Explore the vast range of titles available.

The fiber mode field overlap integral method is employed to analyze the influencing factors of coupling efficiency, as well as the effects of axial and radial alignment errors on coupling efficiency under different relative apertures of coupling lenses. The results indicate that there exists an optimal relative aperture of the coupling lens that maximizes coupling efficiency; however, at this optimal point, coupling efficiency is more susceptible to radial errors. Regarding axial and radial errors, when the relative aperture of the coupling lens is at its optimal value, the tolerance for alignment errors is minimal. Conversely, when the relative aperture exceeds the optimal value, both coupling efficiency and tolerance for alignment decrease. When the relative aperture is less than the optimal value, the requirement for installation accuracy decreases while the tolerance becomes larger. The trend of the simulation results aligns with the experimental data. This study provides instructive significance regarding the trade-off between coupling efficiency requirements and alignment accuracy in the design of actual polarization-maintaining fiber coupling systems.

Rapid urbanization worldwide offers numerous benefits but also introduces challenges, particularly concerning urban climate comfort, which affects the physical and social well-being in cities. This study examined the microclimatic characteristics of the Burç neighborhood in the historical core of Burdur city, using ENVI-Met models with temperature, relative humidity, wind and PET data collected over a 33,665 m2 area at 06:00, 09:00, 12:00, 15:00, 18:00 and 21:00 on 15 August 2023. The analysis revealed that thermal comfort decreases significantly from 09:00 onwards, especially on hard surfaces like asphalt, concrete and parquet, which lack vegetation and intensify heat retention. By contrast, green areas were found to enhance bioclimatic comfort by reducing perceived temperatures by up to 20% in shaded and vegetated zones. Based on these findings, it is recommended that urban areas reduce heat-absorbing materials, such as asphalt and concrete and prioritize green spaces in landscape planning to improve thermal comfort and create more sustainable urban environments.

CO2 pre-fracturing is an innovative technique for enhancing oil and gas production in unconventional reservoirs. Despite its potential, the mechanisms of CO2 pre-fracturing influencing fracture propagation, particularly in ultra-deep reservoirs, remain inadequately understood. This study investigates the CO2 pre-fracturing process in ultra-deep sandstone reservoirs of the central Junggar Basin. A 3D geomechanical model was established using RFPA3D-HF based on rock mechanical parameters from laboratory experiments. The study examines the effect of in situ horizontal stress differences, CO2 pre-injection volume, and slickwater injection rate on fracture complexity index (FCI) and stimulated reservoir volume (SRV). The results reveal that in situ horizontal stress differences are the primary factor influencing fracture propagation. In ultra-deep reservoirs, high horizontal stress difference hinders fracture deflection and bifurcation during slickwater fracturing. CO2 pre-fracturing, through the pre-injection of CO2, reduces formation breakdown pressure and increases reservoir pore pressure due to its low viscosity and high permeability, effectively mitigating the effect of high horizontal stress differences and significantly enhancing fracturing effectiveness. Furthermore, appropriately increasing the CO2 pre-injection volume and slickwater injection rate can increase fracture complexity, resulting in a larger SRV. Notably, adjusting the CO2 pre-injection volume is more effective than adjusting slickwater injection rate in enhancing oil production. This study provides scientific evidence for selecting construction parameters and optimizing oil recovery through CO2 pre-fracturing technology in deep unconventional oil reservoirs and offers new insights into CO2 utilization and storage.

