Search

Zirconia ceramics are widely used in medical and dental industry due to their excellent biocompatibility and aesthetics. In this respect, stereolithography apparatus (SLA) is popular method for 3D printing of ceramic suspensions. However, deformation and cracking of SLA-printed ceramics caused by residual stress seriously limit their application in clinical dentistry. In this investigation, 75 wt. % ZrO2 suspensions were successfully prepared, and ZrO2 ceramic composites were produced via SLA 3D printing. Special attention was paid to the establishment of theoretical model of residual stress in SLA. According to results, the development of residual stress in SLA-printed ceramics depended on various parameters such as laser power, scanning speed, and radius of curvature, as well as linear shrinkage, stress wave, and pure bending stress. The variations of surface residual stress at different laser beam powers, scan speeds, and radii of curvature were found to be consistent with that predicted theoretically. The errors between the theoretical and experimental value were 2.8%, 1.4%, 4.3%, 5.3%, and 4.2%, respectively, when the laser beam powers increased in turn. The errors between the theoretical and the experimental value were 2.5%, 3.9%, 2.8%,2.4%, and 3.8%, respectively, when the scan speeds increased in turn. Thus, theoretical model established was proved high accuracy and conformity to 3D printing standards.

A transmitarray antenna with an elliptical cylindrical shape is presented for a wide-angle multibeam radiation in this communication. The transmitarray has a cylindrical radiating aperture with an elliptical cross-section, namely, elliptical cylindrical shape. Multiple feeds can be placed on the middle horizontal plane to realize multiple beams. Inspired by a 2-D Ruze lens, the antenna shape and the phase compensation are jointly designed according to the desired maximal beam direction. Innovative methods including a feed refocusing analysis and a virtual focal length are utilized to achieve the phase compensation across the 3-D aperture for multiple beam radiations with a small scanning loss. In order to validate the proposed antenna, a prototype operating in the millimeter-wave (mm-wave) $E$ -band has been designed, fabricated, and measured. By changing the position of the feeding gain horn along the refocusing arc, the main beam of antenna can be scanned in 11 directions. The measured peak boresight realized gain is 27 dBi at 70.5 GHz and a beam coverage of ±43° with a less than 2.7 dB scanning loss is obtained.

In this work, novel antibacterial cellulose diacetate/poly(ethylenimine)/hyaluronic acid (CDA/PEI/HA) composite three-dimension (3D) scaffolds were prepared through electrospinning and post-processing. Firstly, CDA/PEI 2D composite nanofiber membranes were prepared by electrospinning, and then the CDA/PEI/HA 3D composite scaffolds were fabricated by post-processing and freeze-drying of CDA/PEI membranes. In particular, HA was added to improve the biocompatibility of composite scaffolds. Compared with CDA/PEI scaffolds, CDA/PEI/HA composite scaffolds showed higher water absorbing rate, higher water retention rate and higher mechanical strength. Moreover, CDA/PEI/HA composite scaffolds with abundant positive charge were benefit for improving antibacterial activity. CDA/PEI/HA scaffolds with biology function of HA were benefit for improving hemocompatibility and cell proliferation of CDA/PEI scaffolds. In summary, antibacterial CDA/PEI/HA scaffolds could protect wound from bacterial infection, improve cellular behavior and accelerate wound healing. Thus, CDA/PEI/HA scaffolds could be potential application for wound dressing.

Video deblurring is a challenging low-level vision task due to variant blur artifacts caused by factors such as depth variations, high-speed movements and camera shakes. Although significant efforts have been devoted to addressing this task, two challenges of capturing temporal patterns and spatial topologies still remain. In this paper, an attention-based interframe compensation scheme is proposed to address the first challenge. The proposed scheme replaces frames in blurry sequences with newly restored frames, and estimates temporal patterns among the replaced sequence to restore the whole sequence. After each replacement, an attention block is employed to exploit dependencies among restored and blurry frames to capture stable temporal patterns. To tackle the second challenge, we propose an adaptive residual block that dynamically fuses multi-level features via learning location-specific weights. Comprehensive experimental results demonstrate that the proposed method achieves state-of-the-art performance in terms of accuracy, visual effect and model size.

Temperature-responsive resistance transition behaviors of the melamine sponges wrapped with different graphene oxide derivatives (i.e. nanoribbon, wide-ribbon and sheet) were investigated. Melamine sponge composites coated by three types of GO derivatives were prepared by a simple dip-coating approach. All these composites show good mechanical flexibility and reliability (almost unchanged compressive stress at 70 % strain after 100 cycles), high hydrophobicity (water contact angle >120°), excellent flame resistance (self-extinguishing) and structural stability even after burning, which was used to construct the resistance-based fire alarm/warning sensor. Notably, the different resistance response behaviors of such sensors are strongly dependent on the GO size and network formed on the MF skeleton surface. Typically, at a fixed high temperature of ∼350 ℃, the three fire alarm sensors show different response time (to trigger the alarm light) of 6.3, 8.4 and 11.1 s for nanoribbon, wide-ribbon and sheet at the same concentration, respectively. The structural observation and chemical analysis demonstrated that the discrepancy of temperature-responsive resistance transition behaviors of various GO derivatives was strongly determined by their different thermal reduction degrees during the high-temperature or flame treating process. This work offers a design and development for construction of smart fire alarm device for potential fire prevention and safety applications.

Multifunctional nanodevices have shown great promise as robust tools in cancer theranostics. Carbon nanohorns (CNHs) of large specific surface area, low toxicity and unique physicochemical properties are among one of ideal platforms to fabricate the nanodevices. Herein, we develop a water-dispersible carbon nanohorn−polyglycerol−gold (CNH-PG-Au) nanohybrid through surface engineering of CNHs, which is further loaded with chemotherapeutic drug doxorubicin (DOX) to afford DOX@CNH-PG-Au nanodevice. Benefiting from the excellent hydrophilicity of PG, as-prepared multifunctional DOX@CNH-PG-Au possesses good aqueous dispersibility and colloidal stability enabling biomedical applications. DOX@CNH-PG-Au nanoparticles are taken up by 4T1 mouse breast cancer cells and mainly disperse in the lysosomes, where the mild acidic condition promotes the release of DOX. DOX@CNH-PG-Au induces more cell apoptosis under X-ray irradiation than DOX@CNH-PG of the same DOX concentrations, which is attributed to increased DNA damage by synergistic effect of DNA intercalation of DOX and radiosensitization of Au nanoparticles. Upon intravenous administration, DOX@CNH-PG-Au nanoparticles steadily accumulate in 4T1 tumor through enhanced permeability and retention (EPR) effect, and the tumor can be clearly observed by photoacoustic imaging. In vivo tumor treatment using DOX@CNH-PG-Au together with X-ray irradiation effectively inhibits tumor growth through radiochemotherapy. Our work demonstrates the great potential of DOX@CNH-PG-Au nanodevice in cancer theranostics.

