Search

A TiB2/AlMgScZr composite was fabricated by selective laser melting (SLM) with direct powder mixing. The microstructure and mechanical properties of SLM-fabricated TiB2/AlMgScZr composites were investigated. Nearly fully dense and crack-free SLM-fabricated TiB2/AlMgScZr parts were obtained. The results showed that TiB2 formed a coherent relationship with the matrix and was swallowed by the solid–liquid interface as a nucleation point to refine grains during the SLM process. In addition, some TiB2 particles did not form a coherent relationship with the matrix, moving the solid–liquid interface at the grain boundary, which reduced the recrystallization fraction and refined the grains significantly. TiB2 also promoted the precipitation of Al3Sc and Al3Zr phases and increased the volume fraction of the Al3Sc and Al3Zr phases during the SLM process. The SLM-fabricated TiB2/AlMgScZr composites showed a very high ultimate tensile strength of ∼483 MPa and excellent ductility of ∼15.8%, which are higher than those of wrought Al alloys. The solid solubility of the Mg element slightly increased in the SLM-fabricated TiB2/AlMgScZr composites because of the high heating and cooling rate during the SLM process. The high heating and cooling rate caused discontinuous strain and Portevin–Le Chatelier effect and weakened the strengthening effect of TiB2 in SLM-fabricated TiB2/AlMgSc composites.

Peer-to-peer (P2P) lending is an emerging field in FinTech and is an alternative source of personal loans. However, P2P lending faces severe credit risk due to high information asymmetry and insufficient collateral. We develop a novel heterogeneous stacking ensemble (HSE) approach by using two real-world datasets to improve the loss given default (LGD) forecasting in the P2P lending domain. Some special data in P2P lending and macroeconomic variables are employed as supplementary data sources to further enhance the model performance. Our proposal is compared with several popular models, including parametric and non-parametric ones, in terms of predictive accuracy and capital requirement. Our finding reveals that special data in P2P lending (e.g., number of investors and loan description) and macroeconomic variables are powerful predictors of LGD in P2P lending. The proposed HSE model outperforms the benchmark models in most cases and significantly achieves optimal average ranks across all the evaluation metrics. The results remain robust under several validations.

Many antigens from Mycobacterium tuberculosis (M. tuberculosis) have been demonstrated as strong immunogens and proved to have application potential as vaccine candidate antigens. Cyclic di-AMP (c-di-AMP) as a bacterial second messenger regulates various bacterial processes as well as the host immune responses. Rv2837c, the c-di-AMP phosphodiesterase (CnpB), was found to be relative to virulence of M. tuberculosis and interference with host innate immune response. In this study, recombinant CnpB was administered subcutaneously to mice. We found that CnpB had strong immunogenicity and induced high levels of humoral response and lung mucosal immunity after M. tuberculosis intranasally infection. CnpB immunization stimulated splenocyte proliferation and the increasing number of activated NK cells but had little effects on Th1/Th2 cellular immune responses in spleens. However, CnpB induced significant Th1/Th2 cellular immune responses with a decreased number of T and B cells in the lungs, and significantly recruits of CD4+ and CD8+ T cells after M. tuberculosis attenuated strain H37Ra infection. Besides, we first reported that CnpB could stimulate IFN-β expression transitorily and inhibit the autophagy of macrophages in vitro. In mice intranasally infection model, CnpB immunization alleviated pathological changes and reduced M. tuberculosis H37Ra loads in the lungs. Thus, our results suggested that CnpB interferes with host innate and adaptive immune responses and confers protection against M. tuberculosis respiratory infection, which should be considered in vaccine development as well as a drug target.

A highly efficient palladium-catalyzed cascade annulation of pyrazolones and aryl iodides to access various benzo[c]cinnoline derivatives has been achieved at 80 °C. A pyridine-type ligand could im...

The selective synthesis of C-2/C-3 aminated five-membered heteroarenes incorporated various functionalities with aniline derivatives in a sustainable way remains an unmet challenge. This protocol presents a practical protocol for the C–H amination of heteroarenes via an electro-oxidative cross coupling process. This electrosynthetic approach enables a facile access to a wide variety of synthetically useful hetero-arene derivatives, tolerates a wide range of functional groups, and is amenable to gram scale synthesis. In addition, a preliminary mechanistic study reveals that a radical cross coupling process between the N-centered radical and indole radical cation may be involved for this amination reaction.

