Search

In contrast to the traditional avatar creation pipeline which is a costly process, contemporary generative approaches directly learn the data distribution from photographs. While plenty of works extend unconditional generative models and achieve some levels of controllability, it is still challenging to ensure multi-view consistency, especially in large poses. In this work, we propose a network that generates 3D-aware portraits while being controllable according to semantic parameters regarding pose, identity, expression and illumination. Our network uses neural scene representation to model 3D-aware portraits, whose generation is guided by a parametric face model that supports explicit control. While the latent disentanglement can be further enhanced by contrasting images with partially different attributes, there still exists noticeable inconsistency in non-face areas when animating expressions. We solve this by proposing a volume blending strategy in which we form a composite output by blending dynamic and static areas, with two parts segmented from the jointly learned semantic field. Our method outperforms prior arts in extensive experiments, producing realistic portraits with vivid expression in natural lighting when viewed from free viewpoints. It also demonstrates generalization ability to real images as well as out-of-domain data, showing great promise in real applications.

We propose a novel multisection cable-driven soft robotic arm inspired by octopus tentacles along with a new modeling approach. Each section of the modular manipulator is made of a soft tubing backbone, a soft silicone arm body, and two rigid endcaps, which connect adjacent sections and decouple the actuation cables of different sections. The soft robotic arm is made with casting after the rigid endcaps are 3D-printed, achieving low-cost and convenient fabrication. To capture the nonlinear effect of cables pushing into the soft silicone arm body, which results from the absence of intermediate rigid cable guides for higher compliance, an analytical static model is developed to describe the relationship between the bending curvature and the cable lengths. The proposed model shows superior prediction performance in experiments over that of a baseline model, especially under large bending conditions. Based on the nonlinear static model, a kinematic model of a multisection arm is further developed and used to derive a motion planning algorithm. Experiments show that the proposed soft arm has high flexibility and a large workspace, and the tracking errors under the algorithm based on the proposed modeling approach are up to 52$\%$ smaller than those with the algorithm derived from the baseline model.

An efficient [3 + 2] cycloaddition reaction between in situgenerated nitrile imines from hydrazonoyl halides and vinylsulfonium salts is developed. The nitrile imines are demonstrated to be a new class of reaction partner for vinylsulfonium salts to conduct the [3 + 2] cycloaddition reaction. The process provides a concise and efficient method for the construction of pyrazole derivatives under mild reaction conditions with broad substrate scope, good product yields, and high regioselectivity.

We have successfully synthesized a series of bidentate ligands by utilizing 2-(trimethylsilyl)phenyl trifluorosulfonate as a precursor for the benzyl group. This method proceeded by inserting a polythiourea into the C═S π-bond, intramolecular ring proton migration, and ring opening. Salient features of this strategy are mild reaction conditions, a novel product structure, excellent stereochemistry, and a good functional group tolerance. Furthermore, a series of density functional theory calculations were performed to gain insights into the transfer mechanism.

•A new microfluidic in-vitro endothelial cell culture system (ECCS) was proposed.•Blood pressure, shear stress, and circumferential strain were accurately reproduced.•The coupling relationship among the above signals was accurately reproduced.•The area for cell study in this system had uniform stress and strain distribution.•The system had better biological properties and potential in mechanobiology study.

This study systematically investigates the flow behavior of 7050 aluminum (Al) alloy produced via spray forming and subjected to a proposed pretreatment through thermal compression tests. The experimental findings reveal that the peak stress increased with decreasing deformation temperature or increasing strain rates. The parameters in the Arrhenius constitutive model were determined and verified by experimental results for spray formed (SFed) 7050 Al alloy. Hot deformation activation energy of the SFed 7050 Al alloy (147.38 kJ/mol) is notably lower than that of cast Al–Zn–Mg–Cu–Zr alloys, which is attributed to factors such as fine grains, low segregation, pores, and dispersed Al3Zr induced by pretreatment. Furthermore, processing maps have been constructed at various strains, revealing optimal stable deformation conditions at 300–450 °C/0.001–0.15s−1and 425–450 °C/0.2–1s−1. Microstructural analysis of deformed samples highlights the critical role of dynamic recrystallization in influencing the stable and unstable areas of the alloy, which may be activated by high temperature and slow strain rate. The hot deformation mechanism of the SFed Al alloy is governed by interactions among η-Mg(Zn,Al,Cu)2precipitates, dispersed Al3Zr particles, dislocation structures, and (sub-)grain boundaries.

