Search

Mesoscale eddies, including surface-intensified eddies (SEs) and subsurface-intensified eddies (SSEs), significantly influence phytoplankton distribution in the ocean. Nevertheless, due to the sparse in-situ data, it is still unclear in understanding the characteristics of SSEs and their influence on chlorophyll-a (Chl-a) concentration. Consequently, the study utilized a deep learning model to extract SEs and SSEs in the North Indian Ocean (NIO) from 2000 to 2015, using long time series of satellite-derived sea surface height (SSH) and sea surface temperature (SST) data. The analysis revealed that SSEs accounted for 44 % of the total eddies in the NIO, and their SST signatures exhibited an opposite behavior compared to SEs. Furthermore, by integrating ocean color remote sensing data, the study investigated the contrasting impacts of SEs and SSEs on Chl-a concentration in two basins of the NIO: the Arabian Sea (AS) and the Bay of Bengal (BoB), known for their disparate biological productivity. In the AS, SEs induced Chl-a anomalies that were two to three times higher than those caused by SSEs. Notably, there were no significant differences in Chl-a anomalies induced by the same type of eddies between summer and winter. In contrast, the BoB exhibited distinct seasonal variations, where SEs induced slightly higher Chl-a anomalies than SSEs during the summer, while substantial differences were observed during the winter. Specifically, subsurface-intensified anticyclonic eddies (SSAEs) led to positive Chl-a anomalies, contrasting the negative anomalies induced by surface-intensified anticyclonic eddies (SAEs) with comparable magnitudes. Moreover, while both subsurface-intensified cyclonic eddies (SSCEs) and surface-intensified cyclonic eddies (SCEs) resulted in positive Chl-a anomalies, the magnitude of SSCEs was only one-fourth of that induced by SCEs. The distinct Chl-a anomalies between SEs and SSEs can be attributed to the contrasting subsurface structures revealed by Argo profiles. The upward displacement of isopycnals within SSAEs and the downward displacement within SSCEs in the upper 30–50 meters lead to higher and lower Chl-a concentrations, respectively. The study provides a valuable approach to investigating subsurface eddies and contributes to a comprehensive understanding of their influence on chlorophyll concentration.

In this paper we apply neural networks with local converging inputs (NNLCI), originally introduced in [arXiv:2109.09316], to solve the two dimensional Maxwell's equation around perfect electric conductors (PECs). The input to the networks consist of local patches of low cost numerical solutions to the equation computed on two coarse grids, and the output is a more accurate solution at the center of the local patch. We apply the recently developed second order finite difference method [arXiv:2209.00740] to generate the input and training data which captures the scattering of electromagnetic waves off of a PEC at a given terminal time. The advantage of NNLCI is that once trained it offers an efficient alternative to costly high-resolution conventional numerical methods; our numerical experiments indicate the computational complexity saving by a factor of $8^3$ in terms of the number of spatial-temporal grid points. In contrast with existing research work on applying neural networks to directly solve PDEs, our method takes advantage of the local domain of dependence of the Maxwell's equation in the input solution patches, and is therefore simpler, yet still robust. We demonstrate that we can train our neural network on some PECs to predict accurate solutions to different PECs with quite different geometries from any of the training examples.

Although Deep Learning (DL) has achieved success in complex Artificial Intelligence (AI) tasks, it suffers from various notorious problems (e.g., feature redundancy, and vanishing or exploding gradients), since updating parameters in Euclidean space cannot fully exploit the geometric structure of the solution space. As a promising alternative solution, Riemannian-based DL uses geometric optimization to update parameters on Riemannian manifolds and can leverage the underlying geometric information. Accordingly, this article presents a comprehensive survey of applying geometric optimization in DL. At first, this article introduces the basic procedure of the geometric optimization, including various geometric optimizers and some concepts of Riemannian manifold. Subsequently, this article investigates the application of geometric optimization in different DL networks in various AI tasks, e.g., convolution neural network, recurrent neural network, transfer learning, and optimal transport. Additionally, typical public toolboxes that implement optimization on manifold are also discussed. Finally, this article makes a performance comparison between different deep geometric optimization methods under image recognition scenarios., Comment: 41 pages

