Search Results (1,996)

Accurate monitoring of shaft voltages and bearing currents in large turbo-generators is essential for promoting the sustainable operation of critical power infrastructure. Conventional monitoring systems often rely on threshold triggers that fail to identify early-stage degradation in shaft-earthing brushes. This paper presents an advanced diagnostic approach based on real-time shaft voltage and current measurements collected from four large utility-scale steam turbine generators. Through detailed analysis of time-domain waveforms, frequency-domain spectra, and current scatter plots, characteristic electrical signatures were established for four operational case studies for faults: (i) a floating voltage brush, (ii) a floating current brush, (iii) a worn brush, and (iv) oil/dust contamination. This study demonstrates that each fault produces a distinctive pattern, such as the suppressed RMS shaft voltage with transient spikes in floating voltage brushes, elevated DC offsets and even-order harmonics in floating current brushes, erratic waveforms and intermittent surges in worn brushes, and elevated DC bias with increased current under contamination. These findings establish actionable thresholds for predictive maintenance, fostering enhanced reliability, optimized asset life, and reduced maintenance-related environmental impact. Full article

This study proposes a novel artificial neural network-based methodology for classifying the incident wave direction during ship navigation using the heave–roll–pitch motion response spectra as input. The proposed model demonstrated a balanced performance with an overall accuracy of approximately 0.888, effectively classifying the wave direction into three major categories: head-sea, beam-sea, and following-sea. The methodology utilizes Response Amplitude Operators derived from linear potential flow theory to generate motion response spectra, which are then used to classify the incident wave direction. The model effectively learns the frequency-distribution characteristics of the response spectrum, enabling wave direction classification without the need for complex inverse analysis procedures. This approach is significant in that it allows wave direction recognition solely based on measurable ship motion responses, without the need for additional external sensors or mathematical modeling. This data-driven approach has strong potential for integration into autonomous ship situational awareness modules and real-time wave monitoring technologies. However, the study simplified the directional domain into three representative groups, and the model was validated primarily using a numerically generated dataset, indicating the need for future improvements. Future research will expand the dataset to include a broader range of sea states, improve directional resolution, and explore continuous wave direction prediction. Additionally, further validation using field-measured data will be conducted to assess the real-time applicability of the proposed model. Full article

Against the backdrop of growing demand for rapid soil testing technologies in precision agriculture, this study proposes a detection method based on pyrolysis-electronic nose and machine olfaction signal analysis to achieve precise measurement of key soil nutrients. An electronic nose system comprising 10 metal oxide semiconductor gas sensors was constructed to collect response signals from 112 black soil samples undergoing pyrolysis at 400 °C. By extracting time-domain and frequency-domain features from sensor responses, an initial dataset of 180 features was constructed. A novel feature fusion method combining Pearson correlation coefficients (PCC) with recursive feature elimination cross-validation (RFECV) was proposed to optimize the feature space, enhance representational power, and select key sensitive features. In predicting soil organic matter (SOM), total nitrogen (TN), available potassium (AK), and available phosphorus (AP) content, we compared support vector machines (SVM), support vector machine-random forest models (SVM-RF), and particle swarm optimization-enhanced support vector machine-random forest models (PSO-SVM-RF). Results indicate that PSO-SVM-RF demonstrated optimal performance across all nutrient predictions, achieving a coefficient of determination (R²) of 0.94 for SOM and TN, with a performance-to-bias ratio (RPD) exceeding 3.8. For AK and AP, R² improved to 0.78 and 0.74, respectively. Compared to the SVM model, the root mean square error (RMSE) decreased by 25.4% and 21.6% for AK and AP, respectively, with RPD values approaching the practical threshold of 2.0. This study validated the feasibility and application potential of combining electronic nose technology with a time-frequency domain feature fusion strategy for precise quantitative analysis of soil nutrients, providing a new approach for soil fertility assessment in precision agriculture. Full article