The Talos stent-graft has extended length to improve aortic remodeling, and distal porous design to decrease the rate of spinal cord ischemia (SCI). This study retrospectively analyzed its mid-term outcomes for uncomplicated type B aortic dissection in a multicenter study.

•Exposure to prior cytotoxic therapy in patients with CCUS independently accounts for inferior PFS and OS.•Patients with CCUS who have received cytotoxic therapy have distinct clinical features, which be considered a unique entity called t-CCUS.

van der Waals (vdWs) dielectrics are widely used in nanoelectronics to preserve the intrinsic properties of two-dimensional (2D) semiconductors. However, achieving aligned growth of 2D semiconductors and their direct utilization on original vdWs epitaxial dielectrics to avoid disorders poses significant challenges. Here, a hydromechanical strategy for aligned epitaxy of 2D materials on naturally occurring vdWs mica dielectrics is developed. By combining density functional theory with Lagrange’s group theorem, a quantitative criterion for 2D material epitaxy on 6-fold symmetric vdWs dielectrics is established. Moreover, the as-grown ultrathin Bi2O2Se-channeled field-effect transistor, with a hybrid dielectric layer, achieves a superior current on/off ratio (1.4 × 107) and high carrier mobility (22.4 cm2V−1S−1) by directly integrating as-grown 2D materials/vdWs dielectrics. This work provides a powerful methodological platform for aligned 2D material synthesis, alignment direction prediction, and intrinsic property investigation, laying the foundation for advanced electronics on as-grown 2D materials/vdWs dielectrics.

A novel three-component cyclization carbonylation reaction of iodoarene-tethered propargyl ethers with amine and CO is reported. This palladium-catalyzed cascade reaction undergoes a sequence of oxidative addition, unsaturated bond migration, carbonyl insertion, and nucleophilic attack to deliver the benzofuran skeleton. Both aromatic amines and aliphatic amines could proceed smoothly in this transformation under one atm of CO.

Impact craters are formed due to the high-speed collisions between small to medium-sized celestial bodies. Impact is the most significant driving force in the evolution of celestial bodies, and the impact craters provide crucial insights into the formation, evolution, and impact history of celestial bodies. In this paper, we present a detailed review of the characteristics of impact craters, impact crater remote sensing data, recognition algorithms, and applications related to impact craters. We first provide a detailed description of the geometric texture, illumination, and morphology characteristics observed in remote sensing data of craters. Then we summarize the remote sensing data and cataloging databases for the four terrestrial planets (i.e., the Moon, Mars, Mercury, and Venus), as well as the impact craters on Ceres. Subsequently, we study the advancement achieved in the traditional methods, machine learning methods, and deep learning methods applied to the classification, segmentation, and recognition of impact craters. Furthermore, based on the analysis results, we discuss the existing challenges in impact crater recognition and suggest some solutions. Finally, we explore the implementation of impact crater detection algorithms and provide a forward-looking perspective.

