Search

The deformation and failure of surrounding rock in soft rock tunnels are significantly exacerbated by the combined effects of mining activities and water immersion erosion. Water immersion leads to the failure of anchoring agents, which reduces the anchorage strength of bolts and cables, thereby diminishing their load-bearing capacity and severely compromising the safety of the tunnel face. To address this issue, a research methodology integrating physical similarity simulation experiments, numerical simulations, and field measurements was employed. The study investigated the failure characteristics of water-immersed soft rock tunnel surrounding rock under mining influence and varying support parameters. The deformation and failure characteristics of surrounding rock under mining and water immersion were identified, and optimal support parameters were proposed. Experimental results indicated that, under mining influence, the deformation and failure of surrounding rock in soft rock tunnels exhibited asymmetry, and under water immersion, the failure range of the surrounding rock expanded further. Increasing the length of the anchor cable from 9 m to 10 m resulted in a 4.3% reduction in the failure range of surrounding rock. When a 12.2 m anchor cable was used to support roof, the failure range on both sides was larger than when using the 9 m or 10 m cables. This was mainly because the 12.2 m cable reached the water-bearing layer in the roof, leading to increased water seepage and further softening of the surrounding rock. The use of a combined support system, consisting of a 9.2 m anchor cable and a 4.2 m short anchor cable, led to a reduction in the surrounding rock's failure range and a significant improvement in water seepage conditions compared to the previous support schemes. Field application results demonstrated that the combined support system, comprising a 9.2 m anchor cable and a 4.2 m short anchor cable, reduced roof subsidence, sidewall deformation, and floor uplift by 77.5%, 65%, and 76.5%, respectively, compared to the original support parameters. This effectively enhanced the surrounding rock's resistance to deformation and failure.

The adsorption of Ni2+ doped TiO2 (001) crystal plane for different gas molecules is studied by combining theory with experiment. The results show that O2 molecules are adsorbed on crystal plane in the form of chemical bond. XPS results prove that the doping of Ni2+ increases the chemisorbed oxygen ability of crystal plane. The adsorption of O2 molecules leads to the increase of the crystal plane conductivity, which is confirmed by the simulation results of the increase of Fermi level DOS and the electrochemical impedance spectroscopy (EIS) experiment. H2O and N2 molecules are adsorbed by crystal plane in the form of hydrogen bond and physical adsorption, respectively, and their contributions to the crystal plane conductivity increase and decrease respectively, which is also confirmed by EIS under different atmospheres. Interestingly, small-angle XRD proves that the sample exists interlayer clearance, and the adsorption of gas molecules in the crystal surface gap further improves the storage capacity of TiO2 materials for gas molecules. The results of this study are of great significance for improving the oxygen carrying capacity and catalytic gas sensitive performance of TiO2 materials. [ABSTRACT FROM AUTHOR]

Bone injury caused changes to surrounding tissues, leading to a large number of osteoclasts appeared to clear the damaged bone tissue before bone regeneration. However, overactive osteoclasts will inhibit bone formation. In this study, we prepared methacrylylated gelatin (GelMA)‐based hydrogel to co‐crosslink with OP3‐4 peptide, a receptor activator of NF‐κB ligand (RANKL) binding agent, to achieve the slow release of OP3‐4 peptide to inhibit the activation of osteoclasts, thus preventing the long‐term existence of osteoclasts from affecting bone regeneration, and promoting osteogenic differentiation. Moreover, CXCL9 secreted by osteoblasts will bind to endogenous VEGF and inhibit vascularization, finally hinder bone formation. Thus, anti‐CXCL9 antibodies (A‐CXCL9) were also loaded in the hydrogel to neutralize excess CXCL9. The hydrogel slow released of OP3‐4 cyclic peptide and A‐CXCL9 to simultaneously inhibiting osteoclast activation and promoting vascularization, thereby accelerating the healing of femur defect. Further analysis of osteogenic protein expression and signal pathways showed that the hydrogel may be through activating the AKT‐RUNX2‐ALP pathway and ultimately promote osteogenic differentiation. This dual‐acting hydrogel can effectively prevent nonunion caused by low vascularization and provide long‐term support for the treatment of bone injury. [ABSTRACT FROM AUTHOR]