Single-atom catalysis recently attracts great attentions, however, whether single atom or their nanoparticle (NP) has the advantage in its intrinsic activity remains under heated debate. Ag/Al2O3 is a widely used catalyst for many catalytic reactions, while the effect of Ag particle size on the activity is seldom investigated due to the great difficulty in synthesizing single atom Ag and Ag clusters/particles with different sizes. Herein, we firstly prepared an atomically dispersed Ag/Al2O3 catalyst using a nano-sized γ-Al2O3 as the support, subsequently obtained a series of Ag0/Al2O3 catalysts with different Ag particle sizes by H2 reducing single-atom Ag/Al2O3 catalyst at various temperatures. The Ag0/Al2O3 treated at 600 °C demonstrated superior CO oxidation performance over single-atom Ag/Al2O3 and the Ag/Al2O3 treated at 400 and 800 °C. Based on experimental data and density functional theory (DFT) calculation results, we reveal that the larger Ag0 particle is beneficial to oxygen activation and improves the valence stability during oxidation reaction, while the aggregation of Ag0 particle also accordingly decreases the concentration of surface active sites, hence, there is an optimum Ag0 particle size. Our findings clearly confirm that Ag0 nanoparticle has the advantage over single-atom Ag species in its intrinsic activity for CO oxidation.

Graphite carbon nitride (g-C3N4), a two-dimensional polymer semiconductor material, has good semiconductor properties, suitable electronic energy band structure, excellent physical and chemical stability. It is widely used in the field of energy and materials science such as photoelectric conversion. In this paper, the progress of g-C3N4 in dye sensitized solar cells (DSSC) and perovskite solar cells (PSC) was reviewed. As a new semiconductor material, g-C3N4 has the advantages of simple preparation, abundant amino and Lewis basic groups. Therefore, on the basis of excellent structure of g-C3N4, its electronic and optical properties are utilized to further expand its application in the field of photoelectric conversion.

Carbon nitride (CN) polymer is considered as a potential material for photocatalysis, but surfers from the low photoconversion efficiency due to its high defect density. Although the defects can be limited through the crystallization of CN, the photocatalytic activity of single-component photocatalyst is still restricted by the recombination of photoexcited carriers. In our work, the ordered Schottky heterojunction of heptazine-based crystalline carbon nitride (HCN) and Ti3C2 MXene was successfully prepared through ionothermal method. The HCN/Ti3C2 composites exhibit higher photocatalytic performance than pristine HCN. Particularly, the HCNT20 sample exhibits the highest photocatalytic H2 evolution activity, which is about 8 and 2 times higher than that of bulk CN and pristine HCN, respectively. The results of charge transfer dynamics and DFT calculation reveal that the photocatalytic enhancement mechanism is primarily attributed to the synergistic effects of crystallization of CN, excellent conductivity of Ti3C2 and well-constructed Schottky heterojunction.

Yttrium borides have attracted much attention for their superconducting and high chemical stability. However, the fundamental knowledge of the mechanical properties and hardness of yttrium borides is rather lack. Here, we performed a systematic investigation on yttrium borides with various stoichiometries based on an unbiased CALYPSO structure search method and first-principles calculations. A new YB6 compound with R3m hexagonal structure is observed, which is more stable than the experimental synthesized Pm3m phase. The calculated formation enthalpy, elastic constants, and phonon dispersions distinctly show that the R3m-YB6 is energetically, mechanically, and dynamically stable. The density of states and electronic band structure reveal that all the stable yttrium borides are metallic. Based on a semiempirical method, the Vicker hardness of R3m-YB6 is expectant to be 37.0 GPa, indicating it is a potential ultrahard metal. Our results provide insights for future synthesis and design of new type functional materials.

Deep neural networks have demonstrated excellent performance in most computer vision tasks in recent years. However, they are vulnerable to adversarial perturbations generated by adversarial attacks. These human-imperceptible perturbations often lead to severe distortion in the high-dimensional intermediate feature space, which is one of the major reasons for the vulnerabilities in deep neural networks. Therefore, input images with perturbations can completely change the predictions of the networks in the decision space. To overcome this drawback, we propose to progressively align the intermediate feature representations extracted from the adversarial domain with feature representations extracted from a clean domain through domain adaptation. The difference between two feature distributions can be accurately measured via an optimal transport-based Wasserstein distance. Thus, the deep networks are forced to learn robust and domain-invariant feature representations, so that the gap between the different domains is minimized and that the networks are no longer easily fooled by diverse adversaries. Extensive evaluations are conducted on four classification benchmark datasets in white-box attack scenarios. The evaluation results demonstrate a significant performance improvement over several state-of-the-art defense methods.

Inhomogeneous structures have significant impact on the drag. In this work, direct numerical simulations of flows past fluidized spheres are performed to investigate the effect of inhomogeneities on the drag at low Reynolds numbers. The inhomogeneities are quantified using various sub-grid quantities including a newly defined quantity named the solid drift flux, which is a measure of the correlation between the gas volume fraction and the particle phase velocity. Following the methodology of Rubinstein et al. (2017), a new drag model that depends simultaneously on the solid drift flux and the scalar variance of solid volume fraction is then developed. Given that these two sub-grid quantities are readily available in Eulerian-Lagrangian (EL) simulations, the developed model can be directly used in EL simulations without any additional closures for sub-grid quantities. An a priori analysis demonstrates that this model achieves satisfactory performance in predicting the drag in inhomogeneous structures.

