Search Results (1,910)

Bearings play crucial roles in industrial machinery. Therefore, the continuous monitoring and effective detection of bearing failures are essential to ensure the safety and reliability of motors. Traditional fault diagnosis methods often require information from both the time and frequency domains; however, converting them into a two-dimensional representation significantly increases computational costs. Conversely, utilizing only time-domain features while ignoring frequency-domain features results in incomplete fault information, reducing accuracy under various operating conditions. This study proposes an efficient dual-stream network with soft-gated fusion for bearing fault diagnosis that simultaneously analyzes acoustic emission signals in the time and frequency domains. Our approach employs two separate feature-learning branches: the time-domain branch directly extracts features from the segmented raw acoustic emission signals, and the frequency-domain branch learns features from one-dimensional spectral vectors obtained using the fast Fourier transform. A gated fusion mechanism adaptively balances the contribution of each domain before classifying fault types. The experimental results show that the proposed method significantly reduces the computational cost compared with that of a two-dimensional-representation-based model and improves accuracy over time-only or frequency-only baselines. Full article

The resolution and accuracy of airborne/spaceborne SAR are continuously improving, making it an effective means for observing ocean dynamic processes and detecting marine targets. In contrast, utilizing its unique orbital angular momentum (OAM) mode, vortex radar does not require temporal accumulation to achieve azimuthal resolution, making it particularly suitable for observing moving sea surfaces. This capability enables stable and continuous monitoring of dynamic ocean scenes. This paper proposes a vortex radar imaging method based on three-dimensional sea surface scattering characteristics: first, a three-dimensional wind-driven sea surface geometric model is established based on the Elfouhaily sea spectrum, and its scattering characteristics under different incident angles, wind speeds, and wind directions are analyzed using the semi-deterministic facet-based two-scale method; then, two-dimensional range-azimuth imaging is achieved through coordinate transformation, echo modeling, pulse compression, and fast Fourier transform (FFT) in OAM mode domain, with the correctness of the imaging algorithm verified through multiple point target imaging results. Finally, simulation results of two-dimensional sea surface vortex imaging under different incident angles are presented, and the influence of wind speed and direction on sea surface vortex imaging is analyzed. The study shows that the vortex imaging system can effectively reflect wave fluctuations and wind direction characteristics, demonstrating the feasibility and potential of vortex radar imaging in oceanographic applications. Full article

To investigate the characteristics of pressure pulsation induced by vortex ropes in the draft tube of a centrifugal pump under runaway conditions, a closed double-layer hydraulic test bench was established in this study. Runaway characteristic experiments were conducted, and pressure pulsation signals were acquired at heads of 7.6 m, 9.6 m, and 11.9 m. The measured pressure data were analyzed in the time–frequency domain using Fast Fourier Transform (FFT) and Wavelet Transform (WT). The results show that both the runaway rotational speed and the reverse flow rate increase with increasing head. Under all three heads, the dominant frequency upstream of the elbow section of the draft tube is 0.53 times the rotational frequency, confirming that the vortex rope in the draft tube serves as the primary excitation source of the flow field. As the vortex rope is conveyed by the main flow through the elbow, it undergoes impingement and fragmentation, causing the dominant frequency downstream of the elbow to decrease to 0.1 times the rotational frequency. The dominant frequency induced by the vortex rope remains continuous over time, whereas the frequency arising from the coupling between the vortex rope and rotor–stator interaction exhibits pronounced time-varying oscillations. These oscillations intensify with increasing head, and their frequency oscillation range broadens from 4 to 6 times the rotational frequency at low head to 2–8 times at high head. These findings provide a theoretical foundation for the preventive and protective design of centrifugal pumps under runaway conditions. Full article