The imbalance in the “production–living–ecology” function (PLEF) has become a major issue for global cities due to the rapid advancement of urbanization and industrialization worldwide. The realization of PLEF coupling and coordination is crucial for a region’s sustainable development. Existing research has defined the concept of PLEF from the perspective of land function and measured its coupling coordination level using relevant models. However, there is still room for improvement in the indicator system, research methods, and other aspects. This work builds a PLEF coupling coordination evaluation-index system based on the perspective of human habitat using multi-source data in order to examine the spatial differences in PLEF coupling coordination level and the influencing factors in the Yellow River Basin (YRB). Using the modified coupling coordination model, the Moran index, spatial Markov chain model, and geographically weighted random forest model were introduced to analyze its spatial and temporal differentiation and influencing factors. The results found that (a) the level of PLEF coupling coordination in the YRB from 2010 to 2022 has been improving, and the number of severely imbalanced cities has been reduced from 23 to 15, but the level of downstream cities’ coupling coordination is significantly higher than that of upstream cities. The probability of cities maintaining their own level is greater than 50%, and there is basically no cross-level transfer. (b) The Moran index of the PLEF coupling coordination level has risen from 0.137 to 0.229, which shows a significant positive clustering phenomenon and is continually strengthening. The intercity polarization effect is being continually enhanced as seen in the LISA clustering diagram. (c) There is significant heterogeneity between the influencing factors in time and space. In terms of importance level, the series is per capita disposable income (0.416) > nighttime lighting index (0.370) > local general public budget expenditure (0.332) > number of beds per 1000 people (0.191) > NO2 content in the air (0.110). This study systematically investigates the dynamic evolution of the coupled coordination level of PLEF in the YRB and its influencing mechanism, which is of great practical use.

Water pollution by hexavalent chromium (Cr(VI)) is widespread and problematic. As a result, more research into economic Cr(VI) removal is needed. In this study, we created and employed an adsorption-reduction mechanism to remove Cr(VI). Magnetically reduced graphene oxide bentonite (MrGO-BT) is acid resistant and can undergo magnetic separation. The hydroxyl group of chitosan (CS) condensed with the functional groups on the surface of bentonite (BT), and the MrGO-BT sandwich has been fabricated and constructed from an Fe3O4 core layer sandwiched by reduced graphene oxide (rGO) and a BT shell, with CS acting as a crosslinker. Cr(VI) elimination by MrGO-BT was exothermic and spontaneous according to thermodynamic analyses. The adsorption kinetics and adsorption isotherms were characterized by the pseudo-second order kinetic theory and the Langmuir model, respectively. Regarding the elimination of Cr(VI), the greatest adsorption ability for Cr(VI) elimination achieved was 91.5 mg g-1. Fourier-transform infrared spectroscopy and X-ray photoelectron spectroscopy suggested that Cr(VI) was reduced by C-O-H on MrGO-BT to produce Cr(III) and H-C=O, and that Cr(III) chelated with amino groups or exchanged with BT after intercalation. In addition, the introduction of Cu2+ increased the positive charge of MrGO-BT and amplified the electrostatic interaction between Cr2O72- and HCrO4-, which is what caused Cr(VI) to be eliminated. Cu2+ and reduced Cr(III) combined with -NH2 on the surface of MrGO-BT to form -NH-Cr(III) or -NH-Cu2+, and Cr(VI) elimination via chelation and ion exchange was confirmed. MrGO-BT is shown to be an adsorbent with high acid resistance and good magnetic responsiveness and stability.

Wet milling of aluminum alloys involves complex interactions among thermal, fluid, and mechanical fields that strongly affect cutting temperature, stress distribution, and surface integrity. To achieve reproducible and physics-based predictions of these coupled phenomena, this study develops a three-dimensional thermo–fluid–solid-coupled Eulerian–Lagrangian (CEL) framework for the wet milling of Al6061. The model system in this study evaluated the effects of milling cutter feed rate and spindle speed, feed per tooth of the milling cutter, axial cutting depth, and coolant flow rate on equivalent stress and peak milling temperature., as well as their correlation with surface roughness metrics (Ra, Sa). Simulation results reveal that higher feed rates significantly raise Tpeak (+12.9%) while reducing σeq (−22.7%) and deteriorating surface quality (Ra +104.2%, Sa +29.9%). Increasing spindle speed lowers both Tpeak (−2.2%) and σeq (−8.5%) and improves surface finish (Ra −39.3%, Sa −16.6%). A greater depth of cut amplifies mechanical and thermal loads, increasing Tpeak (+10.3%) and σeq (+17%). Enhanced coolant flow reduces Tpeak (−23.5%) and σeq (−6.1%) and markedly improves surface quality (Ra −88.8%, Sa −51.3%). Research findings indicate that coolant coverage is the dominant factor determining surface integrity. Although experimental data for Tpeak and σeq were not directly validated, this framework clearly articulates modeling assumptions, quantifies parameter sensitivities, and provides a reproducible methodology for future experimental-numerical verification.

