This paper presents a direct numerical simulation study of the effect of bubble vertical eccentricity on the behavior of a hollow droplet impacting on a super-hydrophobic cylindrical target. Based on the topological changes the droplet undergoes during impact， four typical regimes are identified： direct rebound， breakup-retraction， continuous breakup， and top breakup. The transitions between these regimes are analyzed in relation to the dimensionless parameters including the Weber number and bubble eccentricity ratio. Furthermore， the temporal evolution of key physical quantities， such as the spreading ratio， velocity， and kinetic energy， is discussed for each regime. Quantitative analysis further reveals that， the maximum spreading ratio of the hollow droplet during impact is a non-monotonic function of the bubble’s vertical position within the range of parameters considered. For relatively large Weber numbers， a smaller degree of bubble eccentricity facilitates droplet spreading. On the other hand， the dimensionless rupture time demonstrates an approximately linear relationship with the bubble eccentricity ratio， while exhibiting only a weak correlation with the Weber number.
Devices made of thermoelectric materials can realize the interconversion between heat and electricity， which is based on the Seebeck effect and the Peltier effect respectively， without any moving elements. Different from conventional metallic and inorganic thermoelectric materials， thermoelectric polymer composites possess good flexibility， decent stretchability， and healing ability， thus arousing ever-increasing interests among researchers because they are abundant， inexpensive， light-weighted， low-toxic or non-toxic， and thermally less conductive. Generally， thermoelectric polymer composites comprise nanomaterials which are usually carbon nanotubes， graphene， metal organic frameworks， MXenes， black phosphorus， etc. and conductive polymers which are typically polythiophene and its derivatives， polyaniline， and polypyrrole. Apart from traditional use of generating electricity from heat for thermoelectric materials， polymer composites acting as an indispensable complement to inorganic thermoelectric materials have combined advantages of conductive polymers and nanomaterials and are finding new applications in various sensors although their performance in certain aspects remains below that of conventional metallic and inorganic thermoelectric materials. New toolkits like artificial intelligence and machine learning have been introduced as effective ways in facilitating efficient design， preparation， and performance enhancement of thermoelectric polymer composites. It is expected that thermoelectric polymer composites will make great progress and demonstrate many broader application scenarios.
In a fusion reactor environment characterized by significant temperature variations and intense magnetic fields， the Seebeck effect interacts with the magnetic field， propelling the flow of liquid metal to remove impurities and heat generated during the fusion reaction. This paper conducts experiments on thermoelectric convection driven by the Seebeck effect in a horizontal magnetic field. GaInSn and constantan serve as the experimental working substances. An ultrasonic Doppler velocimetry system meticulously measures the convection velocity in a closed cavity. Three thermoelectric convection modes are identified. At lower magnetic fields， thermoelectric effects dominate the convective mode， allowing for an approximation as a two-dimensional flow. With increasing magnetic field strength， the convective mode transitions to a three-dimensional pattern exhibiting variable velocity fluctuations. Subsequently， it converts to an approximate two-dimensional flow， influenced by the magnetic damping effect under strong magnetic fields. The Lorentz force， resulting from the interaction between the thermoelectric effect and the magnetic field， can enhance heat transfer. However， under stronger magnetic fields， it has an inhibiting effect on heat transfer.
This article utilizes three-dimensional numerical simulations on the flow around a cold cylinder with a constant Reynolds number （Re=80） and a Richardson number （Ri） ranging from -5 to -0.5， to investigate the formation mechanism of the wake structure and the influence of buoyancy on the wake structure. The results indicate that within the range of -2.5≤Ri≤-0.5， the wake structure around the cold cylinder exhibits a consistent cyclic pattern due to the influence of buoyancy. As
To investigate the nitrogen contribution of terrestrial ecosystems along Niyaqu， a Namco lake-entering river， and to determine the role of wetlands along the river in the riverine system， the spatial pattern of nitrogen transport was monitored and analysed. During the 2019-2021 growing season， we measured flow and nitrogen component concentrations at multiple cross sections along the Niyaqu River. The results showed that locations （or cross sections） had no significant effect on total nitrogen （TN）， nitrate nitrogen， and ammonia nitrogen concentrations， but there were significant fluctuations observed interannually and in monthly dynamics. The mean of TN was （0.207±0.003） mg·L-1. The wetlands increased TN slightly by （0.022±0.002） mg·L-1（P<0.05）， but not nitrate nitrogen and ammonia nitrogen. Although terrestrial ecosystems recharge water along the way into the lake， TN throughout the river is slightly lower than the wet deposition level of inorganic N， which was reported to be 0.21-0.24 mg·L-1 in two multi-year studies at Namco Station. It could be concluded that terrestrial ecosystems along the Niyaqu River do not contribute additional nitrogen to Namco through the river.
