Search Results (5,884)

This study combines three optimization algorithms, Tunicate Swarm Algorithm (TSA), Whale Optimization Algorithm (WOA), and Jellyfish Search Optimizer (JSO), with random forest (RF) to predict the shear bond strength of cable bolts under different types and grouting conditions. Based on the original dataset, a database of 860 samples was generated by introducing random noise around each data point. After establishing three hybrid models (RF-WOA, RF-JSO, RF-TSA) and training them, the obtained models were evaluated using six metrics: coefficient of determination (R²), root mean squared error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), variance account for (VAF), and A-20 index. The results indicate that the RF-JSO model exhibits superior performance compared to the other models. The RF-JSO model achieved an excellent performance on the testing set (R² = 0.981, RMSE = 11.063, MAE = 6.457, MAPE = 9, VAF = 98.168, A-20 = 0.891). In addition, Shapley Additive exPlanations (SHAP), Partial Dependence Plot (PDP), and Local Interpretable Model-agnostic Explanations (LIME) were used to analyze the interpretability of the model, and it was found that confining pressure (Stress), elastic modulus (E), and a standard cable type (cable type_standard) contributed the most to the prediction of shear bond strength. In summary, the hybrid model proposed in this study can effectively predict the shear bond strength of cable bolts. Full article

Conventional rail transit train running gear fault diagnosis mainly depends on routine maintenance inspections and manual judgment. However, these approaches lack robustness under complex operational environments and elevated noise levels, rendering them inadequate for real-time performance and the rigorous accuracy standards demanded by modern rail transit systems. Furthermore, many existing deep learning–based methods suffer from inherent limitations in feature extraction or incur prohibitive computational costs when processing multivariate time series data. This study represents one of the early efforts to introduce the TimesNet time series modeling framework into the domain of fault diagnosis for rail transit train running gear. By utilizing an innovative multi-period decomposition strategy and a mechanism for reshaping one-dimensional data into two-dimensional tensors, the framework enables advanced temporal-spatial representation of time series data. Algorithm validation is performed on both the high-speed train running gear bearing fault dataset and the multi-mode fault diagnosis datasets of gearbox under variable working conditions. The TimesNet model exhibits outstanding diagnostic performance on both datasets, achieving a diagnostic accuracy of 91.7% on the high-speed train bearing fault dataset. Embedded deployment experiments demonstrate that single-sample inference is completed within 70.3 ± 5.8 ms, thereby satisfying the real-time monitoring requirement (<100 ms) with a 100% success rate over 50 consecutive tests. The two-dimensional reshaping approach inherent to TimesNet markedly enhances the capacity of the model to capture intrinsic periodic structures within multivariate time series data, presenting a novel paradigm for the intelligent fault diagnosis of complex mechanical systems in train running gears. The integrated human–machine interaction system includes a comprehensive closed-loop process encompassing detection, diagnosis, and decision-making, thereby laying a robust foundation for the continued development of train running gear predictive maintenance technologies. Full article

Facial Expression Recognition (FER) is a research topic of great practical significance. However, existing FER methods still face numerous challenges, particularly in the interaction between spatial and global information, the distinction of subtle expression features, and the attention to key facial regions. This paper proposes a lightweight Global-Aware Spatial (GAS) Attention module, designed to improve the accuracy and robustness of FER. This module extracts global semantic information from the image via global average pooling and fuses it with local spatial features extracted by convolution, guiding the model to focus on regions highly relevant to facial expressions (such as the mouth and eyes). This effectively suppresses background noise and enhances the model’s ability to perceive subtle expression variations. In addition, we further introduce a Squeeze-and-Excitation (SE) Attention module into the dual-branch architecture to adaptively adjust the channel-wise weights of features, emphasizing critical region information and enhancing the model’s discriminative capacity. Based on these improvements, we develop the Ada-DF++ network model. Experimental results show that the improved model achieves test accuracies of 89.21%, 66.14%, and 63.75% on the RAF-DB, AffectNet (7cls), and AffectNet (8cls) datasets, respectively, outperforming current state-of-the-art methods across multiple benchmarks and demonstrating the effectiveness of the proposed approach for FER tasks. Full article