Human motion gesture recognition is the most challenging research direction in the field of computer vision, and it is widely used in human-computer interaction, intelligent monitoring, virtual reality, human behaviour analysis, and other fields. This paper proposes a new type of deep convolutional generation confrontation network to recognize human motion pose. This method uses a deep convolutional stacked hourglass network to accurately extract the location of key joint points on the image. The generation and identification part of the network is designed to encode the first hierarchy (parent) and the second hierarchy (child) and show the spatial relationship of human body parts. The generator and the discriminator are designed as two parts in the network, and they are connected together in order to encode the possible relationship of appearance and, at the same time, the possibility of the existence of human body parts and the relationship between each part of the body and its parental part coding. In the image, the key nodes of the human body model and the general body posture can be identified more accurately. The method has been tested on different data sets. In most cases, the results obtained by the proposed method are better than those of other comparison methods.

To achieve the satisfactory performance of human action recognition, a central task is to address the sub-action sharing problem, especially in similar action classes. Nevertheless, most existing convolutional neural network (CNN)-based action recognition algorithms uniformly divide video into frames and then randomly select the frames as inputs, ignoring the distinct characteristics among different frames. In recent years, depth videos have been increasingly used for action recognition, but most methods merely focus on the spatial information of the different actions without utilizing temporal information. In order to address these issues, a novel energy-guided temporal segmentation method is proposed here, and a multimodal fusion strategy is employed with the proposed segmentation method to construct an energy-guided temporal segmentation network (EGTSN). Specifically, the EGTSN had two parts: energy-guided video segmentation and a multimodal fusion heterogeneous CNN. The proposed solution was evaluated on a public large-scale NTU RGB+D dataset. Comparisons with state-of-the-art methods demonstrate the effectiveness of the proposed network.

Recently developed hybrid models that stack 3D with 2D CNN in their structure have enjoyed high popularity due to their appealing performance in hyperspectral image classification tasks. On the other hand, biological genome graphs have demonstrated their effectiveness in enhancing the scalability and accuracy of genomic analysis. We propose an innovative deep genome graph-based network (GGBN) for hyperspectral image classification to tap the potential of hybrid models and genome graphs. The GGBN model utilizes 3D-CNN at the bottom layers and 2D-CNNs at the top layers to process spectral–spatial features vital to enhancing the scalability and accuracy of hyperspectral image classification. To verify the effectiveness of the GGBN model, we conducted classification experiments on Indian Pines (IP), University of Pavia (UP), and Salinas Scene (SA) datasets. Using only 5% of the labeled data for training over the SA, IP, and UP datasets, the classification accuracy of GGBN is 99.97%, 96.85%, and 99.74%, respectively, which is better than the compared state-of-the-art methods.

Human skeleton-based action recognition is an important task in the field of computer vision. In recent years, masked autoencoder (MAE) has been used in various fields due to its powerful self-supervised learning ability and has achieved good results in masked data reconstruction tasks. However, in visual classification tasks such as action recognition, the limited ability of the encoder to learn features in the autoencoder structure results in poor classification performance. We propose to enhance the encoder's feature extraction ability in classification tasks by leveraging the latent space of variational autoencoder (VAE) and further replace it with the latent space of vector quantized variational autoencoder (VQVAE). The constructed models are called SkeletonMVAE and SkeletonMVQVAE, respectively. In SkeletonMVAE, we constrain the latent variables to represent features in the form of distributions. In SkeletonMVQVAE, we discretize the latent variables. These help the encoder learn deeper data structures and more discriminative and generalized feature representations. The experiment results on the NTU-60 and NTU-120 datasets demonstrate that our proposed method can effectively improve the classification accuracy of the encoder in classification tasks and its generalization ability in the case of few labeled data. SkeletonMVAE exhibits stronger classification ability, while SkeletonMVQVAE exhibits stronger generalization in situations with fewer labeled data.