Search Results (185)

The increasing integration of generative artificial intelligence (AI) in English as a Foreign Language (EFL) education requires a clearer understanding of the factors associated with students’ adoption intentions. Guided by the Technology Acceptance Model (TAM), this explanatory sequential mixed-methods study examines how prompt engineering competence and motivational identity are associated with students’ behavioral intention (BI) to use generative AI tools, with perceived usability conceptualized as a second-order construct comprising perceived usefulness (PU) and perceived ease of use (PEU). Survey data from 470 undergraduate students were analyzed using structural equation modeling. The results indicate that both prompt engineering competence (PEC) and motivational identity (MI) are associated with the perceived usability of AI tools (PUAI), which in turn predicts BI. Mediation analysis yielded findings consistent with a full mediation pattern under the bootstrapped estimation procedure, suggesting that perceived usability mediated the relationships between prompt engineering competence, motivational identity, and behavioral intention. Follow-up interviews with 10 students provided explanatory insights, indicating that competence and motivation contribute to AI use primarily when they enhance perceptions of usefulness and ease of use. Overall, the study provides context-specific evidence extending TAM within AI-supported EFL learning and offers practical implications for fostering effective prompting skills and learner engagement. Full article

Recent advances have been achieved in intelligent fault diagnosis of rolling bearings. However, noise interference and data imbalance remain critical challenges for achieving accurate and reliable fault diagnosis in practical industrial applications. To address the limited robustness of conventional deep learning models under noisy conditions and their bias toward majority classes in imbalanced scenarios, this study proposes a robust bearing fault diagnosis method based on an adaptive weighted heterogeneous ensemble learning framework. The proposed method begins with continuous wavelet transform (CWT), which is employed to preprocess raw vibration signals and convert them into time–frequency images. Subsequently, a residual convolutional denoising autoencoder augmented by the convolutional block attention module is developed, namely CBAM-RCDAE. CBAM-RCDAE is capable of effectively reducing and eliminating noise interference in two-dimensional image data, thus enhancing fault diagnosis accuracy. Furthermore, a heterogeneous ensemble learning framework consisting of three base learners, including Swin Transformer, a multi-scale convolutional neural network, and BiLSTM, is developed to enhance generalization capability. An adaptive weight selection (AWS) strategy is introduced to adjust the weights and aggregate the outputs of the three base learners for final fault classification. The proposed method is extensively evaluated on the PU and CWRU bearing datasets. Experimental results demonstrate that, under the most challenging imbalanced conditions, the proposed method improves the G-mean metric by 5.89% and 4.95% compared with the state-of-the-art methods on the PU and CWRU datasets, respectively. In addition, the proposed method exhibits superior noise robustness, enabling reliable fault diagnosis performance across various noise levels. Full article

Bio-based alternatives to conventional rigid foams have proven to be good substitutes owing to their enhanced sustainability and competitive performance. However, because their manufacturing processes are complex and destructive testing is often impractical, this study investigates whether microstructural features can be correlated with mechanical properties in lignin-containing rigid polyurethane (PU) foams using machine learning approaches. Various types and percentages of lignin-based polyols were investigated as partial replacements for polyol, including LHO, DCA, DCA-D, LHO-O, Kraft lignin (KL), and LHO-MD, at polyol replacement levels ranging from 12.5% to 50%, together with a control formulation. Scanning electron microscopy (SEM) images and corresponding mechanical compression data were used to train a custom state-of-the-art dual-head convolutional neural network (CNN) targeting the specific prediction of density, specific compression modulus, specific yield stress, and specific compression strength. The CNN was optimized with a weighted multi-output loss function, achieving strong predictive performance with R² values ranging from 0.850 to 0.91 and correlation coefficients above 0.92, while maintaining mean absolute error percentages below ≈9%. This proves the trained network’s capability to predict and capture morphological features governing load-bearing responses. On the other hand, Grad-CAM visualization revealed that the network focused its predictions on physically meaningful microstructural regions such as cell walls and strut junctions, which confirms that the proposed network can be classified as an interpretable, non-destructive, and data-driven framework for predicting and understanding bio-based PU foams’ mechanical behavior, hence reducing the inconvenience caused by time-consuming manufacturing and destructive testing. Full article

