基于推理注意力机制的二阶段网络图像修复

现有的图像修复方法在处理大面积缺失或高度纹理化的图像时，通常会产生扭曲的结构或与周围区域不一致的模糊纹理，无法重建合理的图像结构。为此，提出了一种基于推理注意力机制的二阶段网络图像修复方法。首先通过边缘生成网络生成合理的幻觉边缘信息，然后在图像补全网络完成图像的重建工作。为了进一步生成视觉效果更逼真的图像，提高图像修复的精确度，在图像补全网络采用推理注意力机制，有效控制了生成特征的不一致性，从而生成更有效的信息。所提方法在多个数据集上进行了实验验证，结果表明该图像修复方法的结构相似性指数达到了88.9%，峰值信噪比达到了25.56 dB，与现有的图像修复方法相比，该方法具有更高的图像修复精确度，生成的图像更逼真。     

高面板坝全寿命周期变形控制方法及应用

高面板坝的变形对面板的安全运行有着特别重要的影响，国内外已建的高面板坝工程中，因坝体变形大导致防渗面板挤压破损，坝体渗漏量大的实例较多，不得不降低水库水位进行修复处理，造成较大的经济损失乃至给大坝的长期运行留下安全隐患。通过发生挤压破损的实例分析，发现变形控制缺乏系统性是发生面板挤压破损的主要因素，为预防面板破损，系统提出了“控制坝体总变形，转化有害变形，适应纵向变形”的坝体变形控制方法，并在使用软硬岩混合料筑坝的董箐面板堆石坝中得到的应用，取得了良好效果，该工程运行至今达十余年，未见面板有挤压破损迹象，该方法对建设200 m以上乃至300 m级超高面板坝具有重要借鉴意义。     

Grid Search for Predicting Coronary Heart Disease by Tuning Hyper-Parameters

Diagnosing the cardiovascular disease is one of the biggest medical difficulties in recent years. Coronary cardiovascular (CHD) is a kind of heart and blood vascular disease. Predicting this sort of cardiac illness leads to more precise decisions for cardiac disorders. Implementing Grid Search Optimization (GSO) machine training models is therefore a useful way to forecast the sickness as soon as possible. The state-of-the-art work is the tuning of the hyperparameter together with the selection of the feature by utilizing the model search to minimize the false-negative rate. Three models with a cross-validation approach do the required task. Feature Selection based on the use of statistical and correlation matrices for multivariate analysis. For Random Search and Grid Search models, extensive comparison findings are produced utilizing retrieval, F1 score, and precision measurements. The models are evaluated using the metrics and kappa statistics that illustrate the three models’ comparability. The study effort focuses on optimizing function selection, tweaking hyperparameters to improve model accuracy and the prediction of heart disease by examining Framingham datasets using random forestry classification. Tuning the hyperparameter in the model of grid search thus decreases the erroneous rate achieves global optimization.     

Optical properties and radiation stability of SiO2/ZnO composite pigment prepared by co-sintering method

Thermal control coatings (TCCs) are an essential part of the thermal control systems in the spacecraft. Solar absorptance and emittance are the key performance parameters of TCCs. To develop an ultra-low solar absorption and stable inorganic TCCs for surface radiator, different TCCs were prepared by co-sintering ZnO and SiO₂ nanoparticles to form Zn₂SiO₄/SiO₂ pigment in this work, and the optical properties and radiation stability were systematically studied. It is found that the coating based on composite pigment has high reflectivity in the ultraviolet band and excellent optical performance possessing the low solar absorption of 0.06. In addition, the Zn₂SiO₄/SiO₂ coating demonstrates the highest proton and electron radiation stability because that SiO₂ between Zn₂SiO₄ particles acts as the relaxation center of the defects caused by radiation.     

低压静电场辅助冷冻对竹笋品质的影响

为研究低压静电场辅助冷冻对竹笋冻结特性的影响，以冻结曲线、硬度、水分损失率、水分迁移、冰晶形态和组织微观结构为指标，探究低压静电场辅助冷冻（－35 ℃）和普通冷冻（－35 ℃）条件下竹笋品质的变化规律。结果表明：低压静电场辅助冷冻提高了冻结效率，改变了冰晶形态及分布，减轻了组织微观结构破损程度，改善了解冻汁液流失情况。与静电板间距10、20、30、40 cm处的冷冻竹笋解冻后水分损失率分别为14.16%、12.58%、9.73%、10.44%，显著低于对照组（21.01%）（P＜0.05），硬度分别为461.19、507.48、496.65 g和455.31 g，显著高于对照组（350.70 g）（P＜0.05）。低场核磁共振分析结果表明，在低压静电场辅助冷冻下竹笋解冻后汁液流失减少，扫描电子显微镜观察结果显示，竹笋纤维排列整齐，组织微观结构保持较好。低压静电场辅助冷冻可有效改善竹笋品质，可为利用低压静电场进行果蔬的冷冻贮藏和冷链运输提供参考。     

Microbial coupled photocatalytic fuel cell with a double Z-scheme g-C3N4/ZnO/Bi4O5Br2 cathode for the degradation of different organic pollutants

A novel heterostructure of g-C₃N₄/ZnO/Bi₄O₅Br₂ (ZB-3) was designed, and used in the microbial coupled photocatalytic fuel cell (MPFC). It can effectively improve electron utilization efficiency and pollutant degradation using this double Z-scheme heterojunction structure. The current–time (I–t) curves demonstrated that the current density of ZB-3 was higher than that of ZnO, ZnO/Bi₄O₅Br₂ (ZB-1), g-C₃N₄/ZnO (ZB-2). Electrochemical impedance spectroscopy (EIS) indicated ZB-3 possessed the minimum charge-transfer resistance. This MPFC for degrading rhodamine B (RhB) and tetracycline (TC) under different conditions were developed using these materials. Even in the dark condition, MPFC with g-C₃N₄/ZnO/Bi₄O₅Br₂ demonstrated 93% and 82% degradation efficiency for RhB and TC, respectively. Furthermore, the electron transport mechanism of the MPFC and ZB-3 were proposed. It paves the approach for more efficient pollutant degradation via MFC photocatalysis.     