Photoacoustic (PA) velocimetry offers a promising solution to the limitations of conventional techniques for measuring blood flow velocity. Given its moderate penetration depth and high spatial resolution, PA imaging is considered suitable for measuring low-velocity blood flow in capillaries located at moderate depths. High-resolution measurements based on PA signals from individual blood cells can be achieved using a high-frequency transducer. However, high-frequency signals attenuate rapidly within biological tissue, restricting the measurable depth. Consequently, low-frequency transducers are required for deeper measurements. To date, PA flow velocimetry employing low-frequency transducers remains insufficiently explored. In this study, we investigated the effect of the concentration of particles that mimic blood cells within vessels under low-concentration conditions. The performance of flow velocity measurement was evaluated using an ultrasonic transducer (UST) with a center frequency of 10 MHz. The volume fraction of particles in the solution was systematically varied, and the spatially averaged flow velocity was assessed using two different distinct analysis methods. One method employed a time-shift approach based on cross-correlation analysis. Flow velocity was estimated from PA signal redpairs generated by particles dispersed in the fluid, using consecutive pulsed laser irradiations at fixed time intervals. The other method employed a pulsed Doppler method in the frequency domain, widely applied in ultrasound Doppler measurements. In this method, flow velocity redwas estimated from the Doppler-shifted frequency between the transmitted and received signals of the UST. For the initial analysis, numerical simulations were performed, followed by experiments based on displacement measurements equivalent to velocity measurements. The target was a capillary tube filled with an aqueous solution containing particles at different concentration levels. The time–domain method tended to underestimate flow velocity as particle concentration increased, whereas the pulsed Doppler method yielded estimates consistent with theoretical values, demonstrating its potential for measurements at high concentrations. Full article

We present systematic experimental studies of the impact of subwavelength structural geometries and electromagnetic field polarization on the resonance behavior of metallic metasurfaces at D-band frequencies. The measured influence of the photoconductive receiver antenna design in terahertz time-domain spectroscopy on the frequency-domain spectral features was analyzed. Numerical simulations reveal distinct resonance characteristics in the D-band regime, where extraordinary amplitude transmission is highly dependent on the array dimensions and field polarization orientation. The metasurface enables significant enhancements in surface electric fields and resonance response, attributed to the effective excitation of strong dipolar modes. These results demonstrate the extraordinary transmission capabilities of subwavelength metallic arrays and provide valuable insights for designing compact, low-loss, and tunable terahertz functional components needed in next-generation communications. Full article

Accurate recognition of complex human activities from wearable sensors plays a critical role in sports analytics and human performance monitoring. However, the high dimensionality and redundancy of raw inertial data can hinder model performance and interpretability. This study proposes a hybrid feature selection framework that combines Minimum Redundancy Maximum Relevance (MRMR) and Regularized Neighborhood Component Analysis (RNCA) to improve classification accuracy while reducing computational complexity. Multi-sensor inertial data were collected from field hockey players performing six activity types. Time- and frequency-domain features were extracted from four body-mounted inertial measurement units (IMUs), resulting in 432 initial features. MRMR, combined with Pearson correlation filtering (|ρ| > 0.7), eliminated redundant features, and RNCA further refined the subset by learning supervised feature weights. The final model achieved a test accuracy of 92.82% and F1-score of 86.91% using only 83 features, surpassing the MRMR-only configuration and slightly outperforming the full feature set. This performance was supported by reduced training time, improved confusion matrix profiles, and enhanced class separability in PCA and t-SNE visualizations. These results demonstrate the effectiveness of the proposed two-stage feature selection method in optimizing classification performance while enhancing model efficiency and interpretability for real-time human activity recognition systems. Full article

Static eccentricity is an important early-stage fault in marine diesel generators, as small air-gap deviations caused by misalignment or mechanical wear can escalate into bearing damage and rotor–stator contact. To address the challenge of detecting such subtle faults, this study proposes a current signal analysis method based on the two-dimensional short-time Fourier transform (2D STFT) for early detection of static eccentricity faults in marine diesel generators. Using three-phase currents measured during normal operation and fault data synthesized with a physics-based electromechanical coupling model (1–5% eccentricity), we construct a two-dimensional phase–time representation rather than treating each phase as an independent one-dimensional time series and then apply 2D STFT. This formulation enables the simultaneous capture of inter-phase relationships and spatial patterns in the time–frequency–phase domain. Experiments indicate a distinct energy rise near 1020 Hz as static eccentricity increases. This trend enables the proposed method to distinguish small faults of approximately 5% eccentricity, which remain difficult to detect using conventional 1D STFT. As a result, the approach improves the diagnostic accuracy of non-contact, current-based monitoring for static eccentricity faults. Future work will include validation using real in-service fault data and extensions to other fault modes such as dynamic eccentricity and bearing defects. Full article