The pre-trained model (PTM) is revolutionizing Artificial Intelligence (AI) technology. However, the hardware requirement of PTM training is prohibitively high, making it a game for a small proportion of people. Therefore, we proposed PatrickStar system to lower the hardware requirements of PTMs and make them accessible to everyone. PatrickStar uses the CPU-GPU heterogeneous memory space to store the model data. Different from existing works, we organize the model data in memory chunks and dynamically distribute them in the heterogeneous memory. Guided by the runtime memory statistics collected in a warm-up iteration, chunks are orchestrated efficiently in heterogeneous memory and generate lower CPU-GPU data transmission volume and higher bandwidth utilization. Symbiosis with the Zero Redundancy Optimizer, PatrickStar scales to multiple GPUs on multiple nodes. The system can train tasks on bigger models and larger batch sizes, which cannot be accomplished by existing works. Experimental results show that PatrickStar extends model scales 2.27 and 2.5 times of DeepSpeed, and exhibits significantly higher execution speed. PatricStar also successfully runs the 175B GPT3 training task on a 32 GPU cluster. Our code is available at https://github.com/Tencent/PatrickStar. [ABSTRACT FROM AUTHOR]

Aging is a physiological process mediated by numerous biological and genetic pathways, which are directly linked to lifespan and are a driving force for all age-related diseases. Human life expectancy has greatly increased in the past few decades, but this has not been accompanied by a similar increase in their healthspan. At present, research on aging biology has focused on elucidating the biochemical and genetic pathways that contribute to aging over time. Several aging mechanisms have been identified, primarily including genomic instability, telomere shortening, and cellular senescence. Aging is a driving factor of various age-related diseases, including neurodegenerative diseases, cardiovascular diseases, cancer, immune system disorders, and musculoskeletal disorders. Efforts to find drugs that improve the healthspan by targeting the pathogenesis of aging have now become a hot topic in this field. In the present review, the status of aging research and the development of potential drugs for aging-related diseases, such as metformin, rapamycin, resveratrol, senolytics, as well as caloric restriction, are summarized. The feasibility, side effects, and future potential of these treatments are also discussed, which will provide a basis to develop novel anti-aging therapeutics for improving the healthspan and preventing aging-related diseases. [ABSTRACT FROM AUTHOR]

Encryption is an important technique to improve information security for many real-world applications. The Advanced Encryption Standard (AES) is a widely-used efficient cryptographic algorithm. Although AES is fast both in software and hardware, it is time-consuming to do data encryption especially for large amount of data. Therefore, it is a lasting effort to accelerate AES operations. This paper presents SW-AES, a parallel AES implementation on the Sunway TaihuLight, one of the fastest supercomputers in the world that takes the SW26010 processor as the basic building block. According to the architectural features of SW26010, SW-AES exploits parallelism from different levels, including (1) inter-CPE (Computing Processing Element) data parallelism that distributes tasks among the 256 on-chip CPEs, (2) intra-CPE data parallelism enabled by the Single-Instruction Multiple-Data (SIMD) instructions inside each CPE, and (3) instruction-level parallelism that pipelines memory access and the computation. In addition, corresponding to the two application scenarios, SW-AES presents scalable ways to efficiently run AES on many nodes. As a result, SW-AES can gain a maximum throughput of 13.50 GB/s on a single SW26010 node, which is 216.23 × higher than the latest parallel AES implementation on the Sunway TaihuLight, and about 37.3% higher than the latest AES implementation on the GTX 480 GPU. When running on 1024 computing nodes with each one processing 1 GB data, SW-AES can achieve a throughput of 13819.25 GB/s. On the contrast, only a throughput of 63.91 GB/s can be achieved by the latest related work on the Sunway TaihuLight. • A data layout enabling SIMD operations inside one Computing Processor Element. • An S-Box lookup strategy improving the performance of nonvectorized memory access. • A data parallelism to fully use the 256 Computing Processor Elements on a chip. • New mechanisms to scale AES operations of different scenarios to many nodes. [ABSTRACT FROM AUTHOR]

Data parallelism has become a dominant method to scale Deep Neural Network (DNN) training across multiple nodes. Since the bandwidth requirement of synchronizing the gradients of the local model can be a bottleneck for large-scale distributed training, compressing communication traffic has gained widespread attention recently. Among several recent proposed compression algorithms, Residual Gradient Compression (RGC) is one of the most successful approaches—it can significantly compress the transmitting message size (0.1% of the gradient size) of each node and still achieve correct accuracy and the same convergence speed. However, the literature on compressing deep networks focuses almost exclusively on achieving good theoretical compression rate, while the efficiency of RGC in real implementation has been less investigated. In this paper, we develop an RGC method that is able to reduce the end-to-end training time on real-world multi-GPU systems. Our proposed RGC system design called RedSync, introduces a set of optimizations to reduce communication bandwidth while introducing limited overhead. We examine the performance of RedSync on two different multiple GPU platforms, including 128 GPUs of a supercomputer and an 8-GPU server. Our test cases include image classification on Cifar10 and ImageNet, and language modeling tasks on Penn Treebank and Wiki2 datasets. For DNNs featured with high communication to computation ratio, which has long been considered with poor scalability, RedSync shows significant performance improvement. [ABSTRACT FROM AUTHOR]