计算机网络办公自动化及安全策略探究

互联网通讯采取标准化模式主要以TCP/IP协议为载体,通讯的优越特性体现在同时具备便捷性与开放性,为办公提供很大的便利,但基于网络系统也会入侵病毒、也会给信息数据与办公体系安全性造成威胁,直接影响企业综合稳定发展。据此,为保障办公工作的顺利开展,本文对计算机网络办公自动化及安全策略进行详细分析。     

冰下单载波水声通信试验研究

北极地区战略地位的提高促进了极区各种应用的快速发展，冰下水声通信成为其中一个热点问题。为了研究冰下单载波水声通信性能，在黄海冰区开展冰下试验。重点分析了内嵌锁相环的多通道判决反馈均衡(Multi-channelDecision Feedback Equalization, M-DFE)技术在冰下水声通信中抗多途时延的能力，研究比较了改进比例归一化最小均方(Improved Proportional normalized Least Mean Square, IPNLMS)算法和递归最小二乘(Recursive Least Squares,RLS)算法的均衡性能。试验结果表明IPNLMS和RLS都能克服信道多途效应带来的影响。在牺牲计算复杂度的前提下，RLS具有更好的均衡效果，M-DFE在冰下环境中能够有效消除码间干扰。     

Phosphorylated chitosan/poly(vinyl alcohol) based proton exchange membranes modified with propylammonium nitrate ionic liquid and silica filler for fuel cell applications

In our previous work, phosphorylated chitosan was modified through polymer blending with poly(vinyl alcohol) (PVA) polymer to produce N-methylene phosphonic chitosan/poly(vinyl alcohol) (NMPC/PVA) composite membranes. The aim of this work is to further investigate the effects of a propylammonium nitrate (PAN) ionic liquid and/or silicon dioxide (SiO₂) filler on the morphology and physical properties of NMPC/PVA composite membranes. The temperature-dependent ionic conductivity of the composite membranes with various ionic liquid and filler compositions was studied by varying the loading of PAN ionic liquid and SiO₂-PAN filler in the range of 5–20 wt%. As the loading of PAN ionic liquid increased in the NMPC/PVA membrane matrix, the ionic conductivity value also increased with the highest value of 0.53 × 10^?3 S cm^?1 at 25 °C and increased to 1.54 × 10^?3 S cm^?1 at 100 °C with 20 wt% PAN. The NMPC/PVA-PAN (20 wt%) composite membrane also exhibited the highest water uptake and ion exchange capacity, with values of 60.5% and 0.60 mequiv g^?1, respectively. In addition, in the single-cell performance test, the NMPC/PVA-PAN (20 wt%) composite membrane displayed a maximum power density, which was increased by approximately 14% compared to the NMPC/PVA composite membrane with 5 wt% SiO₂-PAN. This work demonstrated that modified NMPC/PVA composite membranes with ionic liquid PAN and/or SiO₂ filler showed enhanced performance compared with unmodified NMPC/PVA composite membranes for proton exchange membrane fuel cells.     

Co2FeO4@rGO composite: Towards trifunctional water splitting in alkaline media

Development of efficient, low cost and multifunctional electrocatalysts for water splitting to harvest hydrogen fuels is a challenging task, but the combination of carbon materials with transition metal-based compounds is providing a unique and attractive strategy. Herein, composite systems based on cobalt ferrite oxide-reduced graphene oxide (Co₂FeO₄) @(rGO) using simultaneous hydrothermal and chemical reduction methods have been prepared. The proposed study eliminates one step associated with the conversion of GO into rGO as it uses direct GO during the synthesis of cobalt ferrite oxide, consequently rGO based hybrid system is achieved in-situ significantly, the optimized Co₂FeO₄@rGO composite has revealed an outstanding multifunctional applications related to both oxygen evolution reaction (OER) and hydrogen counterpart (HER). Various metal oxidation states and oxygen vacancies at the surface of Co₂FeO₄@rGO composites guided the multifunctional surface properties. The optimized Co₂FeO₄@rGO composite presents excellent multifunctional properties with onset potential of 0.60 V for ORR, an overpotential of 240 mV at a 20 mAcm^?2 for OER and 320 mV at a 10 mAcm^?2 for HER respectively. Results revealed that these multifunctional properties of the optimized Co₂FeO₄@ rGO composite are associated with high electrical conductivity, high density of active sites, crystal defects, oxygen vacancies, and favorable electronic structure arisinng from the substitution of Fe for Co atoms in binary spinel oxide phase. These surface features synergistically uplifted the electrocatalytic properties of Co₂FeO₄@rGO composites. The multifunctional properties of the Co₂FeO₄@ rGO composite could be of high interest for its use in a wide range of applications in sustainable and renewable energy fields.