In the present work, a new surface mechanical strengthening (SMS) technique named stress ultrasonic rolling (SUR), which is ultrasonic rolling (UR) assisted with a uniaxial tensile elastic stress field, was reported and successfully conducted on 5 % wt TiB2/7050Al composite. Based on the experimental analysis and numerical simulation, the effects of SUR on surface integrity and fatigue performance of the composite were comprehensively investigated. Compared with ordinary UR, SUR further enhanced the fatigue strength at 106 cycles of the composite by 5 % and prolonged the fatigue life of the composite under the stress level of 532.8 MPa by 1.8 to 3.3 times, which are attributed to the comprehensive effect of excellent surface finish, larger and deeper compressive residual stress (RSc) layer, dispersed TiB2particles distribution in the surface, and superficial nanocrystalline structures. Besides, the assistance of uniaxial tensile elastic stress field did not increase the amount of surface plastic deformation, which indicates the RSc generated by SUR should have similar mechanical and thermal stability to that produced by ordinary UR treatment.

O-Acylhydroxylamine has been widely employed as an electrophilic amination reagent in transition-metal-catalyzed C–N coupling reactions, but its use as an electrophilic oxygen source has not been disclosed. Here, we report a Pd-catalyzed 1,2-oxyarylation of alkenes with O-acylhydroxylamines as an oxidant and an oxygen source for the first time. With simple amide as the monodentate directing group, this method features a broad substrate scope, good functional group tolerance, and mild conditions.

The copolymers of vinyl chloride (VC) with acrylates are widely used as special resins in multiple fields such as coatings, adhesives, and polymer processing aids. Nevertheless, the controllability in the copolymer composition is still a challenge owing to the big gap in reactivity ratio. In this paper, we conducted the nonaqueous copolymerization of VC with methyl acrylate (MA) in a mixed organic solvent (hexane/ethanol) without any suspending agent or emulsifier and intended to regulate the copolymer composition by dropping MA in a continuous mode. The promising results showed that a powdery vinyl chloride-co-methyl acrylate (PVMA) with ultralow number-average molecular weight (Mn, 9000–11?000) and submicrometer size (0.2–0.5 µm) is synthesized successfully, which is highly different from those of conventional aqueous polymerizations such as suspension and emulsion. More interestingly, varied PVMAx(xrepresents the actual mass fraction of MA in the copolymer) with relatively uniform and narrow dispersity in copolymer compositions (x= 1.7 ± 0.4, 3.2 ± 0.5, 5 ± 0.7, and 8.2 ± 0.4%) are obtained during the whole polymerization stage. In the batch process, the content of MA units in the copolymer decreased significantly with the increase of monomer conversion. Subsequently, the PVMA copolymers were employed as processing aids in rigid PVC and exhibited unique features. Stemming from its low Mnand the excellent compatibility with PVC matrix, the addition of PVMA can improve the plasticization efficiency significantly and thereby further enhance the mechanical performance. For instance, addition of 3 phr PVMA8.3in the rigid PVC system can shorten the plasticization time from 104 to 52 s, enhance the tensile strength from 65 to 74 MPa, and increase the elongation at break 2-fold in comparison to the controlled PVC. In summary, this study provides a novel polymerization technique and strategy for the preparation of a vinyl chloride based copolymer resin with a low Mnand further composition control.

We introduce switchable chemoselectivity strategies based on the hydrazone phosphaketene intermediate to synthesize three classes of 1,2,4-diazaphosphol derivatives. First, the five-membered heterocyclic P and O anion intermediates acted as nucleophilic agents in the selective construction of C–P and C–O bonds. Second, the phosphinidene served as a phosphorus synthon, allowing for the formation of C–P and C–N bonds. Finally, a stepwise mechanism, supported by DFT calculations, was invoked to explain the reaction selectivity.