The formation of magnetic minerals is bound up with the tectonic evolution history, whereupon the distribution of magnetic anomalies has great meanings for regional tectonics. In this study, we use the latest global lithospheric magnetic field model EMAG2-v3, processed by various techniques including reduction to the pole (RTP), upward continuation, derivations, Euler deconvolution, estimation of total magnetization direction, and Curie point depth (CPD), to unveil the tectonics around Longmenshan fault zone (LFZ). LFZ is clearly displayed as a positive and negative anomaly transition zone in RTP anomalies and acts as a magnetic basement boundary. The Sichuan Basin (SB), located to the east of LFZ, is marked by strong magnetic basement and NE-strike banded induced positive anomalies which are associated with the Neoproterozoic magmatic activity. The banded shape, absence of radial pattern of anomalies, and existence of fossil subduction zone supports that the magnetic basement was formed in arc environment. The CPD in SB estimated by radial average power spectral is 30–51 km, which allows magnetic minerals in deep crust or even in lithospheric mantle to exhibit high magnetizations. The Songpan-Ganzi fold belt (SGFB), in contrast, is located to the west of LFZ and covered by thick and low-susceptibility Triassic deposits of flysch, manifests as weak negative anomalies caused by relatively shallow CPD and widespread remanent magnetization. Significant positive anomalies, appearing around the Manai and Rilonguan granitic massifs, indicate a strong magnetic basement beneath SGFB, which is conjectured as westward extension of the Yangtze Block at depth.

Many antigens from

Piecewise planar 3D objects are very common in digital furniture manufacturing. In this paper, we propose a novel method for automatic reconstruction of 3D objects with planar facets from unorganized point clouds. We formulate this problem into a point clustering problem where the key difficulty lies in consolidating co-planar points into a cluster. In order to achieve this purpose, the first step is to triangulate the input point cloud into a mesh (may have over connectivity) that is a super-set of the underlying manifold mesh surface. Then the hundreds of thousands of normal vectors of triangles, after being mapped onto a Gauss sphere, are capable of reporting reliable facing directions of the faces of the final 3D model. After grouping the points based on co-planarity, we fit each cluster with a planar facet and then assemble them into a piecewise planarization representation of the whole model. We further introduce an additional regularization term to meet the orthogonality requirement on a particular occasion, and then transform this problem into a purely convex optimization problem. Our method is efficient and requires just a few parameters. Extensive experimental results show that it is able to handle point clouds with various levels of noise and yield a desirable piecewise planar 3D model with a clean and compact representation.

This paper discloses an efficient one-pot protocol to convert easily accessible 3-nitropyridines to 3-acetamidopyridin-2-yl 4-methylbenzenesulfonate derivatives which are core structures of many pharmaceutical molecules. The strategy successfully combined a three-step reaction in one pot via progressively adding different reactants at rt. The reaction displays good functional group tolerance and regioselectivity. Structurally diversified 3-nitropyridine could be time-efficiently (3.5 h) derivatized to various functional 2-O,3-N-pyridines which are apt for further elaborations. The transformation was amenable to gram-scale synthesis.

Multifunctional nanoparticles were simply synthesized by mixing a TICT+AIE featured molecule (TPAPP-CHO) with PBS solution. The fluorescent (FL) dots entered the cells via energy-related endocytosis and were located in lysosome emitting green FL. This indicated that the nanoparticles were dissociated in the lysosome. Moreover, the synthesized nanoparticles (NPs) demonstrate potent cytotoxicity against human U87 glioblastoma cells by inducing cell apoptosis via triggering intracellular ROS overproduction.

Pine wilt disease (PWD) is the most dangerous biohazard of pine species and poses a serious threat to forest resources. Coupling satellite remote sensing technology and deep learning technology for the accurate monitoring of PWD is an important tool for the efficient prevention and control of PWD. We used Gaofen-2 remote sensing images to construct a dataset of discolored standing tree samples of PWD and selected three semantic segmentation models—DeepLabv3+, HRNet, and DANet—for training and to compare their performance. To build a GAN-based semi-supervised semantic segmentation model for semi-supervised learning training, the best model was chosen as the generator of generative adversarial networks (GANs). The model was then optimized for structural adjustment and hyperparameter adjustment. Aimed at the characteristics of Gaofen-2 images and discolored standing trees with PWD, this paper adopts three strategies—swelling prediction, raster vectorization, and forest floor mask extraction—to optimize the image identification process and results and conducts an application demonstration study in Nanping city, Fujian Province. The results show that among the three semantic segmentation models, HRNet was the optimal conventional semantic segmentation model for identifying discolored standing trees of PWD based on Gaofen-2 images and that its MIoU value was 68.36%. Additionally, the GAN-based semi-supervised semantic segmentation model GAN_HRNet_Semi improved the MIoU value by 3.10%, and its recognition segmentation accuracy was better than the traditional semantic segmentation model. The recall rate of PWD discolored standing tree monitoring in the demonstration area reached 80.09%. The combination of semi-supervised semantic segmentation technology and high-resolution satellite remote sensing technology provides new technical methods for the accurate wide-scale monitoring, prevention, and control of PWD.

