The efficient removal of uric acid from blood via purification is a novel approach for the treatment of gout and hyperuricemia. In this study, a new polybutylene terephthalate (PBT) biomembrane with good uric acid adsorption capacity and desirable blood compatibility was prepared via radiation grafting of N-(4-vinylcarbonyloxyphenyl) acetamide (NVA) and acrylic acid (AA). The chemical structure and surface morphology of the modified membrane were characterized by Fourier transform infrared (FTIR) spectroscopy, X-ray photoelectron spectroscopy (XPS), and scanning electron microscopy (SEM). The wettability of the modified membrane was determined by measuring the water contact angle and water absorption rate. The results indicated that the modified membrane exhibited a maximum uric acid adsorption rate of (6.80±0.20)% with minimal impact on blood cell counts. This study provides a promising strategy for developing safe and efficient biomaterials for the treatment of gout and hyperuricemia.
In this research, novel rosin-based non-ionic Gemini surfactants (RNGSs) were synthesized from rosin derivatives, dehydroabietic acid (DAA) and dehydroabietylamine (DHAA), via addition and alkylation reactions. The structures of the synthesized surfactants were characterized by Fourier transform infrared (FTIR) spectroscopy and proton nuclear magnetic resonance (1H NMR) spectroscopy. The surface activity, aggregation property, and foaming property of RNGSs in aqueous solution were investigated by surface tension measurement and dynamic light scattering (DLS). Furthermore, the cloud point was investigated to evaluate the upper temperature limit for practical application of the surfactants. To evaluate the potential of RNGSs for enhanced oil recovery (EOR), their interfacial tension (IFT), static adsorption capacity, salt tolerance, and emulsion stability were examined. The lowest critical micelle concentration (CMC) of RNGSs was 8.51 × 10-6 mol/L, lower than that of the commonly used non-ionic surfactant OP-10 (1.29 × 10-4 mol/L). The cloud point of RNGSs reached 72.6 ℃. Furthermore, the tolerance to NaCl and CaCl2 was determined to be 300 and 10 g/L, respectively. The findings indicate that the synthesized surfactants possess the potential for application in EOR.
This study explores the effects of the substrate temperature (25-200 ℃) on SiO2-like thin films prepared via atmospheric-pressure plasma jet. Films deposited at elevated temperatures exhibited a reduced thickness (from 242 to 156 nm) but enhanced mechanical and chemical performance. The pencil hardness improved from grade B to 4H. The adhesion reached class 0 (no detachment) at temperatures of 100 ℃ and above. X-ray photoelectron spectroscopy (XPS) confirmed carbon reduction and silicon dioxide (SiO2) enrichment when the substrate temperature increased from 25 to 200 ℃, indicating intensified precursor decomposition. Hydrochloric acid corrosion testing (3.0 h immersion) revealed that the films deposited at 200 ℃ exhibited improved corrosion resistance, with a mass loss of 6.53% compared to 7.01% for the bare substrates. The results indicate that the substrate temperature is a critical process parameter for preparing protective films applied in corrosive environments.
When preparing biaxially oriented polyethylene terephthalate (BOPET) films through sequential biaxial stretching, bowing distortion occurs within the film plane, leading to inhomogeneous optical and mechanical properties between the edge and central regions of the films. This quality degradation renders the films unsuitable for optical applications. Considering that polyethylene terephthalate (PET) macromolecules are in a rubbery state and have crystalline physical crosslinking points when the film is stretched above its glass transition temperature Tg but below the melting point Tm, this study establishes a visco-hyperelastic model. The model parameters are obtained by numerically fitting the experimental results of uniaxial stretching and stress relaxation tests. These fittings yield the second-order coefficient of the Mooney-Rivlin polynomial (describing hyperelasticity) and the third-order coefficient of the Prony series (describing viscosity). Then, ANSYS finite element simulations of the tentering process of PET films are carried out with the isotropic elastic, the orthotropic elastic, the plastic, and the visco-hyperelastic material models, respectively. The simulated bowing distortion obtained from the visco-hyperelastic model shows better agreement with experimental results, demonstrating that this model can accurately capture the mechanical response behavior of PET films during biaxial stretching.
To investigate the differences in the carding effect between fully fixed flat carding machines and movable flat carding machines, this study employed the Fluent software to conduct numerical simulations of airflow velocity distributions between the cylinder and flat of both machine types, with particular focus on analyzing the airflow patterns near microscopic needle teeth. The simulation results were further verified by practical carding and spinning experiments of multi-fibers, namely diacetate, viscose, heat-shrinkable acrylic, micro-denier acrylic and wool fibers. The results demonstrate that the fully fixed flat carding machine exhibits a more stable airflow boundary layer near the needle teeth, which enhances the carding effect as evidenced by improved yarn quality parameters.