Grid-forming inverters (GFMIs) are essential for future low-inertia power systems because they establish voltage and frequency rather than simply following the grid. However, compared with synchronous machines, they have limited overload and fault current capability, so providing reliable fault ride-through (FRT)/low-voltage ride-through (LVRT) behavior requires specially designed control strategies. This study proposes a multi-agent reinforcement learning inspired technique for the parameter selection of virtual synchronous generator (VSG)-controlled GFMIs through independent twin delayed deep deterministic policy gradient (TD3PG) agents, considering symmetrical and asymmetrical grid faults on the IEEE 13 bus network. The methodology utilizes power flow errors and voltage unbalance factors as key observational inputs within the MATLAB/Simulink^® 2023b environment. The policies are designed to modify the inertia and damping coefficients of the active power controller, as well as the proportional–integral gains of the reactive power controller to enhance stability in response to grid disturbances. The efficacy of this approach was evaluated against a conventional VSG control approach and VSG with virtual impedance and dynamic current saturation across multiple inverters with different power ratings. The proposed reinforcement learning assisted control embedded with current limiting consistently showed reductions in peak fault current of approximately 10–12% as well as reductions in active and reactive power settling times from around 3.50–5 s to about 1–1.50 s. In addition, it also limits fault currents to below 1.25 p.u. during disturbance intervals, thereby enabling continuous operation of inverters with a wide range of power ratings under both symmetrical and asymmetrical fault conditions. Full article

This study investigates the role of metacognitive awareness in shaping students’ adoption of artificial intelligence (AI) tools for learning within the framework of the Technology Acceptance Model (TAM). Data were collected from 236 students at the University of Ha’il, and structural equation modeling (SEM) was employed to examine the relationships among metacognitive awareness, perceived usefulness (PU), perceived ease of use (PEU), and behavioral intention (BI). The findings indicate that metacognitive awareness has significant positive effects on both PU and PEU. In turn, PU and PEU significantly influence students’ BI to use AI tools for learning, with PU emerging as the stronger predictor. However, metacognitive awareness does not have a significant direct effect on BI. These results suggest that students’ intention to use AI tools is primarily driven by their perceptions of usefulness and ease of use, while metacognitive awareness plays an important role in shaping these perceptions. The study extends the Technology Acceptance Model by positioning metacognitive awareness as an upstream cognitive factor that enhances students’ evaluation of AI technologies in educational contexts. The findings offer practical implications for higher education institutions, emphasizing the importance of developing students’ metacognitive skills to improve their engagement with AI tools. By fostering students’ ability to reflect on and regulate their learning processes, educators can enhance the perceived value and usability of AI technologies, thereby supporting their effective integration into learning environments. Full article

The accelerated deployment of electric vehicles requires planning tools able to quantify how much charging infrastructure can be integrated into distribution systems without violating operational constraints. This paper proposes a multi-scenario optimization framework for the siting and sizing of electric vehicle charging stations in radial distribution networks. The problem is formulated as a mixed-integer nonlinear programming model in which candidate-station slots, binary siting decisions, integer EV assignments, hourly power-flow constraints, voltage limits, thermal limits, charger power factor, and charging strategy are coordinated. The objective function combines hosting capacity maximization with active energy losses and voltage deviation terms through a scalarized formulation. Unmanaged and managed charging strategies are evaluated under weekday and weekend operating scenarios. Four adaptive population-based optimizers are analyzed under identical computational conditions: particle swarm optimization, a population-based genetic algorithm, JAYA, and the multi-verse optimizer. Monte Carlo random sampling is included separately as a non-adaptive baseline without memory or learning. The methodology is tested on a modified 33-bus distribution system using Colombian demand profiles and line-current limits. The campaign includes 720 cases and 7200 independent runs. In the 720-case stochastic campaign, the largest feasible solution serves 765 EVs, equivalent to 5.508 MW, with a minimum voltage of 0.9084 p.u. and a maximum loading of 99.83%. Statistical validation shows no significant Holm-adjusted pairwise differences among the adaptive algorithms in hosting capacity, while PSO provides the most robust feasibility behavior. Supplementary robustness analyses quantify the influence of candidate-site definition, objective scaling, voltage limits, base charging-power scale, and native-load growth. A complementary deterministic 69-bus assessment under a normalized branch-current envelope preserves the qualitative managed-versus-unmanaged trend, with feasible sequential allocations of 779 and 225 equivalent EV charging units, respectively. The proposed framework provides a reproducible basis for identifying robust EVCS locations, estimating hosting capacity, and quantifying tradeoffs among charging capacity, network losses, voltage performance, and computational effort. Full article