Doxorubicin (DOX) has powerful anticancer properties, but its clinical application is affected by its serious cardiotoxicity. Wogonin (WG) has been shown to have marked cardiovascular protection potential. However, it is not known whether this potential can protect the heart from DOX damage. The aim of the present study was to investigate whether WG could ameliorate the cardiotoxicity of DOX. DOX and WG were used to establish a model of cardiac damage. Echocardiography, brain natriuretic peptide, creatine kinase MB and cardiac troponin T were used to detect the degree of cardiac damage. The levels of superoxide dismutase, malondialdehyde, glutathione and catalase in serum were measured to observed oxidative stress state. The mRNA levels of cyclophilin D, voltage-dependent anion-selective channel 1 and adenine nucleotide transporter 1 were detected by reverse transcription-quantitative PCR. Western blotting was used to detect the expression of cytochrome c in mitochondria and cytoplasm and cleaved-caspase-9 and pro/cleaved-caspase-3 in cytoplasm in cardiac tissue and primary cardiomyocytes to verify the related signaling pathways. DOX rats showed a series of cardiac damage. However, these damages were alleviated following WG treatment. Further studies showed that WG antagonized DOX cardiotoxicity through inhibiting the release of cytochrome c. WG protected rat heart from DOX damage. The mechanism may be closely related to inhibiting the release of cytochrome c from mitochondria and reducing cardiomyocyte apoptosis caused by caspase activation.

A practical protocol to synthesize 3-substituent-2-(azol-1-yl)indole derivatives has been developed via an electrochemical oxidative cross coupling process under mild conditions. This electro-oxidative C-N bond formation strategy tolerates a range of functional groups and is amenable to gram scale synthesis. Moreover, this method was applied to the late-stage functionalization of bioactive molecules.

The cardiotoxicity of pirarubicin (THP) seriously affects its clinical application, which cannot be ignored. The antioxidant effect of schisandrin B (SchB) has been extensively reported in the context of dietotherapy. However, whether this antioxidant effect can protect the heart from THP damage remains unknown. The aim of the present study was to investigate whether the antioxidant effect of SchB can antagonize the cardiotoxicity of THP. Changes in electrocardiogram (ECG), echocardiography and serum lactate dehydrogenase, brain natriuretic peptide, creatine kinase MB and cardiac troponin T levels were used to detect the degree of cardiac damage. The levels of superoxide dismutase (SOD), malondialdehyde, catalase and total antioxidant capacity in the serum and heart were measured to observe the oxidative stress state of rats. Primary cardiomyocytes were cultured, and cell viability and reactive oxygen species (ROS) production were detected. Western blotting was used to detect the expression levels of SOD2, NOX2, pro/cleaved-caspase3 and Bcl-2/Bax in heart tissue and primary cardiomyocytes to verify the related signaling pathways. THP-treated rats showed a range of cardiac damage, including an abnormal ECG, echocardiography and myocardial enzymes. In the cellular experiments, cell viability decreased and ROS increased. However, this damage was alleviated after SchB treatment. Further studies demonstrated that SchB antagonized THP cardiotoxicity via its antioxidant effect. In conclusion, SchB protects the heart from THP damage in rats, and the mechanism may be closely associated with its antioxidant effect.

Deep neural network (DNN) models have been widely used for click-through rate (CTR) prediction in online advertising. The training framework typically consists of embedding layers and multi-layer perceptions (MLP). At Baidu Search Ads (a.k.a. Phoenix Nest), the new generation of CTR training platform has become PaddleBox, a GPU-based parameter server system. In this paper, we present Baidu's recently updated CTR training framework, called Gating-enhanced Multi-task Neural Networks (GemNN). In particular, we develop a neural network based multi-task learning model to predict CTR in a coarse-to-fine manner, which gradually reduces ad candidates and allows parameter sharing from upstream tasks to downstream tasks to improve the training efficiency. Also, we introduce a gating mechanism between embedding layers and MLP to learn feature interactions and control the information flow fed to MLP layers. We have launched our solution in Baidu PaddleBox platform and observed considerable improvements in both offline and online evaluations. It is now part of the current production~system.