To convert lignin into high-valued carbon materials and understand the lignin structure function, oxidized lignin, a by-product from lignocellulose PHP-pretreatment (phosphoric acid plus hydrogen peroxide), was carbonized by one-step KOH-activation; the physico-chemical characteristics and electrochemical performances of the harvested carbons were also investigated. Results indicated the resultant carbons displayed 3-dimensional hierarchical porous morphology with maximum specific surface area of 3094 m2 g−1 and pore volume of 1.72 cm3 g−1 using 3:1 KOH/lignin ratio for carbonization. Three-electrode determination achieved a specific capacitance of 352.9 F g−1 at a current of 0.5 A g−1, suggesting a superior rate performance of this carbon. Two-electrode determination obtained an excellent energy density of 9.5 W h kg−1 at power density of 25.0 W kg−1. Moreover, 5000 cycles of charge/discharge reached 88.46% retention at 5 A g−1, implying an outstanding cycle stability. Basically, low molecular weight and abundant oxygen-containing functional groups of employed lignin mainly related to the excellent porous morphology and the outstanding electrochemical performances, suggesting the oxidized lignin was an ideal precursor to facilely prepare activated carbon for high-performance supercapacitor. Overall, this work provides a new path to valorize lignin by-product derived from oxidative pretreatment techniques, which can further promote the integrality of lignocellulose biorefinery.

T-junction is widely used as a phase separation component to separate two-phase flow because of its simple structure and low cost. Previous studies suggest that an increase in the number of T-junction branches is conducive to raise phase separation efficiency. In this paper, the pressure distribution and phase separation performance of two-phase flow of R134a in horizontal double T-junctions are studied. The phase separation performance between two branches is compared. The parameters including inlet quality (0.3-0.6), inlet mass flow rate (10.0 g s−1-25.0 g s−1), diameter ratio (1, 0.75) and the distance between two branches (0.3 m-2.6 m) are studied. The inlet mass flow rate has significant influence on the phase separation, and the separation performance reaches best when it is 10.0 g s−1. Results show that the change of parameters has greater influence on the phase separation performance of the second branch compared with the first branch. When the distance between branches is 2.0 m, the total phase separation capacity of the branches reaches the best. By comparing the phase separation of branches, it can be found that when the distance increases gradually, the phase separation performance of the second branch first increases and then decreases, indicating that there is a specific distance that maximizes the gas phase separation efficiency of branches. The optimal distance is also related to the mass flow ratio of the second branch. Therefore, an empirical correlation equation which can accurately predict the gas phase separation efficiency of the second branch is proposed.

A bi-directional photonics-aided millimeter-wave (MMW) transmission system at E-band is proposed in this article, which can be applied in future mobile networks. Only one pair of horn antennas are adopted in the experiment to support symmetrical wireless signal transmission. The center frequency of the downlink and uplink is set as 73.5 and 83.5 GHz, respectively, to avoid the interference effect. Moreover, the probabilistic shaping (PS) technique is used to adapt transmission rate and improve system capacity under a limited SNR. The time-domain Volterra series based equalizer effectively compensates the nonlinear distortion induced by the photoelectric devices. Based on the experimental results, the proposed full-duplex MMW-RoF system can support OFDM truncated PS (TPS)-4096QAM signal transmission, whose entropy is 10 bits/symbol/Hz. We set the probability of the outermost points to zero and the generated TPS-4096QAM signal is actually a PS-1936QAM one. The system simultaneously allows 7 Gbaud TPS-4096QAM signal downlink and uplink transmission over 2 m wireless distance, satisfying the NGMI threshold at 0.83. Moreover, the bi-directional 7 Gbaud PS-1024QAM signal transmission distance can be extended to 13.4 m when we adopt a higher NGMI threshold at 0.67. So the total rate of our proposed bi-directional transmission system is achieved to 100 Gbit/s. We also investigate the BER performance with different modulation formats. Such a system is a promising solution for future high-speed backhaul networks with multiple modulation format requirements.

Learning-based methods have attracted considerable interest in image dehazing. However, most existing methods are not well adapted to different hazy conditions, especially when dealing with the heavily hazy scene. There is often a significant amount of haze that remains in the images recovered by most methods. To address this issue, we propose an end-to-end Haze Concentration Adaptive Network (HCAN), including a pyramid feature extractor (PFE), a feature enhancement module (FEM), and a multi-scale feature attention module (MSFAM) for image dehazing. Specifically, PFE based on the feature pyramid structure leverages complementary features from different CNN layers to help the clear image prediction. Then, FEM fuses four kinds of images with different haze density (i.e., three recovered images in the FEM with light haze density, and the input hazy image with strong haze condition) to guide the network to adaptively perceive images under different haze conditions. Finally, MSFAM is designed under two principles, multi-scale structure and attention mechanism. It is used to help the network produce a clear image with more details, and ease the network training. Comprehensive experiments demonstrate that the proposed HCAN performs favorably against the state-of-the-art methods in terms of PSNR, SSIM, and visual effect. The results, per-trained models and code are available at https://github.com/TaoWangzj/HCAN .

Clean-energy substitution technology for existing residential buildings in cities is an inevitable choice for sustainable development and low-carbon ecological city construction. In this paper, the current status of energy-saving renovation and renewable-energy applications for existing residential buildings in various cities in China was summarized by using statistical methods. The geographical distribution of clean-energy power generation in primary energy production in China was explored in depth. According to different climatic divisions for existing urban residences, clean-energy production and consumption were analyzed and predicted based on the STIRPAT model. The results show that the energy consumption of urban residential buildings in 2016 increased by 43.6% compared with 2009, and the percentage of clean energy also increased from 7.9% to 13.4%. Different climatic regions have different advantages regarding clean energy: nuclear power generation leads in the region that experiences hot summers and warm winters, whereas wind and solar power generation lead in the cold and severely cold regions. The present results provide basic data support for the planning and implementation of clean-energy upgrading and transformation systems in existing urban residences in China.

Millimeter-wave (MMW) transmission has the advantage of larger available bandwidth compared with traditional wireless communications. Compared with the electrical algorithms, photonics-aided MMW signals generation can support a large modulation bandwidth and is regarded as a potential candidate for future RoF link. In this experiment, we demonstrate a W-band (75 GHz ∼ 110 GHz) photonics-aided MMW communication system. To overcome the high-frequency power loss induced by the limited bandwidth of photoelectric devices, we adopt orthogonal frequency division multiplexing (OFDM) signal with hybrid quadrature amplitude modulation (QAM) formats. Moreover, the probabilistic constellation shaping (PCS) technique is used to further improve the system capacity. In the offline digital signal processing (DSP), a time-domain Volterra series based nonlinear equalizer with I/Q multiple input multiple output (MIMO) structure can compensate for the nonlinearity during the fiber and wireless transmission. Finally, a hybrid time-frequency domain least mean square (LMS) equalizer is proposed to eliminate the crosstalk in the adjacent OFDM symbols and subcarriers. With the aid of advanced modulation and hybrid time-frequency domain equalization, 30 Gbaud OFDM MMW signals with hybrid PS-512QAM and PS-128QAM at 92.5 GHz can be successfully transmitted over 1-m wireless link, whose raw transmission rate is 208.4 Gbit/s. It is worth noting that the hybrid time-frequency domain LMS equalizer can increase the maximal AIR to 178.8 Gbit/s.