This study presents an acoustic membrane design utilizing a thin foil sound resonance mechanism to enhance sound absorption and insulation performance. The membranes incorporate single-layer and double-layer structures featuring parallel foil square wedge-shaped coffers and a flat bottom panel, separated by air cavities. The enclosed air cavity significantly improves the sound insulation capability of the acoustic membrane. Parametric studies were conducted to investigate key factors affecting the sound transmission loss (STL) of the proposed acoustic membrane. The analysis examined the influence of foil thickness, substrate thickness, and back cavity depth on acoustic performance. Results demonstrate that the membrane structure enriches vibration modes in the 500–6000 Hz frequency range, exhibiting multiple acoustic attenuation peaks and broader noise reduction bandwidth (average STL of 40–55 dB across the researched frequency range) compared to conventional resonant cavities and membrane-type acoustic metamaterials. The STL characteristics can be tuned across different frequency bands by adjusting the back cavity depth, foil thickness, and substrate thickness. Experimental validation was performed through noise reduction tests on an air compressor pump. Comparative acoustic measurements confirmed the superior noise attenuation performance and practical applicability of the proposed membrane over conventional acoustic treatments. Compared to uniform foil resonators, the combination of plastic and steel materials with single-layer and double-layer membranes reduced the overall sound level (OA) by an additional 2–3 dB, thereby offering exceptional STL performance in the low- to medium-frequency range. These lightweight, easy-to-manufacture membranes exhibit considerable potential for noise control applications in household appliances and industrial settings. Full article

This study proposes a system identification (SI) technique based on the constrained recursive least squares (CRLS) method to model the dynamics of the P-SUROII. By simplifying the dynamic model in consideration of the inherent characteristics of underwater vehicles and minimizing the number of parameters to be estimated, the proposed approach aims to improve estimation accuracy. In addition, a simplified thruster input model was applied to quantify the actual thruster output and improve the reliability of the input data. To satisfy the persistent excitation (PE) condition during the estimation process, experiments incorporating various motion modes were designed, and free-running and S-shaped maneuvering tests were additionally conducted to validate the model’s generalization capability and prediction performance. The coefficients estimated using the CRLS method, which is robust to noise and bias, were evaluated using quantitative similarity metrics such as root mean squared error (RMSE) and mean absolute error (MAE), confirming their validity. The proposed method effectively captures the actual dynamics of the underwater vehicle and is expected to serve as a key enabling technology for the future development of high-performance control systems and autonomous operation systems. Full article

According to the World Health Organization, cardiovascular diseases, including cardiac arrhythmias, are the leading cause of death worldwide due to their silent, asymptomatic nature. To address this problem, early and accurate diagnosis is crucial. Although this task is typically performed by a cardiologist, diagnosing arrhythmias can be imprecise due to the subjectivity of reading and interpreting electrocardiograms (ECGs), and electrocardiograms are often subject to noise and interference. Deep learning-based approaches present methods for automatically detecting arrhythmias and are positioned as an alternative to support cardiologists’ diagnoses. However, these methods are trained and tested only on open datasets of electrocardiograms from Holter devices, whose results aim to improve the accuracy of the state of the art, neglecting the efficiency of the model and its application in a practical clinical context. In this work, we propose an efficient model based on a 1D CNN architecture to detect arrhythmias from smartwatch ECGs, for subsequent deployment in a practical scenario for the monitoring and early detection of arrhythmias. Two datasets were used: UMass Medical School Simband for a binary arrhythmia detection model to evaluate its efficiency and effectiveness, and the MIT-BIH arrhythmia database to validate the multiclass model and compare it with state-of-the-art models. The results of the binary model achieved an accuracy of 64.81%, a sensitivity of 89.47%, and a specificity of 6.25%, demonstrating the model’s reliability, especially in specificity. Furthermore, the computational complexity was 1.2 million parameters and 68.48 MFlops, demonstrating the efficiency of the model. Finally, the results of the multiclass model achieved an accuracy of 99.57%, a sensitivity of 99.57%, and a specificity of 99.47%, making it one of the best state-of-the-art proposals and also reconfirming the reliability of the model. Full article