Focusing on the issue that the accuracy of object detection is reduced when the foreground target moves slowly and there is interference in the background, a moving object detection method based on weighted kernel norm and saliency constraint RPCA is proposed. In the new method, the weighted kernel norm is used to restore a relatively clean background, which is helpful to separate the slow moving target from the background. The l 1 norm is used to constrain the sparsity of moving objects, and the Frobenius norm is used to detect noise, and a saliency constraint is introduced to detect slow-moving objects. Experiments show that this method can effectively deal with the problems of slow foreground motion and background interference. Compared with the suboptimal algorithm, the average measured value F of the proposed algorithm is improved by 15%.

Traditional vibration sensor of ship machinery using cable for power supply cannot meet the requirements of a large number of distributed sensors of the intelligent ship. A bouncing-ball based triboelectric nanogenerator (BB-TENG) to be used as a self-powered vibration frequency monitoring sensor is proposed. In the sensing test, the vibration frequency of 15 ∼ 50 Hz under different amplitudes can be measured accurately. The frequency calculation of the voltage signal can accurately reflect external vibration frequency, while the change of short-circuit current has linear relationship with the frequency, and the correlation coefficient is about 0.99. In addition, electrical signal test results and photos taken by high-speed cameras show that the limit of the available detection frequency of BB-TENG increases with the increase of amplitude and support height. The electrical output of BB-TENG under different working conditions are tested, which provides the basis for selecting suitable BB-TENG according to different vibration characteristics.

In this paper, we propose a general purpose, simple and fast fusion algorithm based on guided image filter. The proposed method can well combine useful source image information into the fused image supported by multi-scale image decomposition, structure transferring property, visual saliency detection, and weight map construction. Multi-scale image decomposition is appropriate to represent and manipulate image features at various scales. Structure transferring property enabled by our algorithm can induce structures of one source image into the other. A new visual saliency detection based on guided image filter introduced in this paper is able to extract significant regions from visually different images of the same scene. The choice of weight maps helped to integrate the complementary information pixel by pixel at each scale. Experimental outcomes of the proposed method are compared and analyzed with traditional and recent guided image filter-based fusion algorithms in terms of visual quality, fusion metrics and run time. In addition, to enhance fusion results further we made an effort to find a suitable image and video enhancement algorithm. The fusion performance analysis clearly indicates that the proposed method is very promising along with less run time.

In this paper, an extensive survey on PLD methods for low-altitude aircraft safety based on image processing is presented. First, primary features of power lines are introduced. Then, main studies on PLD are summarized, including PLD approaches based on tradition image processing methods and machine learning. Besides, a few latest research works, such as PLD based on deep learning and infrared images, are discussed. By discussing and comparing current methods, we have proposed future development directions of PLD, with the aim of providing a starting point for researchers in PLD. Specifically, image fusion and multimodal sensor fusion should be able to provide better detection performance and lower false alarm rate by exploiting complementary information from different sensors. In addition, AI will be an important technique in PLD, with the help of transfer learning to utilize limited training data and of multimodal learning to deal with different kinds of information.

This paper describes a facile one-pot protocol to synthesize 2-aminophenol derivatives via a palladium-catalyzed C–H acetoxylation strategy with 5-nitropyrimidine as a directing group (DG), which can be easily preinstalled and readily removed under mild condition after the coupling. In addition, the transformation is operationally simple, has high functional group tolerance, and is amenable to gram-scale. Moreover, several examples were shown that introduction/removal of 5-nitropyrimidine and the C–H oxylation sequence could be integrated in one pot.

Pervasive and continuous energy solutions are highly desired in the era of the Internet of Things for powering wide-range distributed devices/sensors. Wind energy has been widely regarded as an ideal energy source for distributed devices/sensors due to the advantages of being sustainable and renewable. Herein, we propose a high-performance flag-type triboelectric nanogenerator (HF-TENG) to efficiently harvest widely distributed and highly available wind energy. The HF-TENG is composed of one piece of polytetrafluoroethylene (PTFE) membrane and two carbon-coated polyethylene terephthalate (PET) membranes with their edges sealed up. Two ingenious internal-structure designs significantly improve the output performance. One is to place the supporting sponge strips between the PTFE and the carbon electrodes, and the other is to divide the PTFE into multiple pieces to obtain a multi-degree of freedom. Both methods can improve the degree of contact and separation between the two triboelectric materials while working. When the pair number of supporting sponge strips is two and the degree of freedom is five, the maximum voltage and current of HF-TENG can reach 78 V and 7.5 μA, respectively, which are both four times that of the untreated flag-type TENG. Additionally, the HF-TENG was demonstrated to power the LEDs, capacitors, and temperature sensors. The reported HF-TENG significantly promotes the utilization of the ambient wind energy and sheds some light on providing a pervasive and sustainable energy solution to the distributed devices/sensors in the era of the Internet of Things.