The Atlantic Meridional Overturning Circulation (AMOC) is an essential component of the climate system due to its vital role in the global distribution of heat, carbon, and water masses. The downwelling of North Atlantic surface waters connecting the upper and lower AMOC limbs is an essential yet vulnerable partof this global circulation. This downwelling partly occurs along continental boundaries. This dissertation presents an extended observational investigation on the quantification of Eulerianmean downwelling along the continental slopes of the North Atlantic subpolar gyre (SPG) and an examination of the underlying mechanisms, with an emphasis on the role of mesoscale eddies. A volume budget of the SPG boundary reveals a total Eulerian-mean (2002-2019) downwelling of -4.41±0.96 Sv at 1300 m depth between Denmark Strait and Flemish Cap, with the barotropic transport (BT) contributing 2.66±0.40 Sv and the baroclinic transport (BC) contributing 1.75±0.43 Sv. To investigate the processes that cause the BC boundary downwelling, i.e., the boundary heat loss and associate along-boundary density gradient, the long-term mean heat budget of the boundary current system is studied. Both lateral heat fluxes, driven by the boundary current/mesoscale eddies, and air-sea heat flux play significant roles in the boundary heat loss. In a Lagrangian framework, it is found that cross-shore eddy propagation generally cools the SPG boundary.; La circulation méridienne de retournement de l'Atlantique (AMOC) est une composante essentielle du système climatique en raison de son rôle vital dans la distribution globale de la chaleur, du carbone et des masses d'eau. La descente des eaux de surface de l'Atlantique Nord reliant les branches supérieure et inférieure de l'AMOC est une composante essentielle mais vulnérable de cette circulation globale. Le transport vertical associé se produit en partie le long des frontières continentales. Cette thèse présente une étude observationnelle étendue sur la quantification du « downwelling » moyen eulérien le long des pentes continentales du gyre subpolaire de l'Atlantique Nord (SPG) et un examen des mécanismes sous-jacents, en mettant l'accent sur le rôle des tourbillons de méso-échelle. Un bilan de volume révèle un « downwelling » total moyen (2002-2019) de 4.41±0.96 Sv à 1300 m de profondeur entre le détroit du Danemark et le Cap de Flemish, le transport barotrope (BT) contribuant pour 2.66±0.40 Sv et le transport barocline (BC) pour 1.84±0.44 Sv. Afin d'étudier les processus à l'origine de la plongée BC le long des bords du gyre SPG, i.e. la perte de chaleur et le gradient de densité associé, le bilan de chaleur moyen du courant de bords est étudié. A la fois les flux de chaleur latéraux, induits par les courants de bord et les tourbillons à méso-échelle, et les flux de chaleur air-mer jouent un rôle important dans la perte de chaleur du courant de bord. Dans un cadre lagrangien basé, on constate que la propagation des tourbillons contribue à refroidir la région de bord du gyre SPG.

The Atlantic Meridional Overturning Circulation (AMOC) is an essential component of the climate system due to its vital role in the global distribution of heat, carbon, and water masses. The downwelling of North Atlantic surface waters connecting the upper and lower AMOC limbs is an essential yet vulnerable part of this global circulation. This downwelling partly occurs along continental boundaries. This dissertation presents an extended observational investigation on the quantification of Eulerian-mean downwelling along the continental slopes of the North Atlantic subpolar gyre (SPG) and an examination of the underlying mechanisms, with an emphasis on the role of mesoscale eddies. A volume budget of the SPG boundary reveals a total Eulerian-mean (2002-2019) downwelling of - 4.41±0.96 Sv at 1300 m depth between Denmark Strait and Flemish Cap, with the barotropic transport (BT) contributing 2.66±0.40 Sv and the baroclinic transport (BC) contributing 1.75±0.43 Sv. To investigate the processes that cause the BC boundary downwelling, i.e., the boundary heat loss and associate along-boundary density gradient, the long-term mean heat budget of the boundary current system is studied. Both lateral heat fluxes, driven by the boundary current/mesoscale eddies, and air-sea heat flux play significant roles in the boundary heat loss. In a Lagrangian framework, it is found that cross-shore eddy propagation generally cools the SPG boundary., La circulation méridienne de retournement de l'Atlantique (AMOC) est une composante essentielle du système climatique en raison de son rôle vital dans la distribution globale de la chaleur, du carbone et des masses d'eau. La descente des eaux de surface de l'Atlantique Nord reliant les branches supérieure et inférieure de l'AMOC est une composante essentielle mais vulnérable de cette circulation globale. Le transport vertical associé se produit en partie le long des frontières continentales. Cette thèse présente une étude observationnelle étendue sur la quantification du « downwelling » moyen eulérien le long des pentes continentales du gyre subpolaire de l'Atlantique Nord (SPG) et un examen des mécanismes sous-jacents, en mettant l'accent sur le rôle des tourbillons de méso-échelle. Un bilan de volume révèle un « downwelling » total moyen (2002-2019) de 4.41±0.96 Sv à 1300 m de profondeur entre le détroit du Danemark et le Cap de Flemish, le transport barotrope (BT) contribuant pour 2.66±0.40 Sv et le transport barocline (BC) pour 1.84±0.44 Sv. Afin d'étudier les processus à l'origine de la plongée BC le long des bords du gyre SPG, i.e. la perte de chaleur et le gradient de densité associé, le bilan de chaleur moyen du courant de bords est étudié. A la fois les flux de chaleur latéraux, induits par les courants de bord et les tourbillons à mésoéchelle, et les flux de chaleur air-mer jouent un rôle important dans la perte de chaleur du courant de bord. Dans un cadre lagrangien basé, on constate que la propagation des tourbillons contribue à refroidir la région de bord du gyre SPG.