Industry 5.0 requires resilient, adaptive, and trustworthy manufacturing systems capable of maintaining reliable diagnostic performance across heterogeneous industrial environments. However, data-driven fault diagnosis models often experience substantial performance degradation when transferred across machines, operating conditions, and data acquisition platforms because of domain distribution shifts. This study proposes an explainable domain-adaptive CNN–Transformer framework for unsupervised bidirectional cross-domain bearing fault diagnosis using the Case Western Reserve University (CWRU) and Paderborn University (PU) datasets. The framework integrates one-dimensional convolutional layers for extracting local high-frequency vibration patterns, Transformer encoders for modelling long-range temporal dependencies, and Maximum Mean Discrepancy (MMD)-based feature-distribution alignment for learning transferable domain-invariant representations. Under the Unsupervised Domain Adaptation (UDA) protocol, the source domain supplies labelled samples for classification learning, whereas the target domain contributes unlabelled features only for MMD-based alignment; target labels are withheld from training and model selection and are used only for final evaluation. Conventional 1D-CNN and bidirectional long short-term memory baselines achieve over 90% accuracy in-domain but fall to 82.14% and 79.88%, respectively, for CWRU→PU, corresponding to domain-drop magnitudes of 12.07 and 12.99 percentage points. The proposed framework achieves 98.63% and 96.82% in-domain accuracy on CWRU and PU, respectively, and 92.46% for CWRU→PU and 94.18% for PU→CWRU, with domain-drop magnitudes of 6.17 and 2.64 percentage points. Ablation results confirm the complementary contributions of convolutional feature extraction, Transformer-based temporal modelling, and domain alignment. Furthermore, attention, saliency, and feature-importance analyses show that the model focuses on fault-relevant vibration regions and informative diagnostic characteristics, including kurtosis, root-mean-square (RMS), and crest factor, improving prediction transparency. These findings support accurate, transferable, and interpretable vibration-based condition monitoring across heterogeneous bearing datasets. Full article

Cross-domain distribution shifts severely degrade the diagnostic performance of rolling bearing models under unseen variable operating scenarios. Single-source domain generalization (SDG) builds fault diagnosis models using only single-source vibration data, which fits the practical limitations of industrial data collection. Existing contrastive learning methods adopt uniform spectral perturbations for data augmentation, which easily corrupt fault harmonic characteristics and require massive, labeled training samples. To tackle these drawbacks, this paper proposes an unsupervised contrastive learning framework named FDACL. An adaptive frequency-domain augmentation (AFA) module equipped with learnable weights is designed to separate fault-critical frequency bands from noise components. Differentiated amplitude perturbations are applied to two categories of spectral signals to generate diverse pseudo-samples while retaining intrinsic fault information. A shared encoder is trained with combined InfoNCE contrast loss and classification loss to learn domain-invariant fault representations. Validations are carried out on three datasets, namely Case Western Reserve University (CWRU), Paderborn University (PU), and the industrial CRRC Qingdao Sifang railway wheelset bearing dataset acquired from physical test benches. FDACL achieves average cross-speed diagnostic accuracies of 92.68% and 77.85% on CWRU and PU, respectively, and maintains competitive performance on the Qingdao Sifang industrial dataset. It outperforms state-of-the-art baselines by 4.23–8.71% across all SDG transfer tasks. Ablation experiments and hyperparameter analysis verify the efficacy of the AFA module and contrastive learning scheme, providing an unsupervised diagnostic approach for railway bearings under unknown working conditions. Full article