The development of artificial intelligence has given rise to the paradigm of machine learning as a service, enabling users to either train the requisite models or employ pre-existing models to make predictions based on the training data that users provide. Despite various measures taken to prevent privacy leakage in machine learning, such as data deletion and anonymization, model inversion attack (MIA) remains capable of inferring sensitive information from user data to a certain extent. This paper reviews the current state of research on black-box MIA, with a particular focus on confidence-based MIA and label-based MIA. It further analyzes the MIA methodologies applied to the emerging modes such as text and audio in recent years, filling the gap in the current review of such modes. Finally, the paper discusses the future challenges and research directions for black-box MIA.
Mobile edge computing (MEC) plays a vital role in supporting communication, computation, and resource management within the maritime industry. It has become an essential element in the advancement of maritime technology. This study adopts a space-air-sea integrated MEC network, in which high-altitude platforms (HAPs) and low-earth orbit (LEO) satellites powered by renewable energy assist unmanned surface vehicles (USVs) in task offloading. The goal of this study is to minimize the total energy consumption of the system subject to a controllable task backlog queue. The Lyapunov optimization method is adopted to transform the original stochastic optimization problem into a deterministic optimization problem, thereby ensuring the stability of the system. Furthermore, an algorithm based on convex optimization is proposed, which is specifically designed to optimize transmission power allocation and channel allocation. The simulation results verify that the proposed scheme can significantly reduce the total energy consumption of the system compared with other benchmark schemes.
The traditional distributed video compressive sensing (DVCS) system effectively reduces transmission costs and encoder complexity for wireless visual sensor networks (WVSN), but struggles with dynamic scenes involving rapid motion or complex textures. Their reliance on temporal correlation and skipping causes reconstruction artifacts and loss of structural details, especially at high skip ratios. This paper proposes the motion-texture joint (MTJ) algorithm to enhance DVCS performance, where the motion module first preliminarily screens those candidate skip-blocks using the sum of absolute differences (SAD) and then generates more robust side information by combining frame interpolation with motion vectors for redundant-content identification and skip-block decision. Further, a texture feature module based on local binary pattern (LBP) analysis jointly exploits LBP texture features, variance textures, and gradient energy textures from adjacent frames, which improves inter-frame correlations. On this basis, an MTJ adaptive weight module dynamically allocates coefficients to motion information and texture information according to their spatio-temporal contributions and controls a skip ratio of DVCS. Experiments on standard video sequences demonstrate that the MTJ algorithm achieves an improvement in average peak signal-to-noise ratio (PSNR) of 0.12 dB over state-of-the-art uniform-reference threshold-dynamic skipping (UTS) with minimal encoder complexity increase. Especially at a 70% skip ratio, the maximum PSNR improvement over UTS reaches up to 0.20 dB on the Foreman sequence. MTJ algorithm effectively balances reconstruction quality and computational complexity, making it suitable for WVSN applications.
Video anomaly detection (VAD) aims to identify snippets or events that deviate from normal behavior in video sequences. With the growing demand for public safety and surveillance, this field has developed rapidly. In recent years, weakly supervised VAD (WSVAD) has attracted increasing attention. However, its performance remains limited due to the challenges of multiple instance learning (MIL), including high dataset complexity and substantial video noise, which severely hinder model learning. To address these issues, we propose a WSVAD method guided by anomaly representations. Specifically, our method leverages anomaly-related representations to correct biases in the learning process. It extracts key snippet and anomaly category representations through the key snippet learning module and anomaly classification module, respectively. These two types of representations are then embedded via an anomaly-guided attention mechanism. Extensive experiments on the ShanghaiTech and UCF-Crime datasets demonstrate that the proposed method significantly improves detection accuracy and achieves competitive performance.
Although the prime attribute problem is nondeterministic polynomial-time complete (NP-complete), the difficulty of determining whether an attribute is prime is related to how the attribute appears in the set of functional dependencies (FDs). The prime attribute problem is extensively studied based on the occurrence of an attribute in the set of FDs. First, in a relation scheme R(U, F), the attribute set U is partitioned into four distinct subsets according to where attributes in U appear in F: U1 (attributes appearing exclusively on the left-hand sides of FDs in F), U2 (attributes appearing exclusively on the right-hand sides of FDs in F), U3 (attributes appearing on both sides of FDs in F), and U4 (attributes not appearing in any FD in F). Second, the occurrence patterns of attributes in the set of FDs are mapped to an F-based derivation tree (F-based DT) forest, and the features of attributes in U1, U2, U3 or U4 are shown in the F-based DT forest. Then, an algorithm for recognizing prime attributes based on F-based DT is proposed. Finally, the following conclusions are obtained: when an attribute belongs to U3, the prime attribute problem is NP-complete; when an attribute belongs to U1, U2, or U4, the problem is in polynomial-time complexity class (P-class), meaning it can be efficiently solved in polynomial time. Compared with traditional methods, the proposed algorithm can simplify the process for recognizing prime attributes significantly, and improve database normalization.