Distributed acoustic sensing (DAS) has been widely applied to injection–production profile monitoring in horizontal wells because it provides continuous full-wellbore coverage, real-time acquisition, and straightforward long-term deployment. In practical downhole operations, however, DAS measurements are frequently compromised by optical-signal attenuation, loss of fiber–casing/formation coupling, and environmental noise. Meanwhile, the mechanisms governing flow-induced acoustic responses remain insufficiently understood, which continues to impede quantitative diagnosis and interpretation of injection–production profiles based on DAS data. To address these challenges, this study performed controlled laboratory-scale physical simulation experiments of single-phase flow in a horizontal wellbore, systematically investigating DAS acoustic responses under two wellbore diameters (25 mm and 50 mm) and a range of flow velocities. Power spectral density (PSD) was derived using the fast Fourier transform to identify flow-sensitive characteristic frequency bands, and frequency-band energy (FBE) was further used to establish an optimal quantitative relationship with flow velocity. The results show that: (1) DAS energy is dominated by low-frequency components (<100 Hz), with the total energy increasing nonlinearly as flow velocity rises, accompanied by a progressive broadening of the characteristic bands; (2) the feature bands identified using an adaptive method based on energy difference statistics applied to PSD frequency-domain features exhibit a higher signal-to-noise ratio and greater physical clarity than traditional wide frequency bands; furthermore, by employing a feature band merging strategy, the distribution characteristics of flow energy can be captured more comprehensively; and (3) FBE exhibits a strong nonlinear dependence on flow velocity, with a power-law model delivering the best theoretical fit, whereas a cubic model (FBE ∝ V³) achieves high accuracy and robustness for practical applications. The proposed workflow—“PSD peak identification–characteristic band delineation–FBE regression”—establishes a methodological foundation for quantitative DAS-based monitoring of horizontal-well injection–production profiles in both laboratory and field settings, and it provides a basis for subsequent intelligent monitoring and interpretation under multiphase-flow conditions. Full article

Broiler respiratory rate (RR) in cage systems is a core physiological indicator of health and stress. However, real-time, non-invasive continuous RR monitoring is difficult in a high-density breeding environment, thereby limiting precise poultry health management. This study developed a feasible non-contact broiler RR measurement method to address this gap. The proposed method integrates infrared thermal imaging and phase-based video magnification (PBVM). Using cage-reared white-feathered broilers as subjects, we selected the thoracodorsal and tail regions as regions of interest (ROI), applied PBVM to amplify subtle respiratory-related body surface movements, and extracted RR features via the Fast Fourier Transform (FFT). Two validation experiments were conducted under controlled laboratory conditions. One was an RR dynamic monitoring experiment covering the entire life cycle (4 to 36 days), which analyzed video data of 198 individual quiet broilers. The other was a multi-gradient heat stress experiment with temperature increases of +2 °C, +4 °C, and +5 °C, and analyzed video data of 162 individual quiet broilers. The method achieved favorable measurement accuracy: in the whole-life-stage experiment, the mean absolute error (MAE) was 0.036 Hz, the mean absolute percentage error (MAPE) was 4.461%, and the coefficient of determination (R²) reached 0.961; in the heat stress experiment, the MAE was 0.042 Hz, the MAPE was 3.270%, and the R² reached 0.928. Linear regression analysis confirmed that healthy broiler RR decreased linearly with increasing age, and verified that RR showed a stepwise response to thermal challenge with a positive correlation between RR increase and temperature increment, accompanied by growth stage specificity. This study provides a feasible non-invasive approach for broiler RR monitoring, offering preliminary reference data for early heat stress detection and sustainable poultry production. Full article

This study investigates the effectiveness of various feature extraction methods applied to vibration signals for the automatic detection of production defects in CNC (Computerised Numerical Control) milling machines. A dataset consisting of real-world data collected from CNC machines equipped with accelerometers was used. The objective of the study is to compare three main groups of techniques: time-domain analysis (TDA), frequency-domain analysis (FDA), and time–frequency-domain analysis (TFA). The findings indicate that basic TDA features lack the necessary sensitivity to accurately distinguish between Good Processing (GP) and Bad Processing (BP) states. Frequency-domain methods, such as the Fast Fourier Transform (FFT), median frequency calculation, and the Welch periodogram, provide better insights but still have limitations. The most effective results are obtained with TFA methods, particularly Empirical Mode Decomposition (EMD) and the Hilbert–Huang Transform (HHT), which reveal deeper signal characteristics. Following the feature optimisation studies, it was determined that a combination of four features—FMED, IMF2, IMF5 and WPT26—yielded the optimal performance, with an accuracy of 91.48%. The incorporation of a fifth feature resulted in information saturation within the model and did not improve performance. This study makes a novel contribution to literature by conducting an in-depth investigation into the most effective feature extraction and selection techniques for achieving robust discrimination between GP and BP productions using vibration signals in CNC milling processes. Conclusively, TFA features, supported by advanced signal processing, offer a strong basis for reliable, automated defect detection in CNC milling operations. Full article

