Search Results (1,212)

Unmanned aerial vehicles (UAVs) are widely used in surveillance and combat for their efficiency and autonomy, whilst complex, dynamic environments challenge the modeling of inter-agent relations and information transmission. This research proposes a novel UAV tactical choice-making algorithm utilizing graph neural networks to tackle these challenges. The proposed algorithm employs a graph neural network to process the observed state information, the convolved output of which is then fed into a reconstructed critic network incorporating a Laplacian convolution kernel. This research first enhances the accuracy of obtaining unstable state information in hostile environments. The proposed algorithm uses this information to train a more precise critic network. In turn, this improved critic network guides the actor network to make decisions that better meet the needs of the battlefield. Coupled with a policy transfer mechanism, this architecture significantly enhances the decision-making efficiency and environmental adaptability within the multi-agent system. Results from the experiments show that the average effectiveness of the proposed algorithm across the six planned scenarios is 97.4%, surpassing the baseline by 23.4%. In addition, the integration of transfer learning makes the network convergence speed three times faster than that of the baseline algorithm. This algorithm effectively improves the information transmission efficiency between the environment and the UAV and provides strong support for UAV formation combat. Full article

In some areas, such as Gansu in China and Texas in the USA, lots of wind power bases are located far away from load centers. Transmitting large amounts of wind power to load centers through long transmission lines will lead to wind turbines being integrated into a weak grid, which decreases the stability limits of wind turbines. To solve this problem, this study investigates the stability limits of a Doubly Fed Induction Generator (DFIG) connected to a weak grid in a DC-link voltage control timescale. To start with, a model of the DFIG in a DC-link voltage control timescale is presented for stability limit analysis, which facilitates profound physical understanding. Through steady-state stability analysis based on sensitivity evaluation, it is found that the critical factor restricting the stability limit of the DFIG connected to a weak grid is ∂P_e/∂ (−i_rd), changing from positive to negative. As ∂P_e/∂ (−i_rd) reaches zero, the system reaches its stability limit. Furthermore, by considering control loop dynamics and grid strength, the stability limit of the DFIG is investigated based on eigenvalue analysis with multiple physical scenarios. The results of root locus analysis show that, when the DFIG is connected to an extremely weak grid, reducing the bandwidth of the PLL or increasing the bandwidth of the AVC with equal damping can increase the stability limit. The aforesaid theoretical analysis is verified through both time domain simulation and physical experiments. Full article

Facial expression plays an important role in human–computer interaction and affective computing. However, existing expression recognition methods cannot effectively capture multi-scale structural details contained in facial expressions, leading to a decline in recognition accuracy. Inspired by the multi-scale processing mechanism of the biological visual system, this paper proposes a wavelet-guided global–local feature aggregation network (WGGLFA) for facial expression recognition (FER). Our WGGLFA network consists of three main modules: the scale-aware expansion (SAE) module, which combines dilated convolution and wavelet transform to capture multi-scale contextual features; the structured local feature aggregation (SLFA) module based on facial keypoints to extract structured local features; and the expression-guided region refinement (ExGR) module, which enhances features from high-response expression areas to improve the collaborative modeling between local details and key expression regions. All three modules utilize the spatial frequency locality of the wavelet transform to achieve high-/low-frequency feature separation, thereby enhancing fine-grained expression representation under frequency domain guidance. Experimental results show that our WGGLFA achieves accuracies of 90.32%, 91.24%, and 71.90% on the RAF-DB, FERPlus, and FED-RO datasets, respectively, demonstrating that our WGGLFA is effective and has more capability of robustness and generalization than state-of-the-art (SOTA) expression recognition methods. Full article

Background/Objectives: Asprosin is the endogenous ligand of the olfactory Olfr734 receptor linked to MASLD and glucose metabolism. Despite the involvement of asprosin in these processes, little has been published on the specific role of Olfr734 in liver function. The aim of this work is therefore to study the specific role of the olfactory Olfr734 receptor in MASLD and glucose metabolism. Methods: To achieve this objective, we performed a genetic inhibition specifically to inhibit Olfr734 in the livers of male mice. We then studied the progression of MASLD in DIO mice. In addition, we studied the glucose metabolism in hypoglycemia states and postprandial glucose production in standard diet-fed mice. Finally, analyses of liver biopsies from patients with obesity and with or without T2DM were conducted. Results: We found that hepatic Olfr734 levels vary according to changes in nutritional status and its knockdown effect in the liver is to increase the hepatic lipid content in DIO mice. Our results also showed that OLFR734 expression is involved in the adaptive response in terms of glucose production to nutrient availability. Finally, the hepatic human Olfr734 ortholog named OR4M1 has been observed to be at significantly higher levels in male patients with T2DM. Conclusions: This study increases understanding of the mechanisms by which the modulation of Olfr734 expression affects liver function. Full article