With the development of intelligent ship, types of advanced sensors are in great demand for monitoring the work conditions of ship machinery. In the present work, a self-powered and highly accurate vibration sensor based on bouncing-ball triboelectric nanogenerator (BB-TENG) is proposed and investigated. The BB-TENG sensor consists of two copper electrode layers and one 3D-printed frame filled with polytetrafluoroethylene (PTFE) balls. When the sensor is installed on a vibration exciter, the PTFE balls will continuously bounce between the two electrodes, generating a periodically fluctuating electrical signals whose frequency can be easily measured through fast Fourier transform. Experiments have demonstrated that the BB-TENG sensor has a high signal-to-noise ratio of 34.5 dB with mean error less than 0.05% at the vibration frequency of 10 Hz to 50 Hz which covers the most vibration range of the machinery on ship. In addition, the BB-TENG can power 30 LEDs and a temperature sensor by converting vibration energy into electricity. Therefore, the BB-TENG sensor can be utilized as a self-powered and highly accurate vibration sensor for condition monitoring of intelligent ship machinery.

Gestalt psychology indicates that simplicity is central to image vectorization, i.e., observers tend to perceive jagged raster edges as piecewise smooth curves and color changes as being either gradual (along edges) or abrupt (across edges). In this paper, we give a pair of simplicity-driven formulations to respectively cope with the two challenges. In detail, we formulate the underlying as-rigid-as-possible edges as the axes of symmetry of the edge saliency map, while reconstructing the color field by enforcing the fidelity and the smoothness at the same time (except on the detected boundaries). We finally convert a rasterized image into gradient-aware vector graphics whose base domain is a high-quality triangle mesh. On the one hand, the rigidity of the boundary curves is naturally achieved based on the assumption of simplicity, instead of by an empirically-grounded curve fitting operation; on the other hand, the color of near-boundary regions is inferred by Hessian energy (an extrapolation-like technique). Our vectorization method is able to yield more visually realistic results than existing approaches and is useful in flexible recoloring, shape editing, and hierarchical level-of-detail (HLOD) image representation.

Hadoop Distributed File System (short for HDFS) is a high availability file system designed to run on commodity hardware. It uses multiple replicas to ensure high reliability, and many data are transmitted between storage nodes. The performance of data transmission has a great impact on the latency of writing operations. Remote Direct Memory Access (short for RDMA) is a protocol with low latency and high through which is running on the Infiniband network. When HDFS runs on the Infiniband network, the default protocol IPoIB can not take advantage of the high-speed network. The latency of the writing process is similar to a TCP/IP network. In this paper, we present a new RDMA-based HDFS writing mechanism. It enables DataNodes to read data parallelly from the Client through RDMA. And by using RDMA primitive, the transmission latency is slower than the original TCP/IP protocol. The experiments show that our approach reduces the latency of the writing process by 10.11%–40.81% compared with the original HDFS.

Recently, computer vision and deep learning technology has been applied in various gait rehabilitation researches. Considering the long short-term memory (LSTM) network has been proved an excellent performance in learn sequence feature representations, we proposed a lower limb joint trajectory prediction method based on LSTM for conducting active rehabilitation on a rehabilitation robotic system. Our approach based on synergy theory exploits that the follow-up lower limb joint trajectory, i.e. limb intention, could be generated by joint angles of the previous swing process of upper limb which were acquired from Kinect platform, an advanced computer vision platform for motion tracking. A customize Kinect-Treadmill data acquisition platform was built for this study. With this platform, data acquisition on ten healthy subjects is processed in four different walking speeds to acquire the joint angles calculated by Kinect visual signals of upper and lower limb swing. Then, the angles of hip and knee in one side which were presented as lower limb intentions are predicted by the fore angles of the elbow and shoulder on the opposite side via a trained LSTM model. The results indicate that the trained LSTM model has a better estimation of predicting the lower limb intentions, and the feasibility of Kinect visual signals has been validated as well.