On an industrial plant, we observed and examined the ammonia-poisoning catalyst for the synthesis of dimethyl oxalate (DMO). We investigated the catalytic activity in response to the amount of ammonia and revealed the mechanism of such poisoning by X-ray photoelectron spectroscopy (XPS) characterization. Our results show that only 0.002% ammonia in the feed gas can significantly deactivate the Pd-based catalyst. Two main reasons were proposed: one is that the competitive adsorption of ammonia on the active component Pd hinders the carbon monoxide (CO) coupling reaction and the redox cycle between Pd0 and Pd2+; and the other is that the high-boiling nitrogen-containing amine compounds formed by reacting with ammonia have adsorbed on the catalyst, which hinders the progress of the catalytic reaction. The deactivation caused by the latter is irreversible. The catalytic activity can be completely restored by a low-temperature liquid-phase in-situ regeneration treatment. Copyright © 2021 by Authors, Published by BCREC Group. This is an open access article under the CC BY-SA License (https://creativecommons.org/licenses/by-sa/4.0).

Fluoro-containing heterocycles are widely applied in agricultural, medicinal, imaging, materials science and other fields due to their unique physical and chemical properties. Thus, development of new and practical methods for efficiently synthesizing these heterocycle skeletons has been a rapidly growing research field in modern synthetic chemistry. In this review, we provide a summary of the established strategies through fluorination and fluoroalkylation of alkenes/alkynes used over the past decade for constructing fluoro-containing heterocycles. In addition, emphasis is also made in this review on the scopes and mechanisms of these reactions and some of the unsolved scientific problems as well as future directions in this field.

With the continuous development of space and sensor technologies during the last 40 years, ocean remote sensing has entered into the big-data era with typical five-V (volume, variety, value, velocity and veracity) characteristics. Ocean remote-sensing data archives reach several tens of petabytes and massive satellite data are acquired worldwide daily. To precisely, efficiently and intelligently mine the useful information submerged in such ocean remote-sensing data sets is a big challenge. Deep learning—a powerful technology recently emerging in the machine-learning field—has demonstrated its more significant superiority over traditional physical- or statistical-based algorithms for image-information extraction in many industrial-field applications and starts to draw interest in ocean remote-sensing applications. In this review paper, we first systematically reviewed two deep-learning frameworks that carry out ocean remote-sensing-image classifications and then presented eight typical applications in ocean internal-wave/eddy/oil-spill/coastal-inundation/sea-ice/green-algae/ship/coral-reef mapping from different types of ocean remote-sensing imagery to show how effective these deep-learning frameworks are. Researchers can also readily modify these existing frameworks for information mining of other kinds of remote-sensing imagery.

Turbot (Scophthalmus maximus), commercially important flatfish species, is widely cultivated in Europe and China. With the continuous expansion of the intensive breeding scale, turbot is exposed to various stresses, which greatly impedes the healthy development of turbot industry. Here, we present an improved high-quality chromosome-scale genome assembly of turbot using a combination of PacBio long-read and Illumina short-read sequencing technologies. The genome assembly spans 538.22 Mb comprising 27 contigs with a contig N50 size of 25.76 Mb. Annotation of the genome assembly identified 104.45 Mb repetitive sequences, 22,442 protein-coding genes and 3,345 ncRNAs. Moreover, a total of 345 stress responsive candidate genes were identified by gene co-expression network analysis based on 14 published stress-related RNA-seq datasets consisting of 165 samples. Significantly improved genome assembly and stress-related candidate gene pool will provide valuable resources for further research on turbot functional genome and stress response mechanism, as well as theoretical support for the development of molecular breeding technology for resistant turbot varieties.

Patient similarity assessment (PSA) is pivotal to evidence-based and personalized medicine, enabled by analyzing the increasingly available electronic health records (EHRs). However, machine learning approaches for PSA has to deal with inherent data deficiencies of EHRs, namely missing values, noise, and small sample sizes. In this work, an end-to-end discriminative learning framework, called SparGE, is proposed to address these data challenges of EHR for PSA. SparGE measures similarity by jointly sparse coding and graph embedding. First, we use low-rank constrained sparse coding to identify and calculate weight for similar patients, while denoising against missing values. Then, graph embedding on sparse representations is adopted to measure the similarity between patient pairs via preserving local relationships defined by distances. Finally, a global cost function is constructed to optimize related parameters. Experimental results on two private and public real-world healthcare datasets, namely SingHEART and MIMIC-III, show that the proposed SparGE significantly outperforms other machine learning patient similarity methods.

We propose a finite difference method to solve Maxwell's equations in time domain in the presence of a perfect electric conductor that impedes the propagations of electromagnetic waves. Our method is a modification of the existing approach by Zou and Liu [36], from a locally perturbed body-fitted grid to a uniform orthogonal one for complicated PEC objects. Similar to their work we extrapolate ghost point values by exploiting the level set function of the interface and the PDE-based extension technique, which allows us to circumvent scrutinizing local geometries of the interface. We stipulate a mild requirement on the accuracy of our extrapolation that the ghost values need only be locally second order accurate. Nevertheless the resulting accuracy of our method is second order thanks to the application of back and forth error correction and compensation, which also relaxes CFL conditions. We demonstrate the effectiveness of our approach with some numerical examples.