High‐performance cyanate ester (CE) resins have attracted intensive interests due to low‐k and heat resistant properties. However, the extreme high curing temperature and inherent brittleness severely limit their practical application. Herein, novel biobased poly(N‐phenylmaleimide‐co‐limonene) (PML) microspheres were specially designed and synthesized by self‐stabilized precipitation polymerization, which could simultaneously serve as effective curing agents and toughening modifiers for 2,2‐bis(4‐cyanatophenyl) propane (BADCy). The high density of unreacted endocyclic vinyl groups in PML could effectively react with BADCy, leading to lower curing temperature and good interfacial compatibility. Benefitting from both the weakly polarizable flexible chains and the highly crosslinked semi‐interpenetrating network formed during the curing process, BADCy/PML resins exhibited highly enhanced toughness and reduced dielectric constant (ε). Specifically, the BADCy/PML resins showed minimum values of ε and dielectric loss of 2.61 and 0.0032 at 106Hz, much lower than BADCy resins (2.88, 0.0052). More importantly, a maximum impact strength of 15.6 kJ/m2was achieved at 10 wt% loading of PML, which was 86% higher than BADCy resin. In summary, this work developed a novel strategy to simultaneously improve toughness and dielectric properties of CE resins using bio‐based PML copolymer, which have great potential in the fields of electronics and information technology.

To compare the 30-day and long-term outcomes between patients with concomitant type B intramural hematoma and intimal disruption upon admission who underwent endovascular repair in the acute or subacute phases.

Herein, low molecular weight poly(vinyl chloride‐co‐n‐butyl acrylate‐co‐vinyl acetate) (PCBV) ternary copolymers with varied copolymer composition was employed as a novel polymeric lubricant in rigid PVC. The lubrication effect of PCBVs on the melting process and mechanical properties of the final PVC products were fully investigated. The results indicated that PCBV can serve either as an internal or as an external lubricant depending on its composition. PCBV‐45.5 terpolymer (PBA% ~ 45.5%) exhibits outstanding balance both internal and external lubricating roles. SEM‐EDS results show that addition PCBV‐45.5 form a film on the surface of PVC grains as similar as that adding both internal and external lubricants. Fusion characteristics and DMTA demonstrate that PCBV‐45.5 prolongs the fusion time (26% extension, 1.5 phr) and reduces the glass transition temperature (4.0 °C, 1.5 phr). Finally, the mechanical properties of rigid PVC have been greatly improved by using 1.5 phr PCBV‐45.5. In detail, the tensile strength increased from 54 to 59 MPa and the elongation at break increased from 10% to 30%, respectively. Meanwhile, the impact strength of PVC/PCBV‐45.5 is 3.6 times that of the blend without addition. The excellent performance of PCBV exhibited great potential as a polymeric lubricant candidate in rigid PVC composites.

Metastasis is the leading cause of death in patients with breast cancer. Detecting high-risk breast cancer, including micrometastasis, at an early stage is vital for customizing the right and efficient therapies. In this study, we propose an enzyme-free isothermal cascade amplification-based DNA logic circuit in situ biomineralization nanosensor, HDNAzyme@ZIF-8, for simultaneous imaging of multidimensional biomarkers in live cells. Taking miR-21 and Ki-67 mRNA as the dual detection targets achieved sensitive logic operations and molecular recognition through the cascade hybridization chain reaction and DNAzyme. The HDNAzyme@ZIF-8 nanosensor has the ability to accurately differentiate breast cancer cells and their subtypes by comparing their relative fluorescence intensities. Of note, our nanosensor can also achieve visualization within breast cancer organoids, faithfully recapitulating the functional characteristics of parental tumor. Overall, the combination of these techniques offers a universal strategy for detecting cancers with high sensitivity and holds vast potential in clinical cancer diagnosis.

The impressive mechanical properties of DNA molecules play important physiological and pathological roles, which arise from their intrinsically ordered and periodic double-helix structures. Nevertheless, bulk fabrication of DNA-based materials that inherit their nanoscale mechanical strength remains a grand challenge. Here, we developed a multiscale engineering approach to assemble with highly ordered DNA bulk fibers. We found that the involvement of metal ions enabled tunable assembly and orientation of DNA-metal ensembles ranging from molecules to nanobundles and to macroscale filaments. This strategy remarkably enhanced the mechanical performance of DNA fibers with the optimal tensile strength and toughness reaching to 513 MPa and 130 MJ m−3, respectively, which outperform even partially protein- and polymer-based fibers. We further demonstrated the high scalability of this method by continuous spinning of 1,000 m of robust fibers with 50 mg DNA, which shows great promise in translations of DNA-based materials in mechanobiology, bioelectronics, and beyond.