Background/Objectives: Tinnitus causes significant cognitive and emotional distress; however, its clinical assessment mostly relies on subjective measures without evaluation of objective indices. In this narrative review, we examined the potential of electroencephalography (EEG)-based neurophysiological markers as objective biomarkers in tinnitus assessment. Methods: The Web of Science, PubMed, EMBASE, and MEDLINE databases were searched to identify research articles on EEG-based analysis of individuals with tinnitus. Studies in which treatment and control groups were compared across four analytical domains (spectral power analysis, functional connectivity, microstate analysis, and entropy measures) were included. Qualitative synthesis was conducted to elucidate neurophysiological mechanisms, methodological characteristics, and clinical implications. Results: Analysis of 18 studies (n = 1188 participants) revealed that tinnitus is characterized by distributed neural dysfunction that extends beyond the auditory system. Spectral power analyses revealed sex-dependent, frequency-specific abnormalities across distributed brain regions. Connectivity analyses demonstrated elevated long-range coupling in high-frequency bands concurrent with diminished low-frequency synchronization. Microstate analyses revealed alterations in spatial configuration and transition probabilities. Entropy quantification indicated elevated complexity, particularly in the frontal and auditory cortices. Conclusions: EEG-derived neurophysiological markers demonstrate associations with tinnitus in group analyses and show potential for elucidating pathophysiological mechanisms. However, significant limitations, including low spatial resolution, small sample sizes, methodological heterogeneity, and lack of validation for individual-level diagnosis or treatment prediction, highlight the need for cautious interpretation. Standardized analytical protocols, larger validation studies, multimodal neuroimaging integration, and demonstration of clinical utility in prospective trials are required before EEG markers can be established as biomarkers for tinnitus diagnosis and management. Full article

In optical sensing, signal demodulation often degrades fine spectral data, particularly in spectroscopic measurements affected by Doppler noise, aliasing, and circuit noise. Existing algorithms often fall short in addressing these issues effectively, as they either necessitate complex parameter tuning and extensive expertise or are limited to handling simple spectral signals. To address these challenges, this study proposes an adaptive spectral extraction algorithm combining Variational Mode Decomposition (VMD) and Savitzky-Golay (SG) filtering. The algorithm optimizes parameters through an innovative adaptation strategy. By analyzing key parameters such as SG frame length, order, and VMD mode number, it leverages signal time-domain and frequency spectrum information to adaptively determine the optimal VMD modes and SG order, ensuring effective noise suppression and feature preservation. Validated through simulations and experiments, the method significantly enhances spectral signal quality by restoring absorption peaks and eliminating manual parameter adjustments. This work provides a robust solution for improving measurement accuracy and reliability in optical sensing instrumentation, particularly in applications involving complex spectral analysis. Full article