Inspired by the tremendous success of the self-attention mechanism in natural language processing, the Vision Transformer (ViT) creatively applies it to image patch sequences and achieves incredible performance. However, the scaled dot-product self-attention of ViT brings about scale ambiguity to the structure of the original feature space. To address this problem, we propose a novel method named Orthogonal Vision Transformer (O-ViT), to optimize ViT from the geometric perspective. O-ViT limits parameters of self-attention blocks to be on the norm-keeping orthogonal manifold, which can keep the geometry of the feature space. Moreover, O-ViT achieves both orthogonal constraints and cheap optimization overhead by adopting a surjective mapping between the orthogonal group and its Lie algebra.We have conducted comparative experiments on image recognition tasks to demonstrate O-ViT's validity and experiments show that O-ViT can boost the performance of ViT by up to 3.6%.

Aggregation equations are broadly used to model population dynamics with nonlocal interactions, characterized by a potential in the equation. This paper considers the inverse problem of identifying the potential from a single noisy spatial-temporal process. The identification is challenging in the presence of noise due to the instability of numerical differentiation. We propose a robust model-based technique to identify the potential by minimizing a regularized data fidelity term, and regularization is taken as the total variation and the squared Laplacian. A split Bregman method is used to solve the regularized optimization problem. Our method is robust to noise by utilizing a Successively Denoised Differentiation technique. We consider additional constraints such as compact support and symmetry constraints to enhance the performance further. We also apply this method to identify time-varying potentials and identify the interaction kernel in an agent-based system. Various numerical examples in one and two dimensions are included to verify the effectiveness and robustness of the proposed method.

In our prior work [arXiv:2109.09316], neural network methods with inputs based on domain of dependence and a converging sequence were introduced for solving one dimensional conservation laws, in particular the Euler systems. To predict a high-fidelity solution at a given space-time location, two solutions of a conservation law from a converging sequence, computed from low-cost numerical schemes, and in a local domain of dependence of the space-time location, serve as the input of a neural network. In the present work, we extend the methods to two dimensional Euler systems and introduce variations. Numerical results demonstrate that the methods not only work very well in one dimension [arXiv:2109.09316], but also perform well in two dimensions. Despite smeared local input data, the neural network methods are able to predict shocks, contacts, and smooth regions of the solution accurately. The neural network methods are efficient and relatively easy to train because they are local solvers., Comment: arXiv admin note: substantial text overlap with arXiv:2109.09316

In this paper, we address the problem of trajectory tracking control of autonomous vehicles by considering the nonlinear characteristics of tires. By considering the influence of the tires’ dynamics on steering stability, the proposed predictive controller can track the desired trajectory and desired velocity in the presence of road curvature while minimizing the lateral tracking deviation. First of all, a hierarchical control structure is adopted, in which the upper-level controller is used to calculate the desired acceleration and the desired front-wheel angle to maintain the control target, and the lower-level controller realized the command through the corresponding component devices. Moreover, a force estimator is designed based on the radial basis function (RBF) neural network to estimate the lateral force of the tires, which is incorporated into the boundary conditions of the vehicle envelope constraint to improve the adaptability of the controller to the vehicle performance. Finally, the proposed controller is tested by co-simulation of CarSim (a simulation software specifically for vehicle dynamics)/Simulink (a modular diagram environment for multidomain simulation as well as model-based design) and hardware-in-loop simulation system. The co-simulation and experimental results demonstrate the controller safely driving at the vehicle’s handling limits and effectively reduce the slip phenomenon of the vehicle.

The stratum and microenvironment temperatures in a greenhouse are important factors that affect crop yield. In order to solve the problem of temperature imbalance caused by solar radiation in greenhouses, this paper proposes the application of a solar radiation array tube in a greenhouse. By adding water or phase change materials to the array tube, the influence of the array tube on the formation and microenvironment temperature changes was studied, and a 10-day test was carried out. A test group and control group were set up to monitor test results, and the ground was divided into six areas. The depths of each area were 10 cm, 30 cm, and 50 cm, and the heights of the greenhouse centers were 0 cm, 30 cm, 60 cm, 90 cm, 120 cm, 150 cm, and 180 cm. Via an analysis of the test results obtained for the formation and microenvironment temperature, the arrangement of the array tube was found to exert a constant temperature regulation effect on the microenvironment of the greenhouse at a formation depth of 10 cm and was able to improve this formation depth to a certain extent. The temperature at 30 cm and 50 cm plays a positive role in building a good vegetation growth environment.

There are some complicated steps in a conventional data processing procedure, including signal preprocessing, feature extraction and classification. This increases the complexity and difficulty of selecting suitable approaches. Thus, the simplification of processing procedure has become a crucial part for electronic noses (e-noses) applications. To overcome the shortcomings, we present a simplified e-nose data processing method based on glomerular microcircuits. Inspired by the structure and function of olfactory glomeruli, a glomerular microcircuit model is established by the synaptic interactions of olfactory receptor neurons, mitral cells, external tufted cells, periglomerular cells and short axon cells. The continuous sensor data is transformed into spike time sequences by the microcircuit model to enhance computational efficiency. To refine the olfactory information contained in spike time, a SpikeProp neural network is designed for feature extraction and classification. The proposed method, which can automatically implement feature learning without signal preprocessing, significantly improves data processing procedure as well as detection performance. The glomerular microcircuits based method was validated for different brands of Chinese liquors with conventional data processing methods. Experimental results show that the proposed method has higher classification rate and simpler procedure than conventional methods.