This paper develops a methodology for simultaneously modeling lane-changing and car-following behavior of automated vehicles on freeways. Naturalistic driving data from the Safety Pilot Model Deployment (SPMD) program are used. First, a framework to process the SPMD data is proposed using various data analytics techniques including data fusion, data mining, and machine learning. Second, pairs of automated host vehicle and their corresponding front vehicle are identified along with their lane-change and car-following relationship data. Using these data, a lane-changing-based car-following (LCCF) model, which explicitly considers lane-change and car-following behavior simultaneously, is developed. The LCCF model is based on Gaussian-mixture-based hidden Markov model theory and is disaggregated into two processes: LCCF association and LCCF dissociation. These categories are based on the result of the lane change. The overall goal is to predict a driver’s lane-change intention using the LCCF model. Results show that the model can predict the lane-change event in the order of 0.6 to 1.3 s before the moment of the vehicle body across the lane boundary. In addition, the execution times of lane-change maneuvers average between 0.55 and 0.86 s. The LCCF model allows the intention time and execution time of driver’s lane-change behavior to be forecast, which will help to develop better advanced driver assistance systems for vehicle controls with respect to lane-change and car-following warning functions.

The combination of yttria-zirconia (3Y–ZrO2) and stereolithography (SLA) is essential for forming alternative bioceramics for complex, custom-designed dental implants. In addition, suitable bioceramic suspensions are preconditions for such fabrication using SLA technology. In this study, 3Y–ZrO2 and 3Y–ZrO2/graphene oxide (GO) powders dispersed in ultraviolet curable acrylic-based resins are developed. A sedimentation experiment is performed to evaluate various ratios of monomers and dispersants to ensure good stability and dispersion of the suspensions. The hydrophobic structures of KH560 are introduced on the surfaces of ceramic powders, representing better wettability performances with 40 vol% NPG2PODA and 60 vol% Di-TMPTA. Moreover, the effects of premixed solvents and solid contents on the rheological behaviors of suspensions are investigated to confirm shear thinning and pseudo plasticity, with the experimental results fitting the Krieger–Dougherty curves. Further, the cure depth and width of the ceramic suspensions are studied, the results of which show that the penetration depth and critical exposure of the 3Y–ZrO2 (3Y–ZrO2/GO) pastes are 17.2 ± 0.50 μm and 4.80 ± 0.62 mJ/cm2 (11.8 ± 1.20 μm and 22.80 ± 9.34 mJ/cm2), respectively. The addition of GO results in a sharp increase in critical exposure, which intensifies the light scattering and hinders the curing of the ceramic paste. The research is expected to serve as a technical guide for the fabrication of ceramic suspensions for dental implants using SLA technology.

Middle and deep geothermal resources are abundant and have the advantages of being clean, stable, and reliable. However, a widely recognised theoretical model of deep ground heat exchangers is currently lacking owing to the uncertainty of the underground geotechnical environment and the complex heat transfer mechanism of such heat exchangers. Therefore, considering the stratification of the rock and soil layers, a heat extraction model of a coaxial deep borehole heat exchanger was established in this study. In addition, MATLAB software was used to conduct an unsteady-state numerical simulation based on the finite difference method, and the simulation results is verified through a long-duration comparison with actual measured engineering data. It is observed that (1) the numerical calculation results are in agreement with the actual engineering measurement results; (2) under the test conditions, the radius of influence of the deep borehole heat exchanger on the temperature of the surrounding rock and soil layer is approximately 7 m after continuous operation throughout the heating season; (3) the heat extraction increases with increasing circulation flow rate and outer diameter of the buried pipe; and (4) the linear meter heat exchange power should not be greater than 150 W. This study provides important guidance for research on coaxial deep borehole heat exchangers.

Integration of each aspect of the manufacturing process with the new generation of information technology such as the Internet of Things, big data, and cloud computing makes industrial manufacturing systems more flexible and intelligent. Industrial big data, recording all aspects of the industrial production process, contain the key value for industrial intelligence. For industrial manufacturing, an essential and widely used electronic device is the lithium-ion battery (LIB). However, accurately predicting the remaining useful life (RUL) of LIB is urgently needed to reduce unexpected maintenance and avoid accidents. Due to insufficient amount of degradation data, the prediction accuracy of data-driven methods is greatly limited. Besides, mathematical models established by model-driven methods to represent degradation process are unstable because of external factors like temperature. To solve this problem, a new LIB RUL prediction method based on improved convolution neural network (CNN) and long short-term memory (LSTM), namely Auto-CNN-LSTM, is proposed in this article. This method is developed based on deep CNN and LSTM to mine deeper information in finite data. In this method, an autoencoder is utilized to augment the dimensions of data for more effective training of CNN and LSTM. In order to obtain continuous and stable output, a filter to smooth the predicted value is used. Comparing with other commonly used methods, experiments on a real-world dataset demonstrate the effectiveness of the proposed method.

A series of triphenylamine (TPA) derivatives with various substituent groups were prepared and showed different absorption and fluorescence characteristics due to the substituent effect. On account of the existence of pyridine units, these TPA derivatives exhibited acid-induced tunable multicolor fluorescence emission including white light emission. In addition, acid-induced fluorescence regulation of these compounds has been also realized in the solid state, which enable them to be successfully constructed the stimuli-responsive fluorescent films and fluorescent inks for inkjet printing.

As a pre-processing step of computer vision applications, single image dehazing remains challenging due to existing inefficiencies in the restoration of content and details. In this paper, the self-supporting dehazing network (SSDN) is proposed to overcome these two problems. For the restoration of image content, the self-filtering block is introduced to remove redundant features, hence improving the representation abilities of learned features. For the recovery of image details, a novel self-supporting module is proposed as a crucial component of the proposed SSDN. With this module, the complementary information among support images that are transformed from multi-level features is explored. By incorporating such information, the self-supporting module can learn more intrinsic image characteristics and generate fine-detail images. Experimental results demonstrate that the proposed SSDN outperforms state-of-the-art dehazing methods in terms of both quantitative accuracy and qualitative visual effect.