As a critical precision transmission element in numerical control (NC) machines, the linear rolling guide (LRG) suffers from surface wear degradation, which significantly impairs machining accuracy and operational reliability. Despite its importance, effective identification methods for LRG degradation remain limited. In this study, a hybrid approach combining multi-scale fuzzy entropy (MFE) with a gray wolf-optimized random forest (GWO-RF) algorithm was proposed to identify the surface wear state of the LRG. Preload degradation and vibration signals were collected at three surface wear stages throughout the LGR’s service life. The vibration signals were decomposed and reconstructed using complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), followed by multi-scale fuzzy entropy analysis of the reconstructed signals. After dimensionality reduction via kernel principal component analysis (KPCA), the processed features were fed into the GWO-RF model for classification. Experimental results demonstrated a recognition accuracy of 97.9%. Full article

The Internet of Vehicles (IoV) presents a vast opportunity for optimised traffic flow, road safety, and enhanced usage experience with the influence of Federated Learning (FL). However, the distributed nature of IoV networks creates certain inherent problems regarding data privacy, security from adversarial attacks, and the handling of available resources. This paper introduces Fed-DTB, a new dynamic trust-based framework for FL that aims to overcome these challenges in the context of IoV. Fed-DTB integrates the adaptive trust evaluation that is capable of quickly identifying and excluding malicious clients to maintain the authenticity of the learning process. A performance comparison with previous approaches is shown, where the Fed-DTB method improves accuracy in the first two training rounds and decreases the per-round training time. The Fed-DTB is robust to non-IID data distributions and outperforms all other state-of-the-art approaches regarding the final accuracy (87–88%), convergence rate, and adversary detection (99.86% accuracy). The key contributions include (1) a multi-factor trust evaluation mechanism with seven contextual factors, (2) correlation-based adaptive weighting that dynamically prioritises trust factors based on vehicular conditions, and (3) an optimisation-based client selection strategy that maximises collaborative reliability. This work opens up opportunities for more accurate, secure, and private collaborative learning in future intelligent transportation systems with the help of federated learning while overcoming the conventional trade-off of security vs. efficiency. Full article

This study proposes an improved super-twisting sliding mode (STSM) control method for a brushless doubly fed induction generator (BDFIG) used in standalone ship shaft power generation systems. Focusing on the problem of the low tracking accuracy of the power winding (PW) voltages caused by the parameter perturbation of BDFIG systems, a mismatched uncertain model of the BDFIG is constructed. Additionally, an improved STSM control method is proposed to address the power load variation and compensate for the mismatched uncertainty through virtual control technology. Based on the direct vector control of the control winding (CW), the proposed method ensured that the voltage amplitude error of the power winding could converge to the equilibrium point rather than the neighborhood. Finally, in the experimental investigation of the BDFIG-based ship shaft independent power system, the dynamic performance in the startup and power load changing conditions were analyzed. The experimental results show that the proposed improved STSM controller has a faster dynamic response and higher steady-state accuracy than the proportional integral control and the linear sliding mode control, with strong robustness to the mismatched uncertainties caused by parameter perturbations. Full article

Federated learning (FL), which enables collaborative learning across distributed nodes, confronts a significant heterogeneity challenge, primarily including resource heterogeneity induced by different hardware platforms, and statistical heterogeneity originating from non-IID private data distributions among clients. Neural architecture search (NAS), particularly one-shot NAS, holds great promise for automatically designing optimal personalized models tailored to such heterogeneous scenarios. However, the coexistence of both resource and statistical heterogeneity destabilizes the training of the one-shot supernet, impairs the evaluation of candidate architectures, and ultimately hinders the discovery of optimal personalized models. To address this problem, we propose a heterogeneity-aware personalized federated NAS (HAPFNAS) method. First, we leverage lightweight knowledge models to distill knowledge from clients to server-side supernet, thereby effectively mitigating the effects of heterogeneity and enhancing the training stability. Then, we build random-forest-based personalized performance predictors to enable the efficient evaluation of candidate architectures across clients. Furthermore, we develop a model-heterogeneous FL algorithm called heteroFedAvg to facilitate collaborative model training for the discovered personalized models. Comprehensive experiments on CIFAR-10/100 and Tiny-ImageNet classification datasets demonstrate the effectiveness of our HAPFNAS, compared to state-of-the-art federated NAS methods. Full article