The configuration of low impact development （LID） facilities is crucial for restoring urban natural hydrological processes. Therefore， we conducted this study at the Yanqi Lake campus of University of Chinese Academy of Sciences. It utilized simulations based on five rainfall recurrence intervals to analyze peak surface runoff and total suspended solid （TSS） loads across various subcatchments. A comprehensive stormwater condition index was created to prioritize subcatchments for LID facility configuration. Based on the specific requirements for LID facilities and local geographic conditions， optimal types and area proportions of LID facilities were determined， leading to a strategic LID configuration plan. The plan included different spatial configurations of green roofs， rain gardens， vegetated swales， and permeable pavements， and their impact on reducing local and overall runoff and TSS was assessed. The results indicated that implementing this strategy could reduce total surface runoff and TSS loads by 16%-24% across the study area， showing the most significant reduction compared to other strategies. The effectiveness of rain gardens， permeable pavements， and vegetated swales in reducing runoff and pollution increased with their area proportion and remained stable under different rainfall recurrences. Green roofs showed increased pollution reduction with greater area coverage， though their runoff reduction effectiveness was greatly influenced by surface land cover， improving with increased green space. However， under heavy rainfall， the effectiveness of all types of LID facilities， as well as the overall strategy， was reduced， indicating the need for integrated management with the urban underground drainage network and water systems. These findings provide a crucial reference for the redevelopment and construction of the study area to better manage potential flooding and pollution disasters from extreme rainfall events.
In the field of autonomous driving， accurate vehicle trajectory prediction plays a crucial role. While current deep learning-based algorithms have significantly improved the accuracy of vehicle trajectory prediction， they lack interpretability regarding the decision-making process of the algorithm. To address this issue， we incorporate prior knowledge into the deep learning-based algorithm and propose a trajectory prediction algorithm based on attention mechanisms. Diverging from traditional methods that add constraints for knowledge integration， we employ a tailored model architecture that embeds insights from the social force model to replicate the decision-making processes of drivers in complex traffic scenarios， thereby enhancing the interpretability of the predictions. Knowledge-infused trajectory prediction algorithm（KIT） leverages an attention mechanism to imitate drivers’ perception of their environment and uses a multilayer perceptron network for predicting accelerations influenced by the driver’s intentions， nearby traffic， and surrounding roads. The proposed method is validated on the Argoverse dataset， and the results indicate that KIT demonstrates superior predictive performance compared to current advanced trajectory prediction methods.
The integration of low earth orbit （LEO） satellite with 5G in the mobile communication system presents extensive application prospects. To ensure effective establishment of communication links between users and base stations， as well as reliable data transmission， time-frequency synchronization technology plays a crucial role. However， traditional timing and frequency synchronization algorithms face limitations when dealing with millimeter wave frequency bands， large bandwidths， high Doppler frequency shifts and change rates， and low signal-to-noise ratios in low-orbit satellite channels. These limitations result in decreased estimation accuracy. For the uplink service data channel of a 5G-based low orbit satellite communication system， a joint timing and frequency synchronization algorithm based on weighted embedded synchronization sequences is proposed to improve the accuracy of timing and frequency offset estimation. Simulation results demonstrate that the proposed timing and frequency synchronization algorithm outperforms traditional algorithms with comparable complexity in terms of synchronization performance.
Fake reviews affect online consumers’purchasing decisions. Efficiently identifying fake reviews is a pressing issue in the current development of e-commerce. Traditional methods for detecting fake reviews are often influenced by variations in review text style， syntax， and context， resulting in lower accuracy. Although large language models （LLMs） can address this accuracy issue， their training process is typically time-consuming. To tackle this problem， we propose a novel method called CF-DRI （cluster-based fine-tuning for deceptive review identification）. This method fine-tunes the pre-trained knowledge of LLMs by selecting clustered review samples， significantly enhancing the training efficiency for fake review identification. Compared to traditional methods， CF-DRI demonstrates superior performance with fewer fine-tuning samples. Experimental results on the Yelp.com dataset show that CF-DRI achieves a precision of 92.29% and a recall of 90.03% in fake review identification using only 20% of the clustered samples. This research provides new perspectives and solutions for managing fake reviews on e-commerce platforms， potentially promoting healthy industry development.