We study the problem of Reinforcement Learning from Demonstrations (RLfD), where the agent has access to not only reward signals from the environment, but also some available expert demonstrations. Recent works absorb ingredients from imitation learning and utilize demonstration data as reward reshaping. Despite their success, these methods update policy over these states seen in the demonstration data, in the same way as other states in the state space, overlooking the validity of direct supervision signals on these states. To address this issue, we propose a novel RLfD objective function with a new shaping reward, by optimizing which can directly leverage the supervision signal on these demonstrated states. We propose a general framework for policy optimization of the proposed objective, with convergence guarantees under the classic tabular setting. Based on that, we further make some approximations based on deep neural networks, and then introduce a new practical algorithm, called Demonstration Actor Critic (DAC) in large continuous domains. Extensive experiments on a range of popular benchmark sparse-reward tasks show that our method can lead to significant performance gains over several strong and off-the-shelf baselines.

Dental implants are in direct contact with living tissue, therefore, biological performance as well as mechanical properties and tribological behavior of implant materials are important and must be systematically investigated. In our study, 3Y–ZrO2 was modified with different concentrations of bioactive, biocompatible graphene oxide (GO). The 3Y–ZrO2/GO composites were fabricated by hot-press sintering. Compared to raw 3Y–ZrO2, addition of GO sharply increased bending strength and fracture toughness, representing an increase of 200% and 41%, respectively. Crack deflection, crack bridging, and GO pull-out were main toughening mechanisms, associated with the formation of a C–O–Zr bond. The addition of GO reduced coefficient of friction and wear rate and lowered surface roughness of 3Y–ZrO2, mainly owing to self-lubricating properties of the composites. In addition, thin film of hydroxyl groups (OH-) on the surface of composites improved wetting properties, as indicated by low contact angle. GO did not cause any obvious signs of cytotoxicity and promoted proliferation, growth, and adhesion of cells. Fully elongated cell morphology was observed on composites. Overall, 3Y–ZrO2/GO composite ceramic with 0.15 wt% GO is promising biomaterial for dental implant applications.

Hydrogel-based wearable devices have attracted tremendous interest due to their potential applications in electronic skins, soft robotics, and sensors. However, it is still a challenge for hydrogel-based wearable devices to be integrated with high conductivity, a self-healing ability, remoldability, self-adhesiveness, good mechanical strength and high stretchability, good biocompatibility, and stimulus-responsiveness. Herein, multifunctional conductive composite hydrogels were fabricated by a simple one-pot method based on poly(vinyl alcohol) (PVA), sodium alginate (SA), and tannic acid (TA) using borax as a cross-linker. The composite hydrogel network was built by borate ester bonds and hydrogen bonds. The obtained hydrogel exhibited pH- and sugar-responsiveness, high stretchability (780% strain), and fast self-healing performance with healing efficiency (HE) as high as 93.56% without any external stimulus. Additionally, the hydrogel displayed considerable conductive behavior and stable changes of resistance with high sensitivity (gauge factor (GF) = 15.98 at a strain of 780%). The hydrogel was further applied as a strain sensor for monitoring large and tiny human motions with durable stability. Significantly, the healed hydrogel also showed good sensing behavior. This work broadens the avenue for the design and preparation of biocompatible polymer-based hydrogels to promote the application of hydrogel sensors with comfortable wearing feel and high sensitivity.

The design and synthesis of low-cost and efficient bifunctional electrocatalysts for water splitting are critical and challenging. Hereby, a bimetallic phosphide embedded in a N and P co-doped porous carbon (FeCoP2@NPPC) material was synthesized by using sustainable biomass-derived N- and P-containing carbohydrates and non-noble metal salts as precursors. The obtained material exhibits good catalytic activities in hydrogen evolution reaction (HER), oxygen evolution reaction (OER), and overall water splitting. The bimetallic alloy phosphide (FeCoP2) is the active site for electrocatalysis. Theoretical calculation indicates that the sub-layer Fe atoms and top-layer Co atoms in FeCoP2 exhibit a synergistic effect for enhanced electrocatalytic performance. The carbon matrix around the FeCoP2 can prevent the corrosion during the catalytic reactions. The hierarchically porous structure of the FeCoP2@NPPC material can promote the transfer of electrons and electrolyte, and increase the contact area of the active sites and electrolytes. N- and P-containing functionalities improve the wetting and conductivity properties of the porous carbon. Due to the synergistic effects, FeCoP2@NPPC requires a low overpotential of 114 and 150 mV at the current density of 10 mA cm-2 for HER in 0.5 M H2SO4 and 1.0 M KOH, and an overpotential of 236 mV for OER in 1.0 M KOH solution, which are much lower than those of FeP@NPPC and CoP@NPPC. Based on the density functional theory calculation, FeCoP2 yields the smallest Gibbs free energy change of rate-determining step among the samples, which leads to better electrochemical performances. In addition, when FeCoP2@NPPC was used as a bifunctional catalyst in water splitting, the electrolyzer needed a low voltage of 1.60 V to deliver the current density of 10 mA cm-2. Furthermore, this work provides a strategy for preparing sustainable, stable, and highly active electrocatalysts for water splitting.

A high-efficiency conformal transmitarray with ultrathin dual-layer Huygens element is developed. The element consists of “I” shape patches for magnetic response and “T” shape stubs for electric response printed on two metal layers of a single substrate with only 0.5 mm thickness ( $\lambda _{0}$ /60 at 10 GHz). By tuning the magnetic and electric responses, the transmitting phase of the element can be changed. Eight elements are designed to cover quantized 360° phase range with a maximal 1.67 dB loss. Then, the proposed elements are employed in a small conformal transmitarray design. To improve the antenna efficiency, the elements’ dimensions are calculated by considering the oblique incidence effects. Finally, a cylindrically conformal transmitarray with a larger aperture size is simulated, fabricated, and measured. It can achieve a measured gain of 20.6 dBi with a 47% aperture efficiency.