Deep exploration in mature gold districts requires subsurface alteration evidence that can be related quantitatively to three-dimensional (3D) geological architecture. This study develops a workflow for the Tudui–Shawang deposit in the Muping–Rushan metallogenic belt that integrates drill-core short-wave infrared (SWIR) spectroscopy, 3D alteration modeling, ore-controlling geological constraints, positive–unlabeled (PU) learning, and ensemble prospectivity prediction. A total of 2140 spectra from 10 drillholes were processed to identify mineral assemblages, extract spectral scalars and feature-shape attributes, classify alteration facies, and construct continuous 3D alteration evidence. Discrete smooth interpolation and indicator kriging were used for continuous and categorical attributes, respectively, and CatBoost, LightGBM, XGBoost, and Random Forest were evaluated within a spatially separated PU-bagging design. Quantitative analyses show that individual SWIR attributes have weak deposit-scale relationships with Au grade. Nevertheless, local IC minima, relatively lower pos2200 values near several mineralized intervals, alteration-facies transitions, and a broader shift toward longer pos2250 wavelengths characterize relevant parts of the mineralized system. FUSE performed best under 1 km × 1 km spatial holdout validation, with an ROC AUC of 0.8900 and a PRAUC of 0.8926. Prediction-area analysis and the 3D probability volume delineated three ranked exploration targets (T1–T3). The results show that drill-core SWIR-derived 3D alteration evidence, when integrated with ore-controlling geology and spatially validated machine learning, provides a practical basis for target prioritization in mature gold districts. Full article

Digital ad fraud continuously evolves through device spoofing, behavior simulation, and coordinated traffic attacks, making the detection of sophisticated fraudulent activities a persistent challenge. In real-world settings, only a small fraction of fraudulent traffic receives reliable labels, while vast amounts of unlabeled data may still contain latent fraud. This severely limits the generalization ability of conventional supervised learning models. To address this issue, we propose DiffVC—a novel contrastive learning framework tailored for ad fraud detection under the Positive-Unlabeled (PU) learning paradigm. DiffVC employs a multi-granularity diffusion augmentation strategy that builds upon a denoising diffusion probabilistic model to generate semantically augmented samples at three granularities: weak, medium, and strong. This strategy expands the latent fraud feature space while preserving diverse semantic information. We further introduce a diffusion-distance-calibrated similarity that dynamically adjusts constraints between augmented samples based on their diffusion distance, thereby improving the discrimination of unlabeled fraud. In addition, we design a dual-gated residual Transformer encoder that adaptively captures high-order feature interactions via gated residual connections and a channel recalibration mechanism. Experimental results on multiple real-world ad fraud detection datasets demonstrate that DiffVC consistently outperforms state-of-the-art methods and achieves stronger generalization. Our results confirm the effectiveness and practical applicability of DiffVC for ad fraud detection under label-scarce conditions. Full article