Subsurface-intensified eddies are ubiquitous in the world ocean, and they can transport large amounts of heat and salt and influence the three-dimensional distribution of oceanic tracers at a global scale. There is a lack of knowledge of subsurface-intensified eddies due to the lack of global in situ measurement. The correlations between sea surface height (SSH) and sea surface temperature (SST) within eddies can be used to identify whether they are subsurface-intensified eddies. The paper proposes a deep learning-based model for subsurface-intensified eddy detection from multi-source remote sensing images. The newly proposed method can distinguish surface- and subsurface-intensified eddy using fusion data of satellite SSH and SST images.

Mesoscale eddies are swirling water exiting ubiquitously in the global ocean. They are significant for material and energy transit and global ocean circulation. Using satellite sea surface height anomaly (SSHA) and sea surface temperature (SST) images, this study proposes a new automated eddy tracking model based on deep learning (DL) technology. Compared with existing eddy tracking algorithms, the DL-based model fuses SSHA and SST data for eddy tracking. We aim to solve problems as eddy splits, mergers, and transient ‘disappearance’ that occurred using SSHA data. The DL-based eddy tracking model is applied in the Kuroshio Extension, and the result verified the model's tracking accuracy and efficiency.

Recent research works for solving partial differential equations (PDEs) with deep neural networks (DNNs) have demonstrated that spatiotemporal function approximators defined by auto-differentiation are effective for approximating nonlinear problems, e.g. the Burger's equation, heat conduction equations, Allen-Cahn and other reaction-diffusion equations, and Navier-Stokes equation. Meanwhile, researchers apply automatic differentiation in physics-informed neural network (PINN) to solve nonlinear hyperbolic systems based on conservation laws with highly discontinuous transition, such as Riemann problem, by inverse problem formulation in data-driven approach. However, it remains a challenge for forward methods using DNNs without knowing part of the solution to resolve discontinuities in nonlinear conservation laws. In this study, we incorporate 1st order numerical schemes into DNNs to set up the loss functional approximator instead of auto-differentiation from traditional deep learning framework, e.g. TensorFlow package, which improves the effectiveness of capturing discontinuities in Riemann problems. In particular, the 2-Coarse-Grid neural network (2CGNN) and 2-Diffusion-Coefficient neural network (2DCNN) are introduced in this work. We use 2 solutions of a conservation law from a converging sequence, computed from a low-cost numerical scheme, and in a domain of dependence of a space-time grid point as the input for a neural network to predict its high-fidelity solution at the grid point. Despite smeared input solutions, they output sharp approximations to solutions containing shocks and contacts and are efficient to use once trained.

The emergent two-dimensional (2D) ferromagnetic materials with unique magnetic properties have endowed great potential for next-generation spintronic devices with extraordinary merits of high flexibility, easy controllability, and high heretointegrability, which is expected to promote the development of Moore's Law continuously. However, it is extremely challenging to realize magnetic switching with ultra-low power consumption at room temperature. Here, we demonstrate the room-temperature spin-orbit torque (SOT) driven magnetization switching in a well-epitaxial all-van der Waals (vdW) heterostructure. The topological insulator Bi2Te3 not only helps to elevate the Curie temperature of Fe3GeTe2 (FGT) through interfacial exchange coupling but also works as a spin current source allowing to switch FGT at a low current density of 2.2 * 106 A cm2. A large SOT efficiency of 0.7 is measured at room temperature, and the thickness of FGT is further adjusted to reduce the influence of the thermal contribution on the second-harmonic signal. Furthermore, the temperature and thickness-dependent SOT efficiency prove that the large SOT in our system mainly originates from the nontrivial origin of topological materials. Our experiment has enabled an all-vdW SOT structure and lays a solid foundation for the implementation of room-temperature all-vdW spintronic devices in the future.

In order to improve the smoothness of the movement process of the arm system, this paper adopts Modelica multi-domain modeling simulation technology based on MWorks platform to analyze and improve the boom system of a concrete pump. Firstly, the mechanical rigid-flexible coupling model and hydraulic model of the arm system of a concrete pump truck are built; in order to study the influencing factors of the arm movement smoothness, stiffness and damping of hydraulic system under the joint arm linkage condition in the arm system are simulated and analyzed; finally, the corresponding optimization and improvement scheme is put forward according to the simulation results. The experimental results show that compared with the original system, the proposed improvement scheme significantly improves the smoothness, which provides theoretical basis and technical support for product improvement.