Chronic ethanol use can lead to alcoholic cardiomyopathy (ACM), while the impact on the molecular and cellular aspects of the myocardium is unclear. Accordingly, male Sprague-Dawley rats were exposed to an ethanol-containing diet for 16 weeks and compared with a control group that was fed an isocaloric diet. Histological measurements from H&E slides revealed no significant differences in cell size. A proteomic approach revealed that alcohol exposure leads to enhanced mitochondrial lipid metabolism, and electron microscopy revealed impairments in mitochondrial morphology/density. Cardiac myosin purified from the hearts of ethanol-exposed animals demonstrated a 15% reduction in high-salt ATPase activity, with no significant changes in the in vitro motility and low-salt ATPase or formation of the super-relaxed (SRX) state. A protein carbonyl assay indicated a 20% increase in carbonyl incorporation, suggesting that alcohol may impact cardiac myosin through oxidative stress mechanisms. In vitro oxidation of healthy cardiac myosin revealed a dramatic decline in ATPase activity and in vitro motility, demonstrating a link between myosin protein oxidation and myosin mechanochemistry. Collectively, this study suggests alcohol-induced metabolic remodeling may be the initial insult that eventually leads to defects in the contractile machinery in the myocardium of ACM hearts. Full article

Motor Imagery (MI) based Brain Computer Interfaces (BCIs) have promising applications in neurorehabilitation for individuals who have lost mobility and control over parts of their body due to brain injuries, such as stroke patients. Accurately classifying MI tasks is essential for effective BCI performance, but this task remains challenging due to the complex and non-stationary nature of EEG signals. This study aims to improve the classification of left and right-hand MI tasks by utilizing high-resolution time-frequency features extracted from EEG signals, enhanced with deep learning-based data augmentation techniques. We propose a novel deep learning framework named the Generalized Wavelet Transform-based Deep Convolutional Network (GDC-Net), which integrates multiple components. First, EEG signals recorded from the C3, C4, and Cz channels are transformed into detailed time-frequency representations using the Generalized Morse Wavelet Transform (GMWT). The selected features are then expanded using a Deep Convolutional Generative Adversarial Network (DCGAN) to generate additional synthetic data and address data scarcity. Finally, the augmented feature maps data are subsequently fed into a hybrid CNN-LSTM architecture, enabling both spatial and temporal feature learning for improved classification. The proposed approach is evaluated on the BCI Competition IV dataset 2b. Experimental results showed that the mean classification accuracy and Kappa value are 89.24% and 0.784, respectively, making them the highest compared to the state-of-the-art algorithms. The integration of GMWT and DCGAN significantly enhances feature quality and model generalization, thereby improving classification performance. These findings demonstrate that GDC-Net delivers superior MI classification performance by effectively capturing high-resolution time-frequency dynamics and enhancing data diversity. This approach holds strong potential for advancing MI-based BCI applications, especially in assistive and rehabilitation technologies. Full article

The research on the effect of feed with probiotic additives on the growth and functional state of young steelhead salmon (Salmo gairdneri) is presented in this study. For the first time, target strains selected not only by antagonism to pathogens but also by their ability to produce lytic enzymes or secondary metabolites with antioxidant activity were used to create probiotic preparations for aquaculture. This study presents findings showing that groups of fish fed probiotic feeds showed an improved growth performance and higher survival rate compared to the control. It was noted that the weights of fish in the first variant and the second variant of the experiment were higher by 8.8% and 6.8%, respectively. This research showed that juvenile steelhead salmon reared with probiotic-supplemented feeds had an improved ability to survive in high salinity and sublethal temperatures. This indicates that probiotics may play a significant role in enhancing the adaptive system of fish. Full article

The Super-Twisting Sliding Mode Controller (STSMC) is regarded as one of the most straightforward and most practical nonlinear control systems, due to its ease of application in industrial systems. Its application helps minimize the chattering problem and significantly improves the resilience of the system. This controller possesses multiple deficiencies and issues, as its use does not promote the expected improvement of systems. To overcome these shortcomings and optimize the efficiency and performance of this technique, a new method is suggested for the super-twisting algorithm (STA). This study proposes and uses a new STA approach, named the fast super-twisting algorithm (FSTA), utilized the conventional IFOC technique to mitigate fluctuations in torque, current, and active power. The results from this suggested the IFOC-FSTA method are compared with those of the traditional SMC and STA methods. The results obtained from this study demonstrate that the suggested method, which is based on FSTA, has outperformed the traditional method in terms of ripple ratio and response dynamics. This demonstrates the robustness of the proposed approach to optimize the generator performance and efficiency in the double-fed induction generator-based wind system. Full article