Whether the lunar landing spacesuit can accurately recognize the astronaut’s gait and provide timely buffering upon lower limb impact is a critical factor in ensuring the safety of extravehicular operations. This study addresses this issue by segmenting the human gait phases and collecting gait data in a simulated lunar surface low-gravity environment. Z-score normalization and Gaussian filtering are applied for data preprocessing. The self-organizing map (SOM)-K-means clustering algorithm is adopted to establish the gait recognition model, and the model is then evaluated. Experimental results and comparisons with other clustering algorithms validate the effectiveness of the SOM-K-means algorithm for gait recognition and the generalizability of the model. The results demonstrate that the model achieves satisfactory clustering performance and accurate gait recognition, providing technical support for the design of high-performance lunar landing spacesuits.
In response to the challenges of radiation source localization in shielded environments, a combination-driven algorithm is proposed based on an improved simultaneous perturbation stochastic approximation (SPSA) framework. The proposed algorithm enhances the convergence performance by optimizing one-sided orthogonal perturbation vectors and incorporating a dynamic step-size adjustment strategy. To address path infeasibility in complex environments, a goal relocation mechanism based on breadth-first search (BFS) is introduced. For complex multi-space scenarios with shielding structures, an active exploration strategy is employed to proactively investigate unobserved regions. To evaluate the performance of the proposed algorithm in terms of convergence performance, obstacle adaptability, and exploration capacity, three simulation environments are set up in Gazebo, which respectively represent the following environments: open-field environment, complex environment and radiation shielding environment. The experiments demonstrate that the proposed algorithm outperforms the standard SPSA algorithm and particle filter algorithm in terms of convergence performance, while also exhibiting robust search capability in complex environments. Furthermore, when confronted with shielded environments, the proposed algorithm demonstrates the capability of actively exploring unknown areas within a relatively short period after being obstructed.
To enhance the accuracy of weld region identification under structured light vision guidance while maintaining method robustness, this paper proposes an adaptive weld identification method fusing local morphological and spatial position features. First, a fused feature space for point cloud data is constructed based on curvature estimation and spatial distance calculation, and the Gaussian mixture model (GMM) is adopted for unsupervised iterative learning of point cloud data to realize the extraction of feature points in the weld region. Second, combining with the hierarchical density-based spatial clustering of applications with noise (HDBSCAN) and voxel connectivity block analysis, the precise segmentation and boundary refinement optimization of the weld region are completed. Finally, experimental results demonstrate that this method achieves a precision of 99.38%, a recall of 88.41%, and an F1 score of 93.57% in identifying V-groove regions in circumferential welds, outperforming comparison methods such as random sample consensus combined with density-based spatial clustering of applications with noise (RANSAC-DBSCAN). Furthermore, the boundary localization error is less than 0.3 mm; parameter tests verify that the proposed method has low parameter sensitivity and strong stability, which meets practical production requirements.
The problem of prescribed-time group-bipartite leader-follower consensus for second-order nonlinear multi-agent systems (MASs) is investigated over a signed communication topology. First, by selecting a time-varying function as the control gain, we propose a novel control protocol for the case where each sub-topology is structurally balanced. By introducing an additional gain parameter into the controller, we simplify the sufficient conditions for the stability of the closed-loop system, and the calculations of parameters are more straightforward. Second, by incorporating leader dynamics into the controller, we relax the topological constraints and only require the overall topology of the MAS to be structurally balanced, and propose a novel consensus protocol based on neighbor information and leader dynamics. Based on matrix theory and Lyapunov stability theory, we derive sufficient conditions for achieving consensus. Finally, two numerical simulation examples are provided to validate the effectiveness of the two control protocols.
To address the increasingly prominent issue of high-frequency traffic noise pollution and to overcome the limitation of narrow sound absorption bandwidth of single-layer micro-perforated panels (MPPs), this study presents a systematic investigation into the sound-absorbing structure of multi-layer series MPPs. First, based on Ma’s MPP theory, theoretical sound absorption models for multi-layer structures were derived by using both the acoustic-electrical analogy method and the transfer matrix method. Then, the sound absorption coefficient curves of both methods were compared through COMSOL simulations, confirming the superiority of the transfer matrix method in the analysis of multi-layer structures. Finally, ant colony algorithm (ACA) was introduced to optimize the structural parameters of the multi-layer series MPPs, aiming to broaden the sound absorption bandwidth to match the high-frequency traffic noise spectrum. The results demonstrate that the transfer matrix method outperforms the acoustic-electrical analogy method in predicting the sound absorption characteristics of multi-layer series MPPs. Furthermore, the optimized structure exhibits significantly improved sound absorption performance in the frequency range of 100-4 000 Hz, with MPPs’ effectiveness increasing with the number of layers. This study would provide a valuable theoretical foundation and practical design references for broad-frequency traffic noise control.