The application of lattices as protective materials/structures is rapidly increasing. This requires improving impact absorption capabilities to protect goods in packaging and prevent human injuries in protective devices. This study aims to improve the accuracy of impact and energy absorption measurements for lattices, addressing the limitations of current methods such as energy-impact diagrams and instrumented drop-impact testers. A novel test setup is introduced by utilizing a modified Charpy test machine equipped with appropriate instrumentation to directly measure both energy and acceleration. Other modifications include adjustments to the machine components and the introduction of a new sandwich configuration for the test specimen, ensuring compatibility with the machine’s geometry and the test objectives. The attractiveness of the proposed test setup lies in its simplicity and efficiency. Unlike drop-impact test machines—which require complex, time-consuming, and error-prone data integration and derivation—the proposed method eliminates the need for postprocessing, as both energy and impact are recorded directly and instantaneously by the machine. The advantage over existing setups becomes particularly evident when considering that, in the presence of noise and high-frequency fluctuations—characteristic of sensor data from impact events—errors in numerical operations can range from 30% to over 100%. The functionality of the proposed test setup is evaluated through a series of experiments, and the results are compared with those obtained from existing methods. Our findings demonstrate the effectiveness of the new setup in providing accurate and direct measures of absorption parameters, offering a significant improvement over the traditional approaches. Full article

This study presents a semi-empirical approach to generalizing the acoustic radiation generated by serrated airfoil configurations, based on small-scale aerodynamic/acoustic experiments and functional regression techniques. In the context of passive noise reduction strategies, such as leading-edge and trailing-edge serrations, acoustic measurements are performed in a controlled subsonic wind tunnel environment. Sound pressure level (SPL) spectra and acoustic power metrics are acquired for various geometric configurations and flow conditions. These spectral data are then analyzed using regression-based modeling techniques—linear, quadratic, logarithmic, and exponential forms—to capture the dependence of acoustic emission on key geometric and flow-related variables (e.g., serration amplitude, wavelength, angle of attack), without relying explicitly on predefined nondimensional numbers. The resulting predictive models aim to describe SPL behavior across relevant frequency bands (e.g., broadband or 1/3 octave) and to extrapolate acoustic trends for configurations beyond those tested. The proposed methodology allows for the identification of compact functional relationships between configuration parameters and acoustic output, offering a practical tool for the preliminary design and optimization of low-noise serrated profiles. The findings are intended to support both physical understanding and engineering application, bridging experimental data and parametric acoustic modeling in aerodynamic noise control. Full article

Sign language is a visual language articulated through body movements. Existing approaches predominantly leverage RGB inputs, incurring substantial computational overhead and remaining susceptible to interference from foreground and background noise. A second fundamental challenge lies in accurately modeling the nonlinear temporal dynamics and inherent asynchrony across body parts that characterize sign language sequences. To address these challenges, we propose a novel part-wise graph Fourier learning method for skeleton-based continuous sign language recognition (PGF-SLR), which uniformly models the spatiotemporal relations of multiple body parts in a globally ordered yet locally unordered manner. Specifically, different parts within different time steps are treated as nodes, while the frequency domain attention between parts is treated as edges to construct a part-level Fourier fully connected graph. This enables the graph Fourier learning module to jointly capture spatiotemporal dependencies in the frequency domain, while our adaptive frequency enhancement method further amplifies discriminative action features in a lightweight and robust fashion. Finally, a dual-branch action learning module featuring an auxiliary action prediction branch to assist the recognition branch is designed to enhance the understanding of sign language. Our experimental results show that the proposed PGF-SLR achieved relative improvements of 3.31%/3.70% and 2.81%/7.33% compared to SOTA methods on the dev/test sets of the PHOENIX14 and PHOENIX14-T datasets. It also demonstrated highly competitive recognition performance on the CSL-Daily dataset, showcasing strong generalization while reducing computational costs in both offline and online settings. Full article