The clean transformation of the China’s energy structure is related to the establishment of China as an energy power in the new world pattern. The multi-energy systems based on renewable energy, especially solar energy and biomass energy, have become a popular energy supply mode. However, there is no clear theoretical guidance for the quantitative source integration of these two types of clean energy. Concretely, whether the source grades of solar energy and biomass energy are similar in different solar energy modes has not been considered yet. Aiming at this problem, based on the theory of space-time unity of the special theory of relativity, a new principle of source integration is proposed, and the corresponding mathematical model is established. The theoretical calculation of the combination of solar energy and different types of biomass energy is carried out. Under the premise of the similarity between the source grades of different biomass energy and solar energy, the specific solar energy concentration ratio is obtained. According to this principle, the matching efficiency of the multi-energy integration for rice and corn plants can be increased by 39.2% and 36.2%, respectively.

Occurrence and flow of hydrocarbons in nanopores are two important issues in the effective exploitation of shale oil reservoirs. In this study, molecular dynamics simulations are employed to investigate the mechanisms about occurrence and flow of octane in slit-shaped quartz nanopores. We show that the occurrence state of octane and, therefore, its flow behavior are profoundly affected by the potential field from quartz walls and adsorption layers if the nanopore width w becomes less than 50 Å. Two main adsorption layers are always formed, adjacent to the walls and independent of w, due to two potential wells generated by the attractive potentials of the walls. Each pair of symmetrical adsorption layers, each of which can be considered as a solid-like surface, forms a confined environment similar to a nano-slit. The attractive potentials from them are found to be the cause for the formation of the adsorption layers between them. The obvious bulk phase of octane is formed in the pore of w = 50 Å due to the wide zero potential barrier induced by the innermost two adsorption layers. The nonlinear dependence of flow rate on pressure gradient shows that Darcy's law fails to describe the flow in the nanopore. The non-Darcy behavior mainly arises from adsorption effects from the walls and the adsorption layers, slippage between octane and walls and between adjacent two adsorption layers, and the molecular exchange between adsorption layers. A modified microscopic model is established to predict the dependence of flow rate on potential field, pressure gradient and w, which is in a good agreement with our simulation results and verified by the dodecane flow through the nanopore. Our work can be of great importance for revealing the mechanisms of occurrence and transport and guiding the estimation and exploitation of shale oil resources.

Zirconia ceramics and carbon-based materials are widely adopted in medical and dental applications due to their excellent biocompatibility and aesthetics. However, fracture toughness of ceramic materials limits their application in clinical dentistry because of the existence of residual stress. In this study, zirconia/graphene oxide (ZrO2-GO) composite ceramics were fabricated by hot-press sintering. Residual stresses developed on the surface of ZrO2-GO composite ceramics were evaluated by X-ray residual stress analysis and indentation techniques. The variation of surface residual stress with GO content was evaluated, and found to be consistent with that of fracture toughness. The generation of residual stress was found to be directly related to fracture toughness. Residual stress calculated by theoretical formula of indentation method was consistent with that measured by X-ray diffraction in line with the content of GO. Based on above results, it is concluded that 0.1–0.15 wt% GO composite ceramics possessed better mechanical properties.

Chiral assemblies by combining natural biomolecules with plasmonic nanostructures hold great promise for plasmonic enhanced sensing, imaging, and catalytic applications. Herein, we demonstrate that human serum albumin (HSA) and porcine serum albumin (PSA) can guide the chiral assembly of gold nanorods (GNRs) with left-handed chiroptical responses opposite to those by a series of other homologous animal serum albumins (SAs) due to the difference of their surface charge distributions. Under physiological pH conditions, the assembly of HSA or PSA with GNRs yielded left-handed twisted aggregates, while bovine serum albumin (BSA), sheep serum albumin, and equine serum albumin behaved on the contrary. The driving force for the chiral assembly is mainly attributed to electrostatic interaction. The opposite chiroptical signals acquired are correlated with the chiral surface charge distributions of the tertiary structures of SAs. Moreover, the chirality of the assembly induced by both HSA and BSA can be enhanced or reversed by adjusting the pH values. This work provides new insights into the modulation of protein-induced chiral assemblies and promotes their applications.

The metal salts KSCN, Ni(CH3COO)2·4H2O, and auxiliary ligand pyridine were added respectively to the salamo-like ligand H2L, and it is expected that the Ni(II) complex coordinated by H2L will be ob...

Due to complex and changeable driving cycles in urban roads, it is a challenging task for most of the current control strategies utilized in vehicles to adapt to the driving environment. At the same time, hardware requirements for storing and processing a massive amount of streaming data are increasing, which lead to excessive accumulated errors and high computational cost. To deal with this problem, an innovative prediction method, which is based on Markov chain and data stream mining, is proposed to predict the future driving cycle of vehicles. State transition probability matrix is updated in real time with data stream mining technology, and every time a new record arrives, the expired record is replaced by the new arrived one in the memory, and both state division and the sizes of the sliding window can be adjusted adaptively based on prediction accuracy for the changing driving cycles. The results show that the proposed method is more suitable for predicting changing driving cycles, which is able to maintain better prediction accuracy than the traditional method. In addition, based on the proposed method, the memory space utilized for storing temporary records were saved largely, and the calculation resource required was reduce.

The study of evacuation for buildings with limited space is an important part of improving evacuation efficiency and preventing stampedes. A building evacuation model was proposed based on cellular automata simulation considering different crowd states. Different flow sizes under layout environments with the same facilities as well as evacuation efficiency, bottleneck area density, and escape routes choice under the orderly and disorderly distribution conditions have also been analyzed. The results show that the initial disorderly distribution state is superior to the orderly distribution state in terms of the evacuation efficiency index. The former provides evacuees with maximum room for the corridor and the exit, with the overall evacuation density being lower than that of the latter. Evacuation along the central corridor provides more room compared to that of the two flanks, which explains why evacuees prefer to occupy the central area when space is limited, and this is detrimental to the moving capacity.

This paper presents a simple microwave non-destructive method to assess the uniformity of fiber distribution in the steel fiber reinforced concrete (SFRC). To simplify the assessment, only the ampl...

The h -good neighbor connectivity κ h ( G ) is an important parameter to evaluate the fault tolerance of an interconnection network G . So far, almost all known results of κ h ( G ) are about small h ′ s except the hypercubes, the star graphs, the k -ary n -cubes and hierarchical hypercube network H H C n et al. In this paper, we focus on κ h ( B n , k ) for the ( n , k ) -bubble-sort network B n , k , which is a generalization of the bubble-sort graph B n as it is a special case of ( n , k ) -bubble-sort networks for k = n − 1 . We show that κ h ( B n , k ) = n + h ( k − 2 ) − 1 for 2 ≤ k ≤ ⌈ n + 3 2 ⌉ and 0 ≤ h ≤ ⌊ n − 3 2 ⌋ , which implies that at least n + h ( k − 2 ) − 1 vertices of B n , k have to be deleted to get a disconnected graph without vertices of degree less than h .