3D real-time rendering technology has a wide range of applications. At present， 3D real-time rendering technology has problems such as high computational complexity and high storage overhead， making it difficult to efficiently run on the web side with limited resources. Therefore， researching lightweight 3D real-time rendering technology is of great significance. Edge collapse algorithm is a commonly used technology for lightweight 3D real-time rendering， but it has problems such as easy loss of edge features， single simplification rate， and low quality of the folded mesh that affect visual effects. In response to the above issues， this article proposes an efficient and lightweight method for 3D real-time rendering on the web side. Firstly， an edge collapse optimization algorithm based on 3D-SIFT feature extraction is proposed to freeze key areas and better preserve model edge features. Secondly， during the edge collapse process， local information entropy is introduced to modify the cost of edge collapse， prioritizing the processing of non-feature regions， thereby achieving hierarchical simplification of different feature regions. Finally， the Delaunay algorithm is introduced to reconstruct areas with poor triangular regularity， improving the quality of the mesh.
Image stitching is a crucial prerequisite step for unmanned aerial vehicle （UAV） remote sensing applications， while the stitched images using most of the current image stitching methods often suffer from large irregular boundaries and multiple stitching seams， which can seriously affect subsequent analysis and applications. Existing improved methods typically cannot simultaneously address these two issues， and integrating the two types of methods in sequence is a straightforward way to solve the two problems， while this often can not obtain satisfactory performance because of the inevitable error propagation problem. This paper proposes an inpainting method for the UAV image stitching task based on the denoising diffusion probability model （DDPM）. The method uniformly designs masks for irregular boundaries and stitching seams， and a diffusion model is then utilized with manifold gradient prior constraints to complete the masked regions. By doing so， both irregular boundaries and stitching seams are simultaneously eliminated， thereby improving the quality of the stitching results. Comparative experiments are conducted using four datasets established for different scenarios. The experimental results demonstrated the efficacy of the proposed method in effectively eliminating irregular boundaries and seams in the image stitching. Moreover， from patches to pictures quality （PaQ-2-PiQ） and multi-scale image quality （MUSIQ） scores increased by 4.36% and 15.37%， respectively. Furthermore， at the locations of irregular boundaries， the structural similarity （SSIM） and peak signal-to-noise ratio （PSNR） values improved by 20.22% and 33.69%， respectively. Compared with state-of-the-art methods and other conventional image stitching algorithms， the proposed method performs better in both subjective and objective quality metric scores， has good robustness and generalization， and can be widely applied to UAV image stitching scenarios.
To address the problem of echo gain loss caused by range migration and Doppler frequency migration during long-time coherent accumulation of detection of high-speed maneuvering targets， a fast coherent integration algorithm based on K-adjacent pulse time-frequency double inversion transformation is proposed. By multiplying the distance-frequency and slow-time double-reversed conjugate signal of the Kth adjacent pulse signal with the target signal in the distance-frequency and slow-time domain， the across range unit and Doppler frequency migration in the signal can be simultaneously eliminated. At the same time， the time-frequency inversion cross-correlation algorithm is introduced as a supplement to estimate the target’s distance， velocity， and acceleration information through joint calculation. Finally， the compensation function is constructed to complete the focusing of echo energy on the distance-Doppler plane. Simulation results show that the proposed method can effectively estimate the second-order high-speed target motion parameters without any parameter search， with low computational complexity. Moreover， the K-adjacent pulse time-frequency dual inversion transformation and time-frequency inversion cross-correlation can be implemented in parallel to further improve the computational speed.
An image preprocessing accelerator scheme based on lightweight and low-cost FPGA has been proposed in this paper to meet the accelerated processing requirements of the visual navigation image frontend. Through efficient pipeline design and parallel processing technology， the designed accelerator integrates key functions such as histogram equalization， FAST feature point detection， and multi-source sensor data time synchronization. This solution solves technical difficulties such as achieving multifunctional integration， meeting real-time requirements， balancing cost and performance， synchronizing multi-source sensor information time， and achieving software hardware collaborative design under limited hardware resources. The proposed solution is based on Xilinx's Zynq-7000 series lightweight FPGA implementation， which greatly reduces image processing latency while achieving low cost. When the FPGA operates at a frequency of 160 MHz， it achieves a processing speed of 150 frame/s for 1 280×720 images， providing a low-cost and high-performance visual navigation image front-end acceleration solution.