Silicon nanocomposites with surface coating have been promising candidates for high-performance anodes in Li-ion batteries. Here we report a simple yet effective way to synthesize Si@SiO2 nanocomposites with amorphous SiO2 coating through a fluoride ion-mediated continuous etching–redeposition process. The Si@SiO2 nanocomposite anode exhibits significantly improved lithium storage properties, with a reversible capacity of 1356 mA h g−1 at 2.1 A g−1 for 300 cycles.

Foot-and-mouth disease (FMD) is an infectious and sometimes fatal viral disease that affects cloven-hoofed animals, and Chinese government adopts compulsory immunization measures for FMD. The adverse effects of FMD vaccine to pigs, cattle and goats have been reported increasingly frequent during the spring and autumn seasons when large numbers of farm livestock are vaccinated. The financial losses caused by vaccine adverse effects have been a serious concern for both farmers and primary prevention personnel. There are various causative factors reported to involve into adverse effect of FMD vaccine, including the inappropriate vaccine production, transportation and storage, livestock poor tolerance, and unqualified vaccinating manipulations. Symptomatic treatment and early drug prevention have a certain effect on the adverse effects. To analyze causes and propose countermeasures, in the current study possible reasons during the production and processing procedures of inactivated FMD vaccine were reviewed and corresponding countermeasures were recommended. The review may provide references for better use of vaccine to prevent FMD.

The addition of Sc and Zr into an Al–Mg–Mn alloy produced by selective laser melting exhibited exceptional properties. Multiple factors eventually lead to the improvement of properties, such as densification behaviors, grain characteristics, texture and the precipitated phases. We firstly investigate evolution of the melt pool characteristics which are varied under different processing parameters and lead to different solidified microstructures, densification behaviors and corresponding mechanical properties. Meanwhile, the characteristics of melt pool can be predicted through the normalized enthalpy of input energy, which presents more credible relationship with the melt pool dimensions. To this end, the relative density is up to 99.88% when the volume energy density and normalized enthalpy are combined used. Recrystallisation inhabitations induced by the added Sc and Zr is supposed after comparing the grain sizes under different conditions. Regrettably, the relationship between melt pool characteristics and the equiaxed/columnar ratio can not be described accurately, which due to the pseudo-equiaxed regions in XZ/YZ sections through regulating the melt track orientations via 67° scanning rotation. There are dispersed Al6Mn and Al3(Sc, Zr) particles exist in as-fabricated samples result in better mechanical properties and the subsequent aging treatment can further improve the comprehensive mechanical properties. Finally, the ultimate tensile strength and ductility is separately about 494.32 ± 4.16 MPa and 15 ± 1.27% after aging for 4 h at 325 °C. In contrast, the processing parameters have little impact on mechanical properties, especially for the yield strength, mainly owing to the homologous precipitate phase and grain sizes when the relative density is similar. The connection involving the whole periods from solidification to the final mechanical properties is discussed that may offer feasible ideas for alloy design and corresponding parameter optimization.

The addition of Sc and Zr endows selective laser melted Al–Mg-based alloys with exceptional properties, which are predominately related to the Al3(ScxZr1-x) precipitates. The Al3(ScxZr1-x) precipitates formed in different stage of processing are identified as primary- Al3(ScxZr1-x) and aged- Al3(ScxZr1-x) according to different crystallographic characteristics. The experimental results show most of primary Al3(ScxZr1-x) particles located in the grain boundaries with incoherent interface and irregular shape, which is seldom reported in conventional casting. In contrast, the subsequent aging brings out more spherical Al3(ScxZr1-x) with coherent interface emerged in the grains. Combined the first-principle calculations with pseudo binary-diagram, we search the possible structures of Al3(ScxZr1-x) based on their formation enthalpy and binding energy, whose space group should be altered as Sc enters the Al3Zr (DO23) dispersoids. The most possible structure of primary Al3(ScxZr1-x) phase is evaluated as Al12Sc3Zr, and the calculated crystallo-graphic parameters are consistent with the experimental results. As the primary Al3(ScxZr1-x) is hard to act as nucleation particle, it can be inferred that the grain refinement resulting from the modification and hindrance of these particles during grain growth. A corresponding ultimate tensile strength and ductility is separately about 521.26 ± 12.30 MPa and 15.83 ± 0.94% after aging at 325 °C for 4 h. The results demonstrate the difference between the primary Al3(ScxZr1-x) and the aged Al3(ScxZr1-x) as well as the formation of primary Al3(ScxZr1-x), which offer a new idea to design new alloy for additive manufacturing.