Leaf area index (LAI) is a crucial parameter reflecting the crown structure of the grassland. Accurately obtaining LAI is of great significance for estimating carbon sinks in grassland ecosystems. However, spectral noise interference and pronounced spatial heterogeneity within vegetation canopies constitute significant impediments to achieving high-precision LAI retrieval. This study used hyperspectral sensor mounted on an unmanned aerial vehicle (UAV) to estimate LAI in a typical grassland, Hulun Lake Basin. Multiple machine learning (ML) models were constructed to reveal a relationship between hyperspectral data and grassland LAI using two input datasets, namely spectral transformations and vegetation indices (VIs), while SHAP (SHapley Additive ExPlanation) interpretability analysis was further employed to identify high-contribution features in the ML models. The analysis revealed that grassland LAI has good correlations with the original spectrum at 550 nm and 750 nm–1000 nm, first and second derivatives at 506 nm–574 nm, 649 nm–784 nm, and vegetation indices including the triangular vegetation index (TVI), enhanced vegetation index 2 (EVI2), and soil-adjusted vegetation index (SAVI). In the models using spectral transformations and VIs, the random forest (RF) models outperformed other models (testing R² = 0.89/0.88, RMSE = 0.20/0.21, and RRMSE = 27.34%/28.98%). The prediction error of the random forest model exhibited a positive correlation with measured LAI magnitude but demonstrated an inverse relationship with quadrat-level species richness, quantified by Margalef’s richness index (MRI). We also found that at the quadrat level, the spectral response curve pattern is influenced by attributes within the quadrat, like dominant species and vegetation cover, and that LAI has positive relationship with quadrat vegetation cover. The LAI inversion results in this study were also compared to main LAI products, showing a good correlation (r = 0.71). This study successfully established a high-fidelity inversion framework for hyperspectral-derived LAI estimation in mid-to-high latitude grasslands of the Hulun Lake Basin, supporting the spatial refinement of continental-scale carbon sink models at a regional scale. Full article

Rice, wheat, and maize are important food grains consumed by most of the population in Asian countries (like India, Japan, Singapore, Malaysia, China, and Thailand). These crops’ production is affected by biotic and abiotic factors that cause diseases in several parts of the crops (including leaves, stems, roots, nodes, and panicles). A severe infection affects the growth of the plant, thereby undermining the economy of a country, if not detected at an early stage. This may cause extensive damage to crops, resulting in decreased yield and productivity. Early safeguarding methods are overlooked because of farmers’ lack of awareness and the variety of crop diseases. This causes significant crop damage and can consequently lower productivity. In this manuscript, a lightweight vision transformer (MaxViT) with 814.7 K learnable parameters and 85 layers is designed for classifying crop diseases in paddy and wheat. The MaxViT DNN architecture consists of a convolutional block attention module (CBAM), squeeze and excitation (SE), and depth-wise (DW) convolution, followed by a ConvNeXt module. This network architecture enhances feature representation by eliminating redundant information (using CBAM) and aggregating spatial information (using SE), and spatial filtering by the DW layer cumulatively enhances the overall classification performance. The proposed model was tested using a paddy dataset (with 7857 images and eight classes, obtained from local paddy farms in Lalgudi district, Tiruchirappalli) and a wheat dataset (with 5000 images and five classes, downloaded from the Kaggle platform). The model’s classification performance for various diseases has been evaluated based on accuracy, sensitivity, specificity, mean accuracy, precision, F1-score, and MCC. During training and testing, the model’s overall accuracy on the paddy dataset was 99.43% and 98.47%, respectively. Training and testing accuracies were 94% and 92.8%, respectively, for the wheat dataset. Ablation analysis was carried out to study the significant contribution of each module to improving the performance. It was found that the model’s performance was immune to the presence of noise. Additionally, there are a minimal number of parameters involved in the proposed model as compared to pre-trained networks, which ensures that the model trains faster. Full article