Aiming to address the complex coupling of transient impulses and steady-state components in vibration signals of scroll compressors in air conditioning systems, this study proposes a physically driven heterogeneous dual-path adaptive weighting network (PDW-Net). The approach constructs a physics-inspired weighting module based on kurtosis and energy criteria, enabling adaptive reconstruction of transient impulses and steady-state vibration components. Feature extraction and decision-level fusion are achieved through a heterogeneous dual-branch network comprising a Fast Fourier Transform (FFT)-based one-dimensional convolutional neural network (1D-CNN) and a Short-Time Fourier Transform (STFT)-based two-dimensional convolutional neural network (2D-CNN). In experimental validation covering four typical fault conditions—condenser failure, refrigerant deficiency, refrigerant overcharge, and main shaft wear—the PDW-Net achieved an average diagnostic accuracy of 97.87% (standard deviation: 2.60%), with 100% accuracy in identifying refrigerant deficiency and normal operating states, demonstrating significant superiority over existing mainstream methods. Ablation studies reveal that the adaptive weighting mechanism contributes most substantially to performance, as its removal results in a 34.24 percentage point drop in accuracy. Replacing the heterogeneous dual-branch structure with a homogeneous counterpart reduces accuracy by 16.18 percentage points, robustly validating the efficacy of the physics-guided and heterogeneous fusion design. Full article

The Hough transform (HT) is widely used in computer vision, tomography, and neural networks. Numerous algorithms for HT computation have been proposed, making their systematic comparison essential. However, existing comparative methodologies are either non-universal and limited to certain HT formulations or task-oriented, relying on application-specific criteria that do not fully capture algorithmic properties. This paper introduces a novel unified methodology for the systematic comparison of HT algorithms. It evaluates key characteristics, including computational complexity, accuracy, and auxiliary space complexity, while explicitly accounting for the property of self-adjointness. The methodology integrates both implementation-level and theoretical considerations related to the interpretation of HT as a discrete approximation of the Radon transform. A set of mathematically justified evaluation functions, not previously described in the literature, is proposed to support our methodology. Importantly, the methodology is universal, applicable across diverse HT paradigms, encompasses pattern-based and Fourier-based fast HT (FHT) algorithms, and offers a comprehensive alternative to existing task-specific methodologies. Its application to several state-of-the-art FHT algorithms ($FHT2DT$, $FHT2SP$, $ASD2$, $KHM$, and Fast Slant Stack) yields new experimentally confirmed theoretical insights, identifies $ASD2$ as the most balanced algorithm, and provides practical guidelines for algorithm selection. In particular, the methodology reveals that for image sizes up to 3000, the maximum normalized computational complexity increases as follows: $FHT2DT$ (1.1), $ASD2$ (15.3), and $KHM$ (30.6), while the remaining algorithms exhibit at least 1.1 times higher values. The maximum orthotropic approximation error equals 0.5 for $ASD2$, $KHM$, and Fast Slant Stack; lies between 0.5 and 1.5 for $FHT2SP$; and reaches 2.1 for $FHT2DT$. In terms of worst-case normalized auxiliary space complexity, the lowest values are achieved by $FHT2DT$ (2.0), Fast Slant Stack (4.0, lower bound), and $ASD2$ (6.8), with all other algorithms requiring at least 8.2 times more memory. Full article

This paper aims to present a non-destructive, optimized variational mode decomposition (VMD)-based ground-penetrating radar (GPR) method developed for identifying void defects in reinforced concrete (RC) structures. This study also presents an enhanced framework for defect detection in RC by integrating advanced spectrum analysis with deep learning techniques. A GPR investigation was conducted on an RC bridge deck with known structural defects to generate a representative dataset reflecting both intact and void-defective conditions. In addition to conventional spectral techniques such as fast Fourier transform (FFT), spectrogram, and scalogram, an optimized variational mode decomposition (VMD) method was implemented. The VMD approach decomposes GPR signals into intrinsic mode functions, enabling refined feature extraction beyond traditional spectral methods and allowing clear differentiation between intact and defective signals. The limited availability and quality of GPR small datasets have restricted the application of a functional 1D-CNN which generally requires at least several hundred datasets. To address this challenge, a data augmentation strategy is adopted. FFT-based features were successfully utilized to train a one-dimensional convolutional neural network (1D-CNN) for automated defect identification. The results demonstrate that both the advanced spectrum-based approach and the hybrid framework combining spectral analysis with deep learning significantly improve defect detection performance. Overall, the proposed methodology provides an effective and intelligent solution to support timely, data-driven decision-making for maintenance and safety assurance of bridge infrastructure. Full article