Roasting time strongly influences almond oil yield, flavor and oxidative stability. This study investigated almond oils obtained by pressing almond kernels roasted at 130 °C for 0, 5, 10 and 20 min, corresponding to cold-aroma, light-aroma, strong-aroma and sauce-aroma groups, respectively, by combining physicochemical analysis, fatty acid quantification, gas chromatography–ion mobility spectrometry (GC-IMS), machine learning and lipidomics. Increasing roasting time enhanced oil yield from 45.81% to 50.34%, with no further significant increase after 10 min. Stronger roasting moderately reduced the free radical level. In contrast, acid value, peroxide value and thiobarbituric acid reactive substances (TBARS) increased from 0.28 to 0.38 mg KOH/g oil, 1.20 to 1.46 meq O₂/kg oil and 0.38 to 0.46 mg malondialdehyde/kg oil, respectively, indicating limited oxidative deterioration. Fatty acid analysis showed that oleic acid remained relatively stable at 644.37–647.03 mg/g oil, whereas linoleic acid decreased significantly from 150.42 to 140.19 mg/g oil. GC-IMS detected 59 volatile signals and clearly separated the four processing groups, indicating substantial roasting-induced changes in volatile fingerprints. A positive–unlabeled (PU) learning model used 27 literature-supported aroma-active compounds as the positive class and 32 compounds without confirmed aroma-activity evidence as the unlabeled class. The model showed good cross-validated discrimination, with a mean area under the receiver operating characteristic curve (ROC-AUC) of 0.866 and a mean area under the precision–recall curve (PR-AUC) of 0.756, and prioritized eight unlabeled compounds, mainly alcohols and methional, as candidate aroma-active compounds requiring further sensory validation. Lipidomic analysis showed that accelerated oxidation caused broad lipid remodeling, with 327 differential lipids between cold-aroma oxidized and unoxidized oils, mainly involving glycerophospholipids, glycerol lipids, sphingolipids and fatty acyls. Overall, increasing roasting time improved oil recovery and promoted aroma differentiation while causing only limited initial oxidative deterioration, providing a theoretical basis for optimizing roasting conditions in industrial processing. Full article

To alleviate the severe voltage security and operational efficiency challenges brought about by the increasing penetration of distributed energy resources in active distribution networks, Volt/Var control (VVC) has become a key mechanism to stabilize node voltage and minimize power loss by coordinating reactive power injection. While multi-agent reinforcement learning (MARL) offers a promising decentralized control approach, its static reward functions are prone to creating harsh trade-offs between voltage constraint enforcement and cost-efficiency. In this paper, a hierarchical autonomous control framework featuring large language model-driven dynamic reward shaping (LLM-Driven DRS) is introduced to balance security and efficiency. The dynamic priority shifting (DPS) mechanism lies at the center of the framework and dynamically varies the reward weights through the detection of real-time grid bottlenecks. Under the LLM-Driven DRS framework, this mechanism successfully achieves a fluid transition between a Constraint-Dominant Phase for voltage stabilization and an Objective-Refinement Phase for economic optimization. Validation on a modified IEEE 33-bus system demonstrates that the proposed framework achieves Pareto superiority over conventional static weight strategies. Crucially, compared with the 1.250% static baseline, the absolute voltage violation rate is suppressed to 0.014%, mitigating long-tail risks of hardware degradation and inverter tripping, while active power losses are reduced by up to 33.76%. A robust safety margin is further confirmed by spatiotemporal analysis, which reveals an average minimum voltage margin increase of over 0.011 p.u. under severe stress conditions. Full article

Underground distribution networks increasingly rely on aging cable assets whose parameter drift modifies propagation velocity, attenuation, and fault-initiated transient signatures, thereby reducing the reliability of conventional traveling-wave (TW) fault location and purely data-driven diagnosis. This paper proposes a physics-regularized hybrid learning framework for joint fault-type classification, feeder-area identification, and continuous fault localization in aging underground distribution feeders. The methodology integrates (i) an aging-aware simulation pipeline driven by a normalized aging-stress index $\alpha \in [0,1]$ that perturbs the per-unit-length cable matrices within a controlled domain; (ii) synchronized multi-sensor time–frequency representations of three-phase voltage and current transients; (iii) an area-aware multi-task architecture with fault-type and area-classification heads and area-specific local regression heads; and (iv) a propagation-consistency loss that depends explicitly on the model-predicted fault position and therefore contributes gradients during training. A branched underground feeder is evaluated using five synchronized sensing locations and a stratified dataset of $N_{\mathrm{tot}}=\mathrm{36,000}$ simulated fault events covering 11 fault classes (SLG-A/B/C, LL-AB/BC/CA, DLG-ABG/BCG/CAG, LLL, and LLLG), six non-overlapping feeder areas, fault resistance $R_f\in [0.1,50]\Omega$, measurement noise $\mathrm{SNR}\in [20,40]\mathrm{dB}$, and a $20\mathrm{ms}$ transient window sampled at $200\mathrm{kHz}$. On a held-out test set of 7200 previously unseen event records drawn from the same simulation domain, the Hybrid model achieves a fault-type accuracy of $0.93$, an area-identification accuracy of $0.96$, a localization MAE of $0.011$ p.u., and a 95th-percentile absolute error of $0.027$ p.u. The proposed configuration outperforms the TW-TOA, purely data-driven Baseline, and physics-regularized PINN references across the reported diagnostic tasks within the prescribed simulator and parameter ranges. Time–frequency attribution is included only as a qualitative interpretability illustration and is not treated as quantitative evidence of explanation faithfulness. Accordingly, the results demonstrate comparative in-domain simulation performance rather than field or cross-simulator deployment readiness. Full article