Early detection of breast cancer commonly relies on imaging technologies such as ultrasound, mammography and MRI. Among these, breast ultrasound is widely used by radiologists to identify and assess lesions. In this study, we developed image segmentation techniques and multiclass classification artificial intelligence (AI) tools based on pretrained models to segment lesions and detect breast cancer. The proposed workflow includes both the development of segmentation models and development of a series of classification models to classify ultrasound images as normal, benign or malignant. The pretrained models were trained and evaluated on the Breast Ultrasound Images (BUSI) dataset, a publicly available collection of grayscale breast ultrasound images with corresponding expert-annotated masks. For segmentation, images and ground-truth masks were used to pretrained encoder (ResNet18, EfficientNet-B0 and MobileNetV2)–decoder (U-Net, U-Net++ and DeepLabV3) models, including the DeepLabV3 architecture integrated with a Frequency-Domain Feature Enhancement Module (FEM). The proposed FEM improves spatial and spectral feature representations using Discrete Fourier Transform (DFT), GroupNorm, dropout regularization and adaptive fusion. For classification, each image was assigned a label (normal, benign or malignant). Optuna, an open-source software framework, was used for hyperparameter optimization and for the testing of various pretrained models to determine the best encoder–decoder segmentation architecture. Five different pretrained models (ResNet18, DenseNet121, InceptionV3, MobielNetV3 and GoogleNet) were optimized for multiclass classification. DeepLabV3 outperformed other segmentation architectures, with consistent performance across training, validation and test images, with Dice Similarity Coefficient (DSC, a metric describing the overlap between predicted and true lesion regions) values of 0.87, 0.80 and 0.83 on training, validation and test sets, respectively. ResNet18:DeepLabV3 achieved an Intersection over Union (IoU) score of 0.78 during training, while ResNet18:U-Net++ achieved the best Dice coefficient (0.83) and IoU (0.71) and area under the curve (AUC, 0.91) scores on the test (unseen) dataset when compared to other models. However, the proposed Resnet18: FrequencyAwareDeepLabV3 (FADeepLabV3) achieved a DSC of 0.85 and an IoU of 0.72 on the test dataset, demonstrating improvements over standard DeepLabV3. Notably, the frequency-domain enhancement substantially improved the AUC from 0.90 to 0.98, indicating enhanced prediction confidence and clinical reliability. For classification, ResNet18 produced an F1 score—a measure combining precision and recall—of 0.95 and an accuracy of 0.90 on the training dataset, while InceptionV3 performed best on the test dataset, with an F1 score of 0.75 and accuracy of 0.83. We demonstrate a comprehensive approach to automate the segmentation and multiclass classification of breast cancer ultrasound images into benign, malignant or normal transfer learning models on an imbalanced ultrasound image dataset. Full article

We study statistical dependence in the frequency domain using the integral probability metric (IPM) framework. We propose the uniform Fourier dependence measure (UFDM) defined as the uniform norm of the difference between the joint and product-marginal characteristic functions. We provide a theoretical analysis, highlighting key properties, such as invariances, monotonicity in linear dimension reduction, and a concentration bound. For the estimation of the UFDM, we propose a gradient-based algorithm with singular value decomposition (SVD) warm-up and show that this warm-up is essential for stable performance. The empirical estimator of UFDM is differentiable, and it can be integrated into modern machine learning pipelines. In experiments with synthetic and real-world data, we compare UFDM with distance correlation (DCOR), Hilbert–Schmidt independence criterion (HSIC), and matrix-based Rényi’s $\alpha$-entropy functional (MEF) in permutation-based statistical independence testing and supervised feature extraction. Independence test experiments showed the effectiveness of UFDM at detecting some sparse geometric dependencies in a diverse set of patterns that span different linear and nonlinear interactions, including copulas and geometric structures. In feature extraction experiments across 16 OpenML datasets, we conducted 160 pairwise comparisons: UFDM statistically significantly outperformed other baselines in 20 cases and was outperformed in 13. Full article

Objectives: This prospective, real-world study, designed by the Greek Research Alliance for Studying headache and Pain (GRASP), primarily aimed to examine whether a single administration of eptinezumab 100 mg can provide early therapeutic benefits over weeks 1–4 in preventing high-frequency chronic migraine (HFCM). Methods: We enrolled adults with HFCM (>23 monthly headache days—MHD) who had failed at least three preventive therapies and received a single infusion of 100 mg IV eptinezumab. Its efficacy was assessed using daily headache diaries over weeks 1–4 and at week 12, as well as with the Patients’ Global Impression of Change (PGIC) scale at week 4. Primary outcomes included ≥50% reduction in monthly headache days (MHD) through week 4 and early response at week 1. Secondary outcomes included changes in migraine severity, acute medication use, and incidence of most bothersome symptoms (MBS) accompanying headache. Results: A total of 39 HFCM patients were analyzed. Their mean age was 48.1 years, and most were female. More than half (n = 22; 56%) were fast-responders, showing >50% reduction in MHD at week 1. Early 50% response rates at week 1 favored CGRP-naïve patients, compared to prior non-responders. Likewise, significant improvement was observed early across all other efficacy domains through week 4, with sustained benefits through month 3. MBS, like photophobia and nausea, decreased notably, while osmophobia and allodynia improved less over weeks 1–4, post-eptinezumab. At week 4, 64.1% of patients reported meaningful overall improvement on PGIC. Conclusions: Eptinezumab rapidly prevented HFCM with clinically meaningful benefits seen early at week 1. Patients experienced sustained improvement in all measured clinical domains through week 4 and onward to month 3. Full article