The rupture behavior of the modified 10Cr–1Mo steel multi-layer welded joint is determined by the fine-grain zones of the weld metal adjacent to the fusion line during the long-term creep test at 620 °C. The microstructures of multi-layer weld metal before and after the creep tests were characterized in detail, and its role in creep behavior was systematically investigated. Most grain boundaries of subgrains represented the low-angle boundaries in the weld metal adjacent to the fusion line both before and after the creep test. The widths of grains in the fine-grain zones were about 0.5–1 μm. The fracture morphology appeared as “wave” structure due to the cracking initiating from multi-layer grain boundaries in the fine-grain zones. Some W elements that melted into weld metal adjacent to the fusion line altered the thermodynamic and kinetic conditions of the Laves phase formation during long-term creep exposure. Laves phase particles mainly distributed along the grain boundaries due to the faster diffusion and segregation of Mo, W, and Si elements. Moreover, higher-density grain boundaries in the fine-grain zones led to easier nucleation and growth of Laves phase particles. Compared with other areas in the welded joint, the size of Laves phase particles in the fine-grain zones of the weld metal adjacent to the fusion line was the largest ones. The interface between Laves phase particles and the matrix acted as the nucleation site of creep micro-cavities. The creep micro-cavities grew up at the expense of fine-grain boundaries and even grew across the grain boundary deeply into adjacent grains, and then developed to cracks in the fine-grain zones.

Point cloud registration is crucial for various applications such as robotics and urban planning, yet it remains challenging due to issues such as noise, variable resolutions, and uncertainties in the initial pose. The normal distribution transform (NDT) algorithm has become a standard approach for registering point scans, but its application to large-scale scenes faces significant challenges due to inherent limitations. These include computational complexity, susceptibility to noise interference, limited adaptability to varying voxel resolutions, and substantial initial pose deviations. To address these issues, we propose the multiscale iterative NDT (MI-NDT) through a meticulously designed multiscale iterative optimization framework. The MI-NDT algorithm voxelizes the reference scan into multiscale voxels based on the optimal flatness ratio criterion. At each scale, the classic NDT algorithm computes the pose of the source scan. The pose obtained from coarse voxels serves as input for the next optimization step on finer voxels. This iterative refinement continues until the finest voxel level, where the resulting pose corresponds to the optimal translation parameters. The experimental results demonstrate the effectiveness of the MI-NDT algorithm in terms of registration accuracy and robustness. Comparative experiments with existing algorithms show significant improvements in accuracy, especially under varying voxel resolutions and noise levels. Moreover, the algorithm exhibits enhanced tolerance to initial pose deviations, enabling successful registration even with poor initial poses.

The SPiForest, a new isolation-based approach to outlier detection, constructs iTrees on the space containing all attributes by probability density-based inverse sampling. Most existing iForest (iF)-based approaches can precisely and quickly detect outliers scattering around one or more normal clusters. However, the performance of these methods seriously decreases when facing outliers whose nature “few and different” disappears in subspace (e.g., anomalies surrounded by normal samples). To solve this problem, SPiForest is proposed, which is different from existing approaches. First, SPiForest uses the principal component analysis (PCA) to find principal components and estimate each component’s probability density function (pdf). Second, SPiForest utilizes the inv-pdf, which is inversely proportional to the pdf estimated from the given dataset, to generate support points in the space containing all attributes. Third, the hyperplane decided by these support points is used to isolate the outliers in the space. Next, these steps are repeated to build an iTree. Finally, many iTrees construct a forest for outlier detection. SPiForest provides two benefits: 1) it isolates outliers with fewer hyperplanes, which significantly improves the accuracy and 2) it effectively detects the outliers whose nature “few and different” disappears in subspace. Comparative analyses and experiments show that the SPiForest achieves a significant improvement in terms of area under the curve (AUC) when compared with the state-of-the-art methods. Specifically, our method improves by at most 17.7% on AUC when compared to iF-based algorithms.