Accurate crop classification provides fundamental data for agricultural resource management and ecological research. Hyperspectral image (HSI) classification is the core technique for achieving precise crop mapping. However, existing models often suffer from excessive parameters, limited robustness under few-shot conditions, and a trade-off between efficiency and robustness. To address these issues, this paper proposes a Multi-Scale Wavelet Kolmogorov–Arnold Network (MSWKN). The model employs a Two-Branch Feature Extractor (TBFE) to capture both spectral correlations and spatial textures. a Channel Cross-Spatial (CCS) module to suppress background clutter and highlight discriminative regions. A group convolution-based Fixed Wavelet Multi-Scale Convolutional Layer (FW-MSCL) that leverages the time–frequency localization of wavelets and learnable linear combinations to enhance robustness against spectral distortion while reducing parameters. And a Fourier-based Transformer encoder to enable global frequency–space modeling. Experiments on the WHU-Hi-HanChuan and WHU-Hi-HongHu hyperspectral crop datasets show that MSWKN achieves high overall accuracy and performs favorably on few-shot categories. Under lower parameter counts and fast inference conditions, the model demonstrates a reasonable trade-off between accuracy and computational efficiency. Ablation studies and wavelet kernel comparisons further confirm the contribution of each module and the advantage of the wavelet. The proposed framework provides an efficient and robust solution for fine-grained hyperspectral crop classification. Full article

Railway traffic monitoring requires robust detection technologies capable of operating reliably under real-world vibration and environmental conditions. In this work, we present the design and validation of an optical vibration sensor based on a Single-mode–Multimode–Single-mode (SMS) fiber structure for Light Rail Vehicle (LRV) detection. The sensing mechanism relies on multimodal interference in the multimode fiber (MMF), where rail-induced vibrations modify the guided mode distribution and, consequently, the transmitted optical intensity. The optical signal is converted to voltage and processed through an embedded acquisition system. Additionally, we conducted tests with freight trains and maintenance trains in order to evaluate the applicability of the sensor in other types of trains besides the LRV. We conducted laboratory experiments to assess mechanical stability, sensibility, and packaging strategies, followed by supervised field tests on an operational LRV line. The recorded time-domain signal exhibited clear modulation during train passage, and first-derivative and sliding-window variance analyses were applied to reliably identify vibration events, even in the presence of slow baseline drift. In addition, frequency-domain analysis was performed by applying the Fast Fourier Transform (FFT) to the measured signal, enabling the identification of characteristic low-frequency spectral components induced by train passage. A quantitative sensitivity assessment was further carried out by correlating the integrated spectral energy (0–12 Hz) with vehicle weight, yielding a linear response with a sensitivity of 0.0017 a.u./t and coefficient of determination $R^2=0.933$. The proposed solution demonstrated stable operation using commercially available low-cost components, confirming the feasibility of SMS-based optical sensing for railway monitoring. These results indicate strong potential for future deployment in traffic safety systems and distributed sensing networks. Full article

As the low-altitude economy develops rapidly, precise radar altimetry is crucial for ensuring the safety and reliability of drone flights. In the context of urban radio detection, the presence of numerous buildings and ground surfaces gives rise to electromagnetic wave multipath propagation. This objective factor gives rise to errors in radar altimetry. Existing channel models often lack the intricate details required to accurately quantify multipath error mechanisms in kilometer-scale complex electromagnetic environments. Therefore, there is an urgent need for a high-fidelity simulation framework. The present study has put forward a pioneering approach to radio altimetry simulation and accuracy assessment in intricate urban environments. The objective of this study is to investigate the impact of multipath propagation on radar altimetry precision. The present study has proposed a novel integration of radar altimetry simulation with kilometre-scale urban electromagnetic simulation models. The simulation of echo signals has been achieved through the utilization of the shooting and bouncing rays (SBR) method and inverse fast Fourier transform (IFFT). A comparative analysis has been conducted based on ranging results from radar systems for different urban models, thereby enabling a mechanism analysis of factors affecting radar altimetry. The study has demonstrated that increased building density and height, along with reduced elevation angles during altimetry, exacerbate ranging errors. Full article