Maraging steel is a promising material for additive manufacturing due to its ultrahigh yield strength, reasonable ductility and good weldability. However, the ductility of the fabricated part will be degraded after aging treatment. In this regard, we firstly designed a new composition of maraging steel Fe-18.3Ni-9Co-4.84Mo-0.92Ti-0.27Al-0.13Cr-0.01C (wt.%), whose strength and ductility can be simultaneously improved by selective laser melting. The relationship between laser process parameters and forming defects has been studied. Using single-track and single-layer experiments, fully dense parts were fabricated with a certain range of process parameters, corresponding to the 30–50% lap rate in the X–Y plane and 70% remelted rate along Z-axis. Besides, we found film-like reverted austenite along the martensite lath boundaries in the as-fabricated part. The effects of heat treatment processes on the reverted austenite and mechanical properties of the fabricated parts were also studied; the strength-ductility trade-off of maraging steel after heat treatment can be alleviated. The tensile strength and elongation of printed samples after direct aging treatment can respectively reach 2037 MPa and 6.4%, and 2182 MPa, 4.8% after solution and aging treatment.

The solution of atmospheric chemical kinetics is considered as the dominant computationally intensive task in chemical transport models. Reported strategies for improving computational efficiency have not changed the essence of solving ordinary differential equations. In this study, we emulate a typical gas-phase chemistry solver implemented in chemical transport models in Chinese Mainland using a 28-layer residual regression neural network. Validation results have a total R2 of 0.985 with the original numerical solution for 194 species during a single timestep. Multiple-timestep test for one month verified the accuracy and stability of our model while no exponential accumulated bias was observed, though a slow accumulation of biases existed. There were 102 species with R2 values higher than 0.90 and 137 species with daily NMB values below 30% among the 194 species throughout the month. Meanwhile, the computational efficiency of gas-phase chemistry using the deep learning emulator was promoted by 10.6 times on one CPU and 85.2 times on one GPU. Our study confirms the feasibility of emulating an atmospheric chemical kinetic solver with data-driven models through deep learning and represents a step towards a deep learning solver being used in a chemical transport model. The deep learning emulator is expected to not only accelerate atmospheric chemical kinetic solutions by orders of magnitude but also provide a promising pathway for implementing near-explicit rather than lumped chemical mechanisms in chemical transport models.

Lithium-rich layered compound cathode materials recently attract ever-growing attention in lithium ion batteries for electric vehicles and energy storage devices due to their high discharge capacities of over 250 mAh g −1 and low cost. However, they still suffer from capacity fade, voltage decay and poor rate capability and thermal stability. In this paper, the Al&F co-doped Li 1.2 Ni 0.13 Co 0.13- x Mn 0.54 Al x O 2(1- y ) F 2 y cathode material has been successfully prepared via a co-precipitation reaction and a subsequent solid phase calcination. The resulting material combines the advantages of both Al-doping and F-doping, which can effectively mitigate the intrinsic layered-to-spinel phase transformation, and as a consequence, inhibits the capacity fade and voltage decay: delivering a discharge capacity of 217 mAh g −1 with retention of 88.21% and an average discharge voltage decay of 0.4019 V upon 150 cycling at 0.5 C. On the other side, with an increased electronic and ionic conductivity, this co-doping practice enhances the rate capability by a considerable extent that it delivers a discharge capacity of 157 mAh g −1 at 10 C, thus facilitating its application for quick charging devices. In addition, the thermal stability related with battery safety is significantly meliorated. Specifically, the co-doped cathode material exhibits an initial exothermal peak at a higher temperature of 273 °C and a much less overall heat generation of 221 J g −1 compared to the 210 °C and 755 J g −1 for the pristine one. In conclusion, the anion/cation co-substitution strategy has been demonstrated to be of great promise for the improvement of lithium-rich layered cathode materials.

Nowadays the Unmanned Aerial Vehicle (UAV) has been widely used due to its low-cost and unique flexibility. Specifically, the high-altitude operational capability makes it the ideal tool in military and civilian surveillance system, in which object tracking based on computer vision is the core ingredient. In this paper, we presented a survey on object tracking methods in aerial surveillance. After briefly reviewing the development history and current research institutions, we summarized frequently used sensors in aerial platform. Then we focused on UAV-based tracking methods by providing detailed descriptions of its common framework (ego motion compensation, object detection, object tracking) and representative tracking algorithms. Through discussing the requirement of a good tracking system and deficiency of current technologies, future directions for aerial surveillance were proposed.