The conventional amplitude measurement of electrolyte-insulator–semiconductor-capacitor (EISCAP) sensor is convenient but susceptible to noise interference, which can deteriorate the measurement accuracy. In this study, a new amplitude-measurement method based on voltage-frequency–voltage (V/F/V) conversion is proposed, which is composed of impedance conversion, hysteresis comparison, and frequency-to-voltage conversion. The impedance conversion technique increases the equivalent capacitance of EISCAP; thus, the measurement sensitivity is improved obviously. The change of EISCAP output voltage is converted into the frequency variation of a periodic pulse signal by the hysteresis comparator. Compared to the amplitude of a signal, its frequency is less susceptible to circuit noise, so that the anti-noise ability of the measurement system is enhanced. Voltage output is more convenient and flexible; at the end of the detection circuit, the frequency-to-voltage conversion is performed. With this proposed amplitude-measurement method, the relative error of measurement is decreased from 2.2643% to 0.0045%. After using impedance converter, the detection sensitivity is increased by 88.66% from 0.0344 to 0.0649/pH at fixed bias voltage mode and by 64.94% from 39.53 to 65.20 mV/pH at fixed amplitude mode, respectively. In addition, the linearity was improved by 7.77%, and the repeatability standard deviation was decreased by 20% from 0.0045 to 0.0036.

In this article, we point out that the essential differences between convolutional neural network (CNN)-based and transformer-based detectors, which cause worse performance of small object in transformer-based methods, are the gap between local information and global dependencies in feature extraction and propagation. To address these differences, we propose a new vision transformer, called Hybrid Network Transformer (Hyneter), after preexperiments that indicate the gap causes CNN-based and transformer-based methods to increase size-different objects results unevenly. Different from the divide-and-conquer strategy in previous methods, Hyneters consist of hybrid network backbone (HNB) and dual switching (DS) module, which integrate local information and global dependencies, and transfer them simultaneously. Based on the balance strategy, HNB extends the range of local information by embedding convolution layers into transformer blocks in parallel, and DS adjusts excessive reliance on global dependencies outside the patch. Ablation studies illustrate that Hyneters achieve the state-of-the-art performance by a large margin of $+2.1\sim 13.2 \text{AP}$ on COCO, and $+3.1\sim 6.5 \text{mIoU}$ on VisDrone with lighter model size and lower computational cost in object detection. Furthermore, Hyneters achieve the state-of-the-art results on multiple computer vision tasks, such as object detection ($60.1 \text{AP}$ on COCO and $46.1 \text{AP}$ on VisDrone), semantic segmentation ($54.3 \text{AP}$ on ADE20K), and instance segmentation ($48.5 \text{AP}^{\text{mask}}$ on COCO), and surpass previous best methods. The code will be publicly available later.

We have demonstrated the realization of a coherent vesicle random lasing (VRL) from the dye doped azobenzene polymer vesicles self-assembled in the tetrahydrofuran-water system, which contains a double-walled structure: a hydrophilic and hydrophobic part. The effect of the dye and azobenzene polymer concentration on the threshold of random laser has been researched. The threshold of random laser decreases with an increase in the concentration of the pyrromethene 597 (PM597) laser and azobenzene polymer. Moreover, the scattering of small size group vesicles is attributed to providing a loop to boost the coherent random laser through the Fourier transform analysis. Due to the vesicles having the similar structure with the cell, the generation of coherent random lasers from vesicles expand random lasers to the biomedicine filed.