Large synchronous motors are typically used to drive various load equipment, such as reciprocating compressors. Due to the continuous oscillation of the load, the pulsating torque acting on the main shaft of the synchronous motor will continuously vary with the load changes. This leads to forced oscillations during the dynamic stable operation of the unit, subsequently causing severe problems such as overheating, noise, and failures. Moreover, the rotor length of large synchronous motors is generally greater than the rotor diameter, giving the rotor certain flexible characteristics. During a motor’s operation, it is necessary to cross the first-order critical speed, making resonance highly likely to occur. Therefore, the analysis of dynamic characteristics of large synchronous motors is particularly important. This study investigates the dynamic characteristics of a 7800 kW-18P large synchronous motor rotor system through comprehensive theoretical and experimental analyses. The research encompasses three key aspects: (1) modal analysis comparing fan-equipped and fan-free configurations, (2) harmonic response evaluation, and (3) critical speed determination under concentrated mass conditions. Experimental validation was performed via impact hammer testing, with measured natural frequencies showing a strong correlation with simulated results for the magnetic pole core assembly. The findings not only confirm the operational speed validity but also establish a reliable foundation for the subsequent structural optimization of high-power synchronous machines. Full article

Although contact-based vibration signal methods for mechanical equipment fault diagnosis demonstrate superior performance, their practical deployment faces significant limitations. In contrast, acoustic signals offer notable deployment flexibility due to their non-contact nature. However, acoustic diagnostic methods are susceptible to environmental noise interference, and fault samples are typically scarce, leading to insufficient model generalization capability and robustness. To address this, this paper proposes an acoustic–vibration feature fusion strategy based on heterogeneous transfer learning, further integrated with a knowledge distillation framework. By doing so, it aims to achieve efficient transfer of vibration diagnostic knowledge to acoustic models. In the proposed approach, a teacher model learns diagnostic knowledge from highly reliable vibration signals and uses this to guide the training of a student model on acoustic signals. This process significantly enhances the diagnostic capability of the acoustic-based student model. Experimental studies conducted on a custom-built test rig and public datasets demonstrate that the proposed method exhibits excellent diagnostic accuracy and robustness under unseen working conditions. Full article

Diffuse correlation tomography (DCT) is an emerging technique for detecting diseases associated with localized abnormal perfusion from near-infrared light intensity temporal autocorrelation functions (g₂(τ)). However, a critical drawback of traditional reconstruction methods is the imbalance between optical measurements and the voxels to be reconstructed. To address this issue, this paper proposes Conv-TransNet, a convolutional neural network (CNN)–Transformer hybrid model that directly maps g₂(τ) data to blood flow index (BFI) images. For model training and testing, we constructed a dataset of 18,000 pairs of noise-free and noisy g₂(τ) data with their corresponding BFI images. In simulation validation, the root mean squared error (RMSE) for the five types of anomalies with noisy data are 2.13%, 4.43%, 2.15%, 4.05%, and 4.39%, respectively. The MJR (misjudgment ratio)of them are close to zero. In the phantom experiments, the CONTRAST of the quasi-solid cross-shaped anomaly reached 0.59, with an MJR of 2.21%. Compared with the traditional Nth-order linearization (NL) algorithm, the average CONTRAST of the speed-varied liquid tubular anomaly increased by 0.55. These metrics also demonstrate the superior performance of our method over traditional CNN-based approaches. The experimental results indicate that the Conv-TransNet model would achieve more accurate and robust reconstruction, suggesting its potential as an alternative for blood flow imaging. Full article

This work focuses on time and frequency domain analyses of IMU-based orientation estimation algorithms, including indirect Kalman (IKF), Madgwick (MF), and complementary (CF) filters. Euler angles and quaternions are used for orientation representation. A 6-DoF IMU is attached to a 6-joint UR5e robotic arm, with the robot’s orientation serving as the reference. Robotic arm data is obtained via an RTDE interface and IMU data via a CAN bus. Test signals include pose sequences, which are big-amplitude, slowly changing signals used to evaluate stationary and low-dynamics responses in the time domain, and small-amplitude, fast-changing generalized binary noise (GBN) signals used to evaluate dynamic responses in the frequency domain. To prevent poor filters’ performance, their parameters are tuned. In the time domain, RMSE and MaxAE are calculated for roll and pitch. In the frequency domain, composite frequency response and coherence are calculated using the Ockier method. RMSEs are computed for response magnitude and coherence, and averaged equivalent time delay (AETD) is derived from the response phase. In the time domain, MF and CF show the best overall performance. In the frequency domain, they again perform similarly well. IKF consistently performs the worst in both domains but achieves the lowest AETD. Full article