Carbon Capture, Utilization and Storage (CCUS) is regarded as one of the most economical solutions to achieve reasonable emission reductions for small and medium-sized point sources in the IEA Clean Technology Scenario (CTS). In order to reduced dependency of fossil energy, the solar-assisted post-combustion CO2 capture system are proposed and studied in past decade. Nevertheless, the most of existing feasibility studies on solar-assisted CO2 capture system are steady-state analysis under certain fixed working conditions, which tend to ignore the impact of fluctuations in weather conditions and even lead to a misguiding on feasibility study. Hence, the dynamic simulation model is necessary as it could provide more detailed information and close to the actual situation. This study presents a design of a solar-assisted 5-step temperature-vacuum swing adsorption CO2 capture (SOL-TVSA) system for an industrial emission source. Verified by experiments of core unit, a dynamic simulation system is established and the influences of cyclic parameters on dynamic performance are evaluated. Results show that when vacuum pressure is lower than 6 kPa and desorption temperature exceeds 70 °C, the recovery rate and purity of TVSA system can exceed 95% and 85% respectively. When the volume-area ratio is below 0.3, the annual amount of CO2 captured by SOL-TVSA system could exceed 2.11 ton and solar fraction can exceed 38.81%. The maximum values of coefficient of performance and exergy efficiency can reach 2.20 and 9.81% respectively.

Redox-responsive ferrocene (Fc)-containing hydrogels have attracted increasing interest for their potential applications in materials science, biomedicine and catalysis. Herein, we report the synthesis of a novel polynorbornene-based tetrablock terpolymer containing consecutively aldehyde, Fc, and dentritic triethylene glycol and aldehyde groups in the side chain by the ring-opening metathesis polymerization with the aid of the 3rd generation Grubbs catalyst. Owing to the presence of Fc and triazole groups, this copolymer could be used as both reducing and stabilizing agents for the formation of silver nanoparticles (AgNPs). Also, this copolymer could be used as a crosslinking agent to fabricate gelatin hydrogels based on the Schiff's base reaction between the aldehyde groups and the side-chain amino groups of gelatin. Based on the above properties, the AgNPs composites of gelatin hydrogels were further successfully fabricated by blending this copolymer, gelatin and AgNO3 together or soaking this Fc-containing gelatin hydrogels into the AgNO3 solution, in which the AgI was in situ reduced into Ag0 by the Fc groups. Results of inhibition zone experiments confirmed the antibacterial activities of the obtained composites against both Escherichia coli and Staphylococcal aureus.

Carbon dots (CDs) have received increasing attention as fluorescence sensors for the detection of different metal ions and organic small molecules. In this work, κ-carrageenan (κ-Carr)-derived carbon dots (C-CDs) were synthesized via a facile hydrothermal treatment with κ-Carr as a carbon source. The produced C-CDs exhibited a high fluorescence with fluorescent quantum yield up to 20.6% and excellent fluorescent stability under diverse conditions (UV radiation, pH value and ionic strength). Then, the C-CDs were utilized as dual functional sensing probe to detect both Fe3+ and oxytetracycline (Otc). The fluorescence intensity of C-CDs was selectively quenched by Fe3+, in sharp contrast to dozens of other metal ions like Al3+, Ba2+, Ca2+, Co2+, Mg2+, Pb2+, Na+, Hg2+, Ni2+, Zn2+, Cd2+ and Cu2+. Coincidentally, this fluorescence could also be selectively quenched by Otc, entirely distinct from related physiological molecules like glycine, ammonium hydroxide, glucose, urea. The prepared C-CDs could be used for label-free sensitive and selective fluorescent detector for Fe3+ in a wide concentration range of 0–500 μM with a detection limit of 0.21 μM and for Otc as well in the range of 0–100 µM with a detection limit of 0.05 µM. Moreover, the fluorescent C-CDs were successfully used for Fe3+ and Otc detection in some real samples, suggesting great application potential in fluorescence sensing of trace metal elements and antibiotics.

A deep neural network approach to voice conversion usually depends on a large amount of parallel training data from source and target speakers. In this paper, we propose a novel conversion pipeline, DeepConversion, that leverages a large amount of non-parallel, multi-speaker data, but requires only a small amount of parallel training data. It is believed that we can represent the shared characteristics of speakers by training a speaker independent general model on a large amount of publicly available, non-parallel, multi-speaker speech data. Such general model can then be used to learn the mapping between source and target speaker more effectively from a limited amount of parallel training data. We also propose a strategy to make full use of the parallel data in all models along the pipeline. In particular, the parallel data is used to adapt the general model towards the source-target speaker pair to achieve a coarse grained conversion, and to develop a compact Error Reduction Network (ERN) for a fine-grained conversion. The parallel data is also used to adapt the WaveNet vocoder towards the source-target pair. The experiments show that DeepConversion that only uses a limited amount of parallel training data, consistently outperforms the traditional approaches that use a large amount of parallel training data, in both objective and subjective evaluations.

To make full use of the key information in frame-level features, a DNN-based model for speech enhancement is proposed using self-attention on the feature dimension. Two improvement strategies are adopted to strengthen the attention of the fully connected layers to the effective information in the features. First, the model introduces the fusion of feature domains on the input features, using a 136-dimensional combination of features including the MFCC, AMS, RASTA-PLP, cochleagram, and PNCC. The fusion of features complements information from different domains, including the mel domain and gammatone domain, thus providing more effective information for self-attention. Second, a feature-level self-attention mechanism is applied to the output of the fully connected layer to obtain the information related to the task. The feature-level attention enables the fully connected layers to capture the internal correlations between different features and to reduce the redundancy brought by multiple features. The experimental results show that under the matched noisy condition, compared to the noisy signals, the proposed algorithm increased the PESQ, fwsegSNR and STOI by 40.92%, 60.2% and 8.31%, respectively, and by 23.64%, 32.55% and 3.4%, respectively, under the mismatched noisy condition. Comparisons between different neural networks indicate that the proposed algorithm is superior to the compared algorithms in both the matched and mismatched situations using fewer context frames. Therefore, the proposed model can effectively utilize the key information of the features to suppress the noise, thereby improving the speech quality and generalizing the mismatched samples.