Support Vector Machine (SVM) is a classification method that has been widely used in the domain of remote sensing for decades. Although SVM-based classification method achieves good performance for classification accuracy in many studies, it can become very time-consuming in some remote sensing applications such as hyperspectral image classification or large-scale land cover mapping. To improve the efficiency for SVM training and classification in remote sensing applications, we designed and implemented a highly efficient multiclass support vector machine (MMSVM) for × 86-based multicore and many-core architectures such as the Ivy Bridge CPUs and the Intel Xeon Phi coprocessor (MIC) based on our previous MIC-SVM library. Various analysis methods and optimization strategies are employed to fully utilize the multilevel parallelism of our studied architectures. We select several real-world remote sensing datasets to evaluate the performance of our proposed MMSVM. Compared with the widely used serial LIBSVM, our MMSVM achieves 6.3–31.1 $\times$ (in training) and 4.9–32.2 $\times$(in classification) speedups on MIC, and 6.9–14.9$\times$ (in training) and 5.5–22.1$\times$ (in classification) speedups on the Ivy Bridge CPUs. We also conduct a performance comparison analysis on different platforms and provide some ideas on how to select the most suitable architecture for specific remote sensing classification problems in order to achieve the best performance. [ABSTRACT FROM PUBLISHER]

Automatic extraction of building footprints from high-resolution satellite imagery has become an important and challenging research issue receiving greater attention. Many recent studies have explored different deep learning-based semantic segmentation methods for improving the accuracy of building extraction. Although they record substantial land cover and land use information (e.g., buildings, roads, water, etc.), public geographic information system (GIS) map datasets have rarely been utilized to improve building extraction results in existing studies. In this research, we propose a U-Net-based semantic segmentation method for the extraction of building footprints from high-resolution multispectral satellite images using the SpaceNet building dataset provided in the DeepGlobe Satellite Challenge of IEEE Conference on Computer Vision and Pattern Recognition 2018 (CVPR 2018). We explore the potential of multiple public GIS map datasets (OpenStreetMap, Google Maps, and MapWorld) through integration with the WorldView-3 satellite datasets in four cities (Las Vegas, Paris, Shanghai, and Khartoum). Several strategies are designed and combined with the U-Net–based semantic segmentation model, including data augmentation, post-processing, and integration of the GIS map data and satellite images. The proposed method achieves a total F1-score of 0.704, which is an improvement of 1.1% to 12.5% compared with the top three solutions in the SpaceNet Building Detection Competition and 3.0% to 9.2% compared with the standard U-Net–based method. Moreover, the effect of each proposed strategy and the possible reasons for the building footprint extraction results are analyzed substantially considering the actual situation of the four cities. [ABSTRACT FROM AUTHOR]

[Display omitted] • O 2 molecules bond with Sn atoms on the crystal plane and H 2 O molecules bond with Zn atoms to form characteristic adsorbed species. • As the characteristic adsorbed species, H 2 O molecules further promoted the natural adsorption of O 2 molecules in the form of hydrogen bond. • The crystal conductivity changes with the adsorption and desorption. • High temperature facilitates the adsorption of O 2 molecules. The molecular adsorption behavior of Zn 2 SnO 4 (1 1 1) crystal plane was studied. DFT theoretical simulation result shows that O 2 and H 2 O molecules are adsorbed on the crystal plane by bonding with Zn atom and Sn atom respectively and forming characteristic adsorption species. H 2 O molecules as characteristic adsorbed species promote the natural adsorption of O 2 molecules in the form of hydrogen bonds, N 2 molecules adsorb naturally in the form of physical adsorption. Meanwhile, the crystal surface conductivity varies with the adsorption and desorption of different gases. When the adsorbent is only O 2 molecule, the crystal surface conductivity decreases, the DOS of Fermi is from 8.00 to 4.01 electrons/eV. The conductivity of crystal surface increases when H 2 O molecule acts as a characteristic adsorbent and further adsorbs O 2 molecule, the DOS of Fermi is from 5.74 to 8.51 electrons/eV. When oxygen molecules are directly adsorbed on Zn 2 SnO 4 (1 1 1) crystal plane, the adsorption energy is +0.79 eV, while when oxygen molecules are further adsorbed on Zn 2 SnO 4 (1 1 1) crystal plane through water molecule, the adsorption energy is +2.71 eV. This conclusion was confirmed by EIS. In situ Raman spectroscopy detection of acetone molecule gas sensitive experiment showed that the adsorption of O 2 molecules is slow at room temperature and fast at high temperature, which is consistent with the calculation result that O 2 molecule belongs to endothermic adsorption. The results show that, through the method of combining experiment and theory, the influencing mechanism is analyzed, which effects of different gases on adsorption properties of crystal surfaces. This work provides guidance for optimizing catalytic and gas sensitive properties of materials and realizing controllable optimization of surface properties of materials. [ABSTRACT FROM AUTHOR]