Background/Objectives: Ischemic postconditioning (IP) is a well-established intervention that mitigates this damage by activating endogenous cardioprotective pathways. However, the presence of comorbidities such as dyslipidemia can disrupt these protective mechanisms and abolish the infarct-sparing effect typically induced by IP. In this context, identifying pharmacological strategies to restore cardioprotection is of clinical relevance. This study aimed to evaluate whether N-acetylcysteine (NAC), a glutathione precursor with antioxidant properties, can restore the infarct-limiting effect of IP compromised by HFD-induced oxidative stress. Methods: Male mice were fed a control diet (CD) or HFD for 12 weeks. NAC (10 mM) was administered in drinking water for 3 weeks before ex vivo myocardial ischemia/reperfusion (I/R) injury (30 min ischemia/60 min reperfusion). In IP groups, six cycles of brief I/R were applied at the onset of reperfusion. Infarct size, ventricular function, redox status (GSH/GSSG), lipid profile, and histology were evaluated. Results: NAC improved the lipid profile (HDL/non-HDL ratio) and enhanced the infarct-sparing effect of IP in CD-fed mice. In HFD-fed mice, NAC restored the efficacy of IP, significantly reducing infarct size (HFD-I/R-NAC: 39.7 ± 4.5% vs. HFD-IP-NAC: 26.4 ± 2.0%, p < 0.05) without changes in ventricular function. The ratio of oxidized/reduced glutathione (GSSG/GSH) is depicted. Under basal conditions, the hearts of mice fed an HFD exhibited a shift towards a more oxidized state compared to the control diet CD group. In the I/R protocol, a significant shift towards a more oxidized state was observed in both CD and HFD-fed animals. In the IP protocol, the GSSG/GSH ratio revealed a tendency to basal values in comparison to the I/R protocol. The analysis indicates that animals subjected to I/R and IP protocols in conjunction with NAC show a tendency to reach basal values, thus suggesting a potential for the reduction in ROS. Conclusions: NAC treatment mitigates oxidative stress and restores the cardioprotective effect of ischemic postconditioning in a model of early-stage atherosclerosis. These findings support NAC as a potential adjunct therapy to improve myocardial resistance to reperfusion injury under dyslipidemic conditions Full article

This study addresses the long-standing difficulty of predicting the remaining useful life (RUL) of rolling bearings from highly non-stationary vibration signals by proposing WaveAtten, a symmetry-aware deep learning framework. First, mirror-symmetric and bi-orthogonal Daubechies wavelet filters are applied to decompose each raw signal into multi-scale approximation/detail pairs, explicitly preserving the left–right symmetry that characterizes periodic mechanical responses while isolating asymmetric transient faults. Next, a bidirectional sparse-attention module reinforces this structural symmetry by selecting query–key pairs in a forward/backward balanced fashion, allowing the network to weight homologous spectral patterns and suppress non-symmetric noise. Finally, the symmetry-enhanced features—augmented with temperature and other auxiliary sensor data—are fed into a long short-term memory (LSTM) network that models the symmetric progression of degradation over time. Experiments on the IEEE PHM2012 bearing dataset showed that WaveAtten achieved superior mean squared error, mean absolute error, and R² scores compared with both classical signal-processing pipelines and state-of-the-art deep models, while ablation revealed a 6–8% performance drop when the symmetry-oriented components were removed. By systematically exploiting the intrinsic symmetry of vibration phenomena, WaveAtten offers a robust and efficient route to RUL prediction, paving the way for intelligent, condition-based maintenance of industrial machinery. Full article

Electric submersible progressive-cavity pumps (ESPCPs) deliver high lifting efficiency but are prone to failure in the high-temperature, high-pressure, and multiphase down-hole environment, leading to production losses and elevated maintenance costs. To achieve reliable condition recognition under these noisy and highly imbalanced data constraints, we fuse deep residual feature learning, ensemble decision-making, and generative augmentation into a unified diagnosis pipeline. A class-aware TimeGAN first synthesizes realistic minority-fault sequences, enlarging the training pool derived from 360 field records. The augmented data are then fed to a CNN backbone equipped with ResNet blocks, and its deep features are classified by a Random-Forest head (CNN-ResNet-RF). Across five benchmark architectures—including plain CNN, CNN-ResNet, GRU-based, and hybrid baselines—the proposed model attains the highest overall validation accuracy (≈97%) and the best Macro-F1, while the confusion-matrix diagonal confirms marked reductions in the previously dominant misclassification between tubing-leakage and low-parameter states. These results demonstrate that residual encoding, ensemble voting, and realistic data augmentation are complementary in coping with sparse, noisy, and class-imbalanced ESPCP signals. The approach therefore offers a practical and robust solution for the real-time down-hole monitoring and preventive maintenance of ESPCP systems. Full article