Although much significant progress has been made in the research field of object detection with deep learning, there still exists a challenging task for the objects with small size, which is notably pronounced in UAV-captured images. Addressing these issues, it is a critical need to explore the feature extraction methods that can extract more sufficient feature information of small objects. In this paper, we propose a novel method called Dense Multiscale Feature Fusion Pyramid Networks(DMFFPN), which is aimed at obtaining rich features as much as possible, improving the information propagation and reuse. Specifically, the dense connection is designed to fully utilize the representation from the different convolutional layers. Furthermore, cascade architecture is applied in the second stage to enhance the localization capability. Experiments on the drone-based datasets named VisDrone-DET suggest a competitive performance of our method., 10 pages, 5 figures

Mesoscale eddies are circular flowing currents that can retain and transport salt, heat, and nutrients all around the ocean. Mesoscale eddies can be detected on remote sensing images, i.e., sea surface height (SSH) images, sea surface temperature (SST) images, chlorophyll concentration images, etc. Most existing automatic eddy detection algorithms are developed based on one kind of remote sensing data. There is a lack of an automatic eddy detection algorithm that can make full use of multi-source remote sensing data to ensure the accuracy and efficiency of eddy detection. The paper proposes a multi-modal U-Net model, a deep neural network-based framework for eddy detection from multi-source remote sensing images. Compared with the previous eddy detection methods, the newly proposed method improves the accuracy and efficiency of eddy detection by using fusion data of SSH and SST.

A novel low-cost gas atomisation technology producing spherical titanium powder (wire induction heating gas atomisation, WIGA) has been developed for additive manufacturing. Combined with the gas a...

Amino acid supplementation is an efficient and effective strategy to increase sperm quality. In our research, a comparative study was conducted to screen free amino acids to improve sperm motility, and we found that leucine was the most efficient one. Leucine treatment increases sperm motility depending on the activation of PI3K/Akt signaling pathway, while the chemical inhibitor of PI3K/Akt signal could reduce the amount of pAkt activated by leucine treatment. Moreover, leucine treatment improved the expression of P62 and LC3-II, substantially suppressed the autophagy process in zebrafish testis. In vitro studies showed that leucine could reduce the fusion of autophagosome and lysosome that was indicated by the co-localization of EGFP-LC3 and lysosome marker. Two chemical modulators of autophagy, such as LY294002 (the inhibitor of PI3K/Akt signal) and chloroquine were administered to investigate the process of autophagy on zebrafish sperm motility. LY294002 inhibited autophagosome formation to reduced sperm motility, while chloroquine inhibited the fusion of autophagosome and lysosome to improve sperm motility. Our data suggest that short-term treatment with leucine could increase zebrafish sperm motility by affecting the autophagy and inhibiting the fusion of autophagosome and lysosomes, depending on the activation of PI3K/Akt signaling pathway.

Hydroxyapatite/Chitosan (HA/CS) composite have significant application in biomedical especially for bone replacement. Inorganic particle shape and size of composite affect the scaffold mechanical property, biological property, and degradation. The aim of this study was to fabricate HA/CS scaffold with good pore connectivity and analyze their biological, degradation properties. Microhydroxyapatite/chitosan (mHA/CS) and nanohydroxyapatite/chitosan (nHA/CS) composite scaffolds with interconnected spherical pore architectures were fabricated. Composite scaffolds structure parameters were analyzed using micro CT. Cell proliferation and morphology were tested and compared between two scaffolds using mouse osteoblastic cell line MC3T3-E1. To research the composite degradation in lysozyme PBS solution, degradation rate and reducing sugar content were tested, and scaffolds morphology were observed by SEM. The results showed that microHA and nanoHA were fabricated by being calcined and synthesis methods, and their infrared spectra are very similar. EDAX composition analysis demonstrated that both of microHA and nanoHA were calcium deficiency HA. Micro-CT results demonstrated the scaffolds had interconnected spherical pores, and the structure parameters were similar. Cell viabilities were significant increased with cultured time, but there were no significant difference between microHA/CS and nanoHA/CS scaffolds. Scaffold structure was gradually destroyed and inorganic composition HA particles are more prominent with degradation time. Significance: (1) Inorganic particle shape and size of composite affect the scaffold mechanical property, biological property, and degradation. NanoHA/CS and microHA/CS scaffolds with good pore connectivity were fabricated and their biological, degradation properties were studied in this manuscript. (2) The scaffold with interconnected porosity construct provides the necessary support for cells to proliferate and maintain their differentiated function, and its architecture related to the structure and morphology of new bone. Polymer scaffolds were fabricated by the technique of compression molding and particulate leaching method, and paraffin microspheres were used as the porogen. (3) MicroHA/CS and nanoHA/CS composite scaffolds are potential materials for use in bone tissue engineering.

This paper proposes a new algorithm for parallel identification of mesoscale eddies from global satellite altimetry data. By simplifying the recognition process and the sea level anomaly (SLA) contours’ search range, the method improves identification efficiency compared with the previous SSH-based method even in the single-threaded process. The global SLA map is divided into several regions. These regions are identified simultaneously with a new SSH-based method. All the eddy identification results of these regions are merged seamlessly into a global eddy map. A β-plane approximation is used to calculate the geostrophic speed in the equatorial band. Compared with the computation complexity of the previous SSH-based method, which is , the computation complexity of the new method is , where K is the number of threads and L is the number of regional SLA maps. When applying the new method to the global SLA map, the computation is ~100 times faster than the previous SSH-based method on an average computer. The new method characterizes an eddy structure by radius, amplitude, eddy core, closed SLA contour, and closed SLA contour with maximum average geostrophic speed. In situ data and another global eddy dataset are applied to validate the reliability of eddies detected by the new algorithm. Global eddy mean properties, variability, and the geographical distribution of both datasets are analyzed to demonstrate the performance of this new method and to help understand eddy activities on a global scale.