The qEEG findings of subjects with Down syndrome (DS) have not been described in the context of bipolar montage. Resting-state EEG (rsEEG) with a bipolar montage was performed in 22 young adults (26.0 ± 1.2 years) with DS but without psychiatric or neurological pathology and matched control subjects of the same sex and age, and the results were conventionally and numerically analyzed. Channels were grouped into frontal, parieto-occipital, and temporal lobes. For every channel, the power spectrum was calculated and used to compute the area for the delta, theta, alpha and beta bands and was log-transformed. Shannon’s spectral entropy (SSE) and coherence by bands were computed. Finally, we also calculated the peak frequency distribution of the alpha band. qEEG revealed alterations in the rsEEG that were not detected visually. Subjects with DS showed a significant generalized increase in the power of the delta and theta bands, along with a decrease in the power of the alpha band in the posterior half of the scalp. This alpha activity also exhibited features corresponding to older euploid subjects, showing interhemispheric asynchrony in one-third of the individuals. The beta band power was significantly increased in the frontal lobes and adjacent regions, such as the parietal and mid-temporal regions. Individuals with DS showed a generalized decrease in parieto-occipital synchronization associated with intelligence quotient. Left temporal synchronization was also lower. The synchronization of specific channel pairs was greater in subjects with DS in the frontal lobe and much lower in the occipital and temporal regions. These results indicate that alterations in band structure and synchronization in subjects with DS are highly specific and can aid in the clinical evaluation of these individuals. Full article

Carbon fiber-reinforced polymer (CFRP) laminates are widely used in aircraft skins due to their advantages of high strength and lightweight properties. However, their laminate structure and energy-absorbing characteristics result in low-energy impact damage, such as delamination, that is often invisible but can lead to catastrophic failure. Consequently, early detection of delamination in CFRP laminates is necessary. Nonlinear ultrasonic guided waves exhibit high sensitivity to delamination, and second harmonics are widely employed. Compared to second harmonics, one-way mixing of ultrasonic guided waves can excite and receive signals simultaneously at the same location, thereby precisely localizing delamination. This capability has the potential for inspecting buried structures. However, existing literature has not yet fully addressed the generation mechanism of one-way mixing in CFRP laminates nor its interaction with delamination. Based on finite element simulation, this study investigates one-way mixing of A₀ modes and S₀ modes in CFRP laminates. Utilizing pulse-inversion techniques and two-dimensional fast Fourier transforms, the modes and propagation directions of difference-frequency components and sum-frequency components are determined. Furthermore, by utilizing the normalized acoustic nonlinearity parameter χ’ and adjusting the position of the mixing zone through different time delays, delamination in the CFRP laminate is successfully localized. Full article

Advances in Industry 4.0 manufacturing have accelerated the adoption of machine learning (ML) for automated classification. Polyester (PES), a widely used synthetic fiber, competes with natural fibers like cotton and other synthetics, highlighting the need for continuous research and improvement. In the textile sector, distinguishing recycled polyester (rPES) from virgin polyester (vPES) remains challenging due to overlapping chemical signatures and material variability. A combination of Fourier transform infrared (FTIR) spectroscopy and ML has not been explored for this purpose. In this study, we evaluated ML models to discriminate three PES fiber types (45 vPES, 65 rPES, and 55 mixed PES) using 165 FTIR spectra across four spectral regions, R1, R2, R3, and R4, as well as their combined representation. Six ML approaches were tested on data reduced with fast independent component analysis (FastICA) (1–30 components) using an 80/20 train–test dataset split. The Decision Tree classifier achieved the highest Accuracy in four of the five spectral evaluations, with classification accuracies ranging from 66.67% to 77.78% for region R4, which also had a balanced classification profile with an area-under-the-curve (AUC) value of 0.81. Notably, despite the moderate overall Accuracy, the model achieved 100% discrimination of rPES when distinguishing it from both mixed and vPES. Mixed fibers remained the most difficult to classify, highlighting the need for improved feature representation. Full article