The cross-regional dependency in the UK housing market is analysed using regional house price indices. In this article, a network approach based on partial correlations is proposed, along with rolling-window analysis to consider potential time-varying dependency. The results show that house prices in the outer South East region have the strongest influence on regional housing market interactions in the UK. This influence is stronger when the markets are highly interconnected, whereas the house prices in London have the strongest influence when the UK regional housing markets are relatively less connected.

In order to expand the family and improve the bioactivity of oral implant ceramics, the phase structures, mechanical and wetting properties of the hot-pressed yttria-zirconia/multilayer graphene oxide composite (3Y-ZrO2/GO) ceramics were investigated. GO was uniformly distributed in 3Y-ZrO2 powders, forming the C–O–Zr bond during the sintering process. In comparison to raw 3Y-ZrO2 ceramics, the flexural strength and fracture toughness improved up to 200% (1489.96 ± 35.71 MPa) in ZG3 (with 0.15 wt% GO) and 40.9% (8.95 ± 0.59 MPa m1/2) in ZG2 (with 0.1 wt% GO), respectively, while the relative density and Vickers hardness increased slightly. The toughening mechanisms included crack deflection, crack bridging, and GO put-out. Meanwhile, the composite ceramics were transformed into a more hydrophilic direction and indicated a good wetting property. In consideration of mechanical and wetting properties, the ZG3 would be a favorable alternative to the yttria-zirconia ceramic (Y-TZP) in dental implant applications. The results are expected to serve as a technical guidance for the fabrication and evaluation of dental implants.

SARS-CoV-2, a β-coronavirus, has rapidly spread across the world, highlighting its high transmissibility, but the underlying morphogenesis and pathogenesis remain poorly understood. Here, we characterize the replication dynamics, cell tropism and morphogenesis of SARS-CoV-2 in organotypic human airway epithelial (HAE) cultures. SARS-CoV-2 replicates efficiently and infects both ciliated and secretory cells in HAE cultures. In comparison, HCoV-NL63 replicates to lower titers and is only detected in ciliated cells. SARS-CoV-2 shows a similar morphogenetic process as other coronaviruses but causes plaque-like cytopathic effects in HAE cultures. Cell fusion, apoptosis, destruction of epithelium integrity, cilium shrinking and beaded changes are observed in the plaque regions. Taken together, our results provide important insights into SARS-CoV-2 cell tropism, replication and morphogenesis., Here, the authors characterize replication of SARS-CoV-2 in human airway epithelial (HAE) cultures and show that it can infect ciliated and secretory cells, affects transepithelial electrical resistance and causes plaque-like cytopathic effects associated with apoptosis.

Nb-doped SrSnO3 (SSNO) thin films were epitaxially grown on LaAlO3(001) single-crystal substrates using pulsed laser deposition under various oxygen pressures and substrate temperatures. The crystalline structure, electrical, and optical properties of the films were investigated in detail. X-ray diffraction results show that the cell volume of the films reduces gradually with increasing oxygen pressure while preserving the epitaxial characteristic. X-ray photoelectron spectroscopy analysis confirms the Nb5+ oxidation state in the SSNO films. Hall-effect measurements were performed and the film prepared at 0.2 Pa with the 780 °C substrate temperature exhibits the lowest room-temperature resistivity of 31.3 mΩcm and Hall mobility of 3.31 cm2/Vs with a carrier concentration at 6.03 × 1019/cm3. Temperature-dependent resistivity of this sample displays metal-semiconductor transition and is explained mainly by electron-electron effects. Optical transparency of the films is more than 70% in the wavelength range from 600 to 1800 nm. The band gaps increase from 4.35 to 4.90 eV for the indirect gap and 4.82 to 5.29 eV for the direct by lowering oxygen pressure from 20 to 1 × 10−3 Pa, which can be interpreted by Burstein-Moss effect and oxygen vacancies generated in the high vacuum.

Exemplar-based image denoising algorithms have shown great potential for image restoration with a multitude of existing models. In this paper, we interpret nonlocal similar patch-based denoising as a problem of low-rank recovery. This offers a physically plausible model and unifies several existing techniques in a single low-rank recovery framework. The framework can handle complex noise models, such as zero-mean Gaussian noise, impulse noise, and any other noise that can be approximated by mixing these two kinds of noise. Moreover, we introduce a new nonconvex surrogate for the $l_{0}$ -norm and find the optimal solution of the optimization problems when the new norm is applied to low-rank recovery. The experimental results with different kinds of noise confirm the effectiveness of the proposed low-rank recovery framework and the new norm.

In this article, a novel scheme to effectively mitigate the nonlinear impairments in a PAM-8 radio-over-fiber (ROF) delivery is proposed by a joint deep neuron network (J-DNN) equalizer, which has more superiority in terms of good training accuracy, satisfactory tracking speed, and over-fitting suppression compared with a typical deep neuron network (DNN) equalizer. Our proposed J-DNN equalization scheme is mainly based upon back-propagation (BP) algorithm and blind cascaded multi-modulus algorithm (CMMA), which can be trained via two steps including DNN initialization and DNN optimization. By using the proposed J-DNN equalizer, 60-Gbps PAM-8 signal generation and transmission over 10-km SMF and 3-m wireless link at 135-GHz can be achieved. For the digital signal processing (DSP) at receiver, comparisons between CMMA equalizer, DNN equalizer, and J-DNN equalizer are demonstrated. The results indicate that J-DNN equalizer has a much better BER performance in receiver sensitivity than the traditional CMMA, and an improvement of receiver sensitivity can be achieved as much as 1 dB compared with a DNN equalizer at the BER of 3.8 × 10−3. To the best of our knowledge, this is the first time to propose a novel joint DNN equalizer, which is promising for the development in integrated microwave photonics and microwave/millimeter-wave photonics for 5G applications and beyond.

An unexpected bistricyclic aromatic ene AF was synthesized in a tin(II) chloride-mediated reductive aromatization reaction. The obtained AF showed a highly overcrowded structural conformation as revealed by X-ray crystallography. Interestingly, AF exhibited reversible high-contrast mechanochromism and thermochromism between pale and red color. The obvious chromism is likely ascribed to the conformation transformation and trace amount of diradical species formation upon stimulus.