基于1D CNN与XGBoost的恶意代码纹理检测

恶意代码数量已经呈现爆炸式增长,对于恶意代码的检测防护显得尤为重要.近几年,基于深度学习的恶意代码检测方法开始出现,基于此,提出一种新的检测方法,将恶意代码二进制文件转化为十进制数组,并利用一维卷积神经网络(1 Dimention Convolutional Neural Networks,1D CNN)对数组进行分类和识别.针对代码家族之间数量不平衡的现象,该算法选择在分类预测上表现良好的XGBoost,并对Vision Research Lab中的25个不同恶意软件家族的9458个恶意软件样本进行了实验.实验结果表明,所提的方法分类预测精度达到了97％.     

一种带有浮体下水防挤拉功能的水上光伏浮体阵列的研制

文章针对水上光伏电站浮体阵列挤拉组件受损的问题,研制出一种新型带有浮体下水防挤拉功能的水上光伏浮体阵列的连接装置,并提出三种方案进行定性定量分析,选取最佳方案为"限位盒"结构,并对最佳方案进行细化,确定基本结构.通过在现有工程中进行技术推广,并跟踪使用效果,结果表明:新装置提高了施工效率,降低了施工成本,效益显著.     

Application of ferrous sulfate alleviates negative impact of cadmium in rice (<i>Oryza sativa</i> L.)

Soil contamination with toxic heavy metals [such as cadmium (Cd)] is becoming a serious global problem due to rapid development of social economy. Iron (Fe), being an important element, has been found effective in enhancing plant tolerance against biotic and abiotic stresses. The present study investigated the extent to which different levels of Ferrous sulphate (FeSO₄) modulated the Cd