In this paper numerical simulation and experimental study was carried out for a tangential inlet air swirl burner which is used for high heat value biomass gas. Study on the different inlet angle effects on Combustion by numerical method. The results of the study show that the 30 degree is the best. The intake angle burner 30 degrees were tested to verify the accuracy of simulation results. Through the comparative analysis of the trajectory and the velocity of the air, the method and the conclusion of the simulation study are reliable.

Miscible displacement of carbon dioxide in low permeability sandstone reservoirs can greatly improve oil recovery. The minimum miscible pressure is the starting point of miscible flooding. Its determination is very important for oilfield development. In this paper, the error analysis of the results measured by empirical formulas is carried out by the internationally accepted slim tube experiment method. Research results show that the minimum miscibility pressure obtained by slender tube displacement experiment is 22.12 MPa. The maximum, minimum and average miscibility pressures predicted by empirical formula method are 32.33 MPa, 7.85 MPa and 19.76 MPa, respectively. The relative error of the average minimum miscibility pressure predicted by all empirical formulas is 10.67%. The least relative error is the Yelling and Metcalfe correlation method, the relative error is 8.82%.

Determining the minimum miscibility pressure is the main research content of CO2 mixed phase displacement. It is a key indicator of whether the formation fluid can reach the miscible phase. There are many ways to determine the minimum miscibility pressure. At present, the laboratory test measures the minimum miscibility pressure of CO2-crude oil. The accepted industry standard is the slim tube test. This paper is aimed at the reservoir profile of Tree 101 block in Yushulin Oilfield. Carry out a slim tube test to determine the minimum miscibility pressure of 32.43MPa.

Major depressive disorder (MDD) is a common disease of serious mental disorders, and it is characterized by the alterations in brain functional connections. Brain network analysis is contributed to understanding the pathological mechanism of MDD. In this paper, we establish a brain network based on the structural Magnetic Resonance Imaging (sMRI) and present a dynamic evolution model of the brain network. The model uses control factors and evolution strategies to simulate the dynamic process of brain lesions of MDD. The experimental results illustrate that our evolution model captures the dynamics of the MDD, which provides a new way for exploring the pathological mechanism of MDD.

Dissertation, This dissertation proposes a family of computational frameworks for fracture modeling in coupled field problems. Fracture mechanics has been a topic of considerable interest for several decades due to the wide existence of fracture in different engineering structures and the important applications of fracture in multiple industries. The present study first develops a continuum-discrete approach to model the fluid-driven fracture of granular media. This approach avoids remeshing by representing the particles as moving interfaces on a fixed background mesh. The effect of particle movement on the flow is characterized by a non-slip boundary condition. A boundary split scheme is proposed to ensure the coercivity of the method. The fluid-driven force on particles is represented by a boundary integral of the viscous drag force around the particles. The corresponding initial-boundary value problem is constructed for the invading fluid and is spatially discretized with the finite element method. A novel quadrature method is developed to handle partial elements that arise due to the mismatch between the mesh and the physical domain. The conditioning issue of partial elements is also addressed in the present study. The present study also aims at developing a general and robust computational approach for the fracture modeling of conventional materials within coupled field problems. We follow the framework of a phase field regularization because of its strength in handling complex fracture patterns. The first application we consider is the fracture simulation of kidney stones during shock wave lithotripsy (SWL), an acoustic-solid-fracture coupling problem. The present study develops a novel computational framework for simulating SWL. The propagation of acoustic pressure is modeled by a wave equation and the deformation of the phantom stone is modeled by the elasto-dynamics equations. The interactions between the acoustic wave and kidney stone is enforced via the continuity condition. The initialization and propagation of fracture within the stone is implicitly represented by the evolution of the phase field.Traditional phase field is designed to model the brittle fracture of homogeneous materials. The present study develops a phase field framework to model fracture propagation in anisotropic and heterogeneous solids. The present model is distinguished from the traditional phase field approach by the fact that it converges to a cohesive type model instead of a Griffith model. A mathematically self-consistent strain energy density functional is proposed that is valid for any anisotropic linear elastic materials. Anisotropy in both the bulk moduli and the crack surface energy are characterized. The model employs multiple phase fields to capture the fracture behavior of the material with more than one preferential cleavage plane. The model develops a novel degradation function which relaxes the strong constraint between the regularization length l and the material properties. The convergence of the model with reducing l and the energy conservation properties of the framework are demonstrated through numerical examples. A robust adaptivity strategy is developed to increase the efficiency of the model. The present framework is applied to model fracture in heterogeneous and anisotropic materials. Coupled with a fine scale analysis, the present model is also used to model the fracture of functionally graded materials.

C12H16CuF6PN8, monoclinic, C2/c (no. 15), a = 13.102(2) Å, b = 9.7902(18) Å, c = 14.727(3) Å, β = 102.336(6)°, V = 1845.3(6) Å3, Z = 4, Rgt (F) = 0.0291, wR ref(F 2) = 0.0733, T = 150(2) K.

