optimization. The framework is validated using high-fidelity simulations incorporating shear currents, density stratification, and wave drift effects, as well as real-world lake trial data. The results demonstrate that PG-HyNet significantly improves robustness against disturbance-induced parameter compensation, enabling physically consistent hydrodynamic parameter recovery while accurately capturing spatially varying environmental disturbance effects. Full article