Silicon nanocomposites with surface coating have been promising candidates for high-performance anodes in Li-ion batteries. Here we report a simple yet effective way to synthesize Si@SiO2 nanocomposites with amorphous SiO2 coating through a fluoride ion-mediated continuous etching-redeposition process. The Si@SiO2 nanocomposite anode exhibits significantly improved lithium storage properties, with a reversible capacity of 1356 mA h g-1 at 2.1 A g-1 for 300 cycles.

Ensemble models have been extensively applied to credit scoring. However, advanced tree-based classifiers have been seldom utilized as components of ensemble models. Moreover, few studies have considered dynamic ensemble selection. To fill the research gap, this paper aims to develop a novel tree-based overfitting-cautious heterogeneous ensemble model (i.e., OCHE) for credit scoring which departs from existing literature on base models and ensemble selection strategy. Regarding base models, tree-based techniques are employed to acquire a balance between predictive accuracy and computational cost. In terms of ensemble selection, the proposed method can assign weights to base models dynamically according to the overfitting measure. Validated on five public datasets, the proposed approach is compared with several popular benchmark models and selection strategies on predictive accuracy and computational cost measures. For predictive accuracy, the proposed approach outperforms the benchmark models significantly in most cases based on the non-parametric significance test. It also performs marginally better than several state-of-the-art studies. Our proposal remains robust in several scenarios. In terms of computational cost, the proposed method provides acceptable performance and benefits from GPU acceleration considerably.

Computing geodesic distances on polyhedral surfaces is an important task in digital geometry processing. Speed and accuracy are two commonly-used measurements of evaluating a discrete geodesic algorithm. In applications, such as parametrization and shape analysis, a smooth distance field is often preferred over the exact, non-smooth geodesic distance field. We use the term Quasi-geodesic Distance Field (QGDF) to denote a smooth scalar field that is as close as possible to an exact geodesic distance field. In this paper, we formulate the problem of computing QGDF into a standard quadratic programming (QP) problem which maintains a trade-off between accuracy and smoothness. The proposed QP formulation is also flexible in that it can be naturally extended to point clouds and tetrahedral meshes, and support various user-specified constraints. We demonstrate the effectiveness of QGDF in defect-tolerant distances and symmetry-constrained distances. The authors would like to thank the anonymous reviewers for their valuable comments and suggestions. This work is supported by National Natural Science Foundation of China (61772016, 61772312), the NSFC-Zhejiang Joint Fund for the Integration of Industrialization and Informatization (U1909210), the Dalian University of Technology 2019 Discipline Platform Fund (1000-82212201), the National Young Talents Program of China, and the Strategic Priority Research Program of Chinese Academy of Science (XDA21010205).

A method for smoke detection in video is proposed. The camera monitoring the scene is assumed to be stationary. With the atmospheric scattering model, dissipation function is reflected the transmissivity between the background objects in the scene and the camera. Dark channel prior and fast bilateral filter are used for estimating dissipation function which is only the function of the depth of field. Based on dissipation function, visual background extractor (ViBe) can be used for detecting smoke as a result of smoke’s motion characteristics as well as detecting other moving targets. Later on, various characteristics of smoke are extracted. Color feature and high-frequency energy based on wavelet transformation is selected to constitute the final recognition vectors, and Support vector machine (SVM) is used as a classification model. The final experimental results show that the accuracy rate of this method for smoke detection can reach 90.1%.

Precise monitoring of human respiration is crucial to the diagnosis of a wide range of respiratory and cardiovascular disorders and (thus) is of great interest to both clinicians and researchers. Existing measurements may include electrocardiography (ECG), nasal cannula and respiratory plethysmography. However, low comfort levels and signal shifts which are often observed in these techniques pose limitations for long-term, accurate monitoring of human respiration. In this paper, we develop a convenient respiratory signal acquisition method based on angular velocity derived from suprasternal notch. Research has found that waveforms collected from the suprasternal notch display higher robustness, as well as less gender variability. We use the median filter method and intersection detection technique to extract the respiratory waveform, respiratory frequency and respiratory phase parameters. And validated the respiratory signals derived from suprasternal angular velocity in both males and females. The extracted parameters are validated against a carbon dioxide concentration acquisition device, which serves as our golden reference platform. The results demonstrate the potential of suprasternal-derived angular velocity as a simple, low-cost and unobtrusive method for monitoring human respiration.