This paper presents a methodology that enables projection-based model reduction for black-box high-fidelity models such as commercial CFD codes. The methodology specifically addresses the situation where the high-fidelity model may be a black-box but there is complete knowledge of the governing equations. The main idea is that the linear operator matrix, resulting from the discretization of the linear differential terms is approximated directly using a suitable discretization method such as the Finite Volume Method and requires only the computational grid as input. In this regard, the governing equations are first cast in terms of a set of scalar observables of the state variables, leading to a linear set of equations. By applying the snapshots of the observables to the discrete linear operator, a right hand side vector is obtained, providing the necessary system matrices for the Galerkin projection step. This way an online database of ROMs are generated for various parameter snapshots which are then interpolated online to predict the state for new parameter instances. Finally, the reduced order model is posed as a non-linear constrained optimization problem that can be solved at a significantly cheaper cost compared to the full order model. The method is successfully demonstrated on on a canonical non-linear parametric PDE with exponential non-linearity, followed by the compressible inviscid ow past the NACA0012 airfoil. As a first step, this paper focuses only on establishing feasibility of the method., 36 pages. Submitted to the AIAA Journal

Background Fatigue damage of the long bones is prevalent in running athletes and military recruits due to vigorous mid- and long-term physical activity. The current study attempted to know the features of bony athletic fatigue damage and to explore the mechanism of fatigue damage repair through bone targeted remodeling process. Methods Right ulnae of the Wistar rats were fatigue loaded on an INSTRON 5865 to construct the athletic fatigue damage model, and several time points (i.e. experimental days: 0, 7, 13 and 19) were selected to simulate physiological status, preliminary, mid-term and perennial stage during continuous high-intensive training, respectively. The multi-level responses of rat ulnae under the athletic fatigue loading, including cellular protein expression, micro damage or micro-crack and macro mechanical properties, were tested and statistically analyzed. Results Wistar rats, subjected to the athletic fatigue loading protocol, experienced a decrease of ulna fatigue mechanical properties and an active bone resorption of the loaded ulnae in the early stage, whereafter, a hyperactive bone formation and significant improvements of ulnae fatigue mechanical properties were detected. However, a deterioration of quasi-static mechanical properties in the subsequent period implied limitations of bone remodeling to maintain the bearing capacity of bone during long-term strenuous exercise. Conclusions In summary, after athletic fatigue loading, bone targeted remodeling is activated and proceeds to repair fatigue damage, but only to a certain extent.

For the problem that industrial robot have a poor ability to perceive external environment, and not able to change the way of operation and path of transformation timely as working condition changes. This paper selects color sensor as the medium of recognition, introduces the embedded system Raspberry Pi to 6 axis industrial robot grabbing assignments, and puts forward a six axis industrial robot intelligent recognition model with color perception, dynamic decision-making and other multi-function. In the model, the color sensor signal will be resolved by color processing algorithm in embedded system, and consequently color information of material on the conveyor can be obtained. The color information after processing by the embedded system will be converted to digital signals. Then input the digital signal into robot controller. Finally, robot controller chooses corresponding motion path according to the input signal. The experimental results show that the six axis industrial robot intelligent recognition system can distinguish colors of material with almost 100% accuracy, and adjust operation way and path in 0.72 s time rapidly. Efficiency could be improved by more than 75%.

In this paper, a high efficient differential chaos shift keying (DCSK) based on frequency division multiplexing (FDM) communication system, namely FDM-HEDCSK is proposed. The scheme has two branches and a serial-parallel converter to achieve four attainable bit rates and improve communication security in comparison to DCSK. In this scheme, the combination of two chaos signals is used as the reference signal, on which four bits data are modulated and each transmitter branch transmits its signals together by FDM. The bit error rate (BER) performance is derived based on Gaussian approximation in AWGN channel. Moreover, compared with several systems such as DCSK, correlation delay shift keying (CDSK), with typical spreading factors the BER performance of the proposed scheme can always outperform them at reasonable signal-to-noise ratio (SNR).

Advances in phytoremediation research require the selection of new hyperaccumulators or accumulators and investigations of their tolerance and accumulation of heavy metals. Two concentration gradient experiments (preliminary and verification), involving the same soil Cd concentrations (0, 25, 50, 75, 100, and 125 mg kg−1), were conducted to study Cd accumulation and tolerance in Capsella bursa-pastoris. In the preliminary and verification concentration gradient experiments, the Cd content in shoots of C. bursa-pastoris exceeded 100 mg kg−1 at a soil Cd concentration of 50 mg kg−1, and the maxima of Cd contents in shoots were 307.67 and 369.20 mg kg−1 at 125 mg kg−1 Cd in soil, respectively. The shoot bioconcentration factor of C. bursa-pastoris was greater than 1, but the translocation factor was less than 1.0 in the preliminary and verification concentration gradient experiments. In the verification concentration gradient experiment, the malondialdehyde content and the activities of superoxide dismutase, peroxidase, and catalase increased compared with the control. Therefore, C. bursa-pastoris is a Cd-accumulator with strong tolerance to Cd, and could be used to remediate Cd-contaminated farmland soil in winter. © 2014 American Institute of Chemical Engineers Environ Prog, 2014 34: 663–668, 2015