The application of deep learning-based semantic segmentation in remote sensing (RS) images has achieved considerable success. However, many supervised methods still heavily rely on a large amount of labeled data, requiring time-consuming and labor-intensive manual annotations. Besides, networks trained on labeled source domain data often perform poorly in inference tasks with target domain data due to the domain shift phenomenon. To address these challenges, we construct a novel end-to-end unsupervised domain adaptation (UDA) framework, named data augmentation and mobile threshold self-supervision (DA&MTSS), which integrates data augmentation with self-supervision. Specifically, we analyze the common factors that cause domain shifts in RS images and adopt different data augmentation techniques to attenuate the domain shift and enhance the generalization ability and robustness of the network in cross-domain inference. In the self-supervision phase, we design a new sample-based mobile threshold method to dynamically control the thresholds of both dominant and long-tail classes during training and generate stable pseudo-labels. Therefore, our method eliminates the need for additional training or expert knowledge and achieves the co-evolution of network parameters and pseudo-label quality in the training process. The results of comprehensive experiments on five tasks across the ISPRS Vaihingen, Potsdam, and LoveDA datasets demonstrate that this method consistently achieves higher mIoU scores, showcasing the performance advantage of DA&MTSS in UDA for the semantic segmentation of RS image.

The building contours, as one of the most important features for representing geometry, are widely used in various applications including urban modeling and reconstruction. Automatic extraction of high-fidelity compact contours from unstructured point clouds is rather challenging, and the existing methods are limited in generating global-aware and artifact-free building contours. Here, we approach contour extraction as a Bayesian inference problem. Two ideas are proposed to obtain building contour points and edges directly from unstructured point clouds. First, we construct a point-edge collaborative (PECo) Bayesian framework to couple the information of contour points and contour edges. The developed model fully takes local contour features, global edge structures, and global geometric priors into account. Second, given the Bayesian framework for building contouring, we leverage an expectation maximization (EM) algorithm to iteratively infer the contour edges in a maximum posteriori manner. The alternate EM iterations between point and edge domains progressively refine the local pointwise information to a global representation of contour edges. As a result, a synergistic effect between the point features and edge structures for global-aware building contouring is attained. Our approach outperforms the state-of-the-art methods in terms of geometric accuracy and structural compactness in modeling buildings with various complexities. Furthermore, the compactness of the contouring results can be more flexible and easily controlled.

Deployment of Synchronous Condensers (SynCons) has become one popular solution in accommodating large-scale renewable energy source (RES) for existing power systems. However, as a rotating device, SynCon may lose transient rotor angle stability caused by injected active power from nearby RES after a fault. To reveal this phenomenon, the transient stability of SynCon is investigated in this article. To derive intuitive mechanism and assist quantitative analysis, the equivalent synchronous generators (ESG) model is proposed to characterize the transient behavior of SynCon and co-located RES. The Lyapunov's direct method is employed for the assessment of transient stability by analyzing attraction domain for a typical benchmark system. On this basis, an index of transient stability margin against grid strength and an adaptive low voltage ride through (LVRT) control strategy are proposed for further assessment and enhancement of transient stability. The accuracy and effectiveness of the proposed methods are verified with EMT simulations of a typical windfarm configuration.

To facilitate emerging applications and demands of edge intelligence (EI)-empowered 6G networks, model-driven semantic communications have been proposed to reduce transmission volume by deploying artificial intelligence (AI) models that provide abilities of semantic extraction and recovery. Nevertheless, it is not feasible to preload all AI models on resource-constrained terminals. Thus, in-time model transmission becomes a crucial problem. This paper proposes an intellicise model transmission architecture to guarantee the reliable transmission of models for semantic communication. The mathematical relationship between model size and performance is formulated by employing a recognition error function supported with experimental data. We consider the characteristics of wireless channels and derive the closed-form expression of model transmission outage probability (MTOP) over the Rayleigh channel. Besides, we define the effective model accuracy (EMA) to evaluate the model transmission performance of both communication and intelligence. Then we propose a joint model selection and resource allocation (JMSRA) algorithm to maximize the average EMA of all users. Simulation results demonstrate that the average EMA of the JMSRA algorithm outperforms baseline algorithms by about 22%.