In this study, we examine positive-unlabeled (PU) learning for classifying Polish texts intended for large language model training. We evaluate six PU methods, Naive, Clust, Strict-LassClust, Non-Strict-LassClust, LassoJoint, and Spy, under both classic and MVC non-SCAR labeling schemes. Across the benchmark, the best F1 reaches 0.906 on an open subtitles corpus under classic labeling and 0.900 under MVC labeling; the largest absolute gain over the Naive baseline is 0.811 F1 points on ulotki medyczne and 0.758 on plwiki in the classic setting. The experimental workflow supports multiple labeling rates, c calc, and repeated random seeds, and evaluates models with AUC, PR-AUC, and F1 as the main metrics. Our study focuses on six SpeakLeash datasets spanning different domains: plwiki, open subtitles corpus, job offers corpus, ISAP corpus, ulotki medyczne, and wolne lektury corpus. For each dataset, we construct a binary PU setup from the quality field, where HIGH is treated as positive and LOW is negative, while other values are removed. The resulting preprocessed tables retain only the selected linguistic and structural features, rescale numeric variables to the [0, 1] range, and remove the raw text column. Full article

Generative artificial intelligence (GenAI) is rapidly transforming teaching, learning, research, and institutional practices in higher education, increasing the need to understand how university professionals engage with AI technologies in both functional and ethical contexts. This study examined AI adoption among sampled Generation Y (Gen Y) and Generation Z (Gen Z) higher education professionals by investigating the relationships among AI Competency Capability (AICC), AI Utilisation Behaviour (AIUB), Perceived Usability (PU), Perceived Strategic Value (PSV), AI Ethical Awareness (AIEA), AI Self-Efficacy (AISE), Responsible AI Utilisation Behaviour (RAIUB), Academic Engagement (AE), Perceived Performance Outcomes (PPO), and AI Adoption Intention (AAI). Adaptive Structuration Theory (AST) and Dual-Process Theory (DPT) were employed as complementary interpretive perspectives rather than as theories directly tested by the structural model. Data were collected from 870 higher education professionals employed at Saudi Arabian universities and analysed using Partial Least Squares Structural Equation Modelling (PLS-SEM), Partial Least Squares Multi-Group Analysis (PLS-MGA), and the Measurement Invariance of Composite Models (MICOM) procedure. The PLS-MGA identified statistically significant between-group differences only for the relationships between AICC and AIUB, which were numerically larger among the sampled Gen Z participants, and between AIEA and Responsible AIUB, which were numerically larger among the sampled Gen Y participants. All remaining differences represented sample-specific numerical variations rather than statistically confirmed between-group differences. Given the cross-sectional design and differences in age, career stage, institutional role, and professional experience, the findings should be interpreted as sample-specific associations rather than fixed generational characteristics. This study introduces the Gen-AI Dual Competency Alignment Framework (GADCAF) as a provisional conceptual and interpretive framework, intended to guide future research on AI competency, responsible AI utilisation, and organisational AI integration rather than as a validated theoretical model. The findings advance understanding of the complementary functional and ethical dimensions of AI adoption in higher education. Full article