Out-of-school time (OST) STEM programs are well-positioned to strengthen family engagement, yet practical, theory-aligned tools remain limited. This early-stage mixed-methods study tests parent/caregiver (P/C) and staff (S) surveys based on Clover for Families developmental theory expressed through the CARE framework: Connect (welcoming climate, clear communication), Act (hands-on participation, at-home supports), Reflect (shared meaning-making, feedback), and Empower (family voice, decision-making). Nine OST STEM programs (eight U.S. states) co-designed/piloted CARE plans, activities, and surveys over six months. Quantitative data included baseline experiences (CARE practice frequency; n = 67 P/C, 42 S across nine programs), program-end reflection (retrospective perceptions of change; n = 26 P/C, 29 S), and forced-ranking (most/least important domains; n = 67 P/C, 42 S). Qualitative data from meetings, open responses, and interviews were analyzed to contextualize quantitative findings, which included strong internal consistency (P/C α = 0.83–0.95; S α = 0.77–0.95) and large retrospective gains in both groups across domains. Forced-ranking elevated Connect and Act over Reflect and Empower, highlighting a need to scaffold family involvement. Staff described CARE as useful and actionable. Findings show that CARE supports measurement and continuous improvement of STEM family engagement. Future work should test large-sample validity, link results to observed practice and youth outcomes, and refine Empowerment-related items for everyday agency. Full article

Offshore wind turbine support structures are in harsh and unsteady marine environments, and their dynamic characteristics could change gradually after long-term service. To better understand the status and improve remaining life estimation, it is essential to conduct in situ measurement and update the numerical models of these support structures. In this paper, the modal properties of a 5.5 MW offshore wind turbine were first identified by a widely used operational modal analysis technique, frequency-domain decomposition, given the acceleration data obtained from eight sensors located at four different heights on the tower. Then, a finite element model was created in MATLAB R2020a and a set of model parameters including scour depth, foundation stiffness, hydrodynamic added mass and damping coefficients was updated in a Bayesian inference frame. It is found that the posterior distributions of most parameters significantly differ from their prior distributions, except for the hydrodynamic added mass coefficient. The predicted natural frequencies and damping ratios with the updated parameters are close to those values identified with errors less than 2%. But relatively large differences are found when comparing some of the predicted and identified mode shape coefficients. Specifically, it is found that different combinations of the scour depth and foundation stiffness coefficient can reach very similar modal property predictions, meaning that model updating results are not unique. This research demonstrates that the Bayesian inference framework is effective in constructing a more accurate model, even when confronting the inherent challenge of non-unique parameter identifiability, as encountered with scour depth and foundation stiffness. Full article

Polyurethane (PU) has proven to be an effective material for reinforcing frozen-soil roadbeds; however, the excessive use of PU increases cost and contamination and limits its large-scale application in practical projects. To fill this gap, laboratory tests were conducted to determine the optimal content that achieved the best reinforcement effect at the lowest cost. A continuous frost-heave strain profile and its variation features were obtained through laboratory tests using advanced Rayleigh optical frequency-domain reflectometry technology (OFDR). A calibration method for OFDR at negative temperatures was introduced. The influences of the PU content, water content, and ambient temperature on frost heave were determined based on distributed measurements. The results indicate that a linear function is suitable for describing the relationship between the strain shift and temperature variation above 0 °C, whereas a cubic function is suggested below 0 °C, with a fitted R² of 1. When the moisture content is 4.7% and the ambient temperature is −20 °C, compared with the original reinforced soil, the frost-heave displacement decreased by 33.27%, 47.43%, 71.65%, and 72.77%, respectively, after reinforcement with PU contents of 4%, 8%, 12%, and 16%. When the moisture content increased from 4.7% to 10% and the ambient temperature was −20 °C, compared to the original reinforced soil, the frost-heave displacement of the reinforced soil with PU contents of 4%, 8%, 12%, and 16% increased by 49.34%, 14.93%, 7.48%, and 0.16%, respectively. When the PU content was less than 4%, the reinforcement effect was insignificant. The freezing point and frost heave rate decreased after the addition of PU owing to the changes in the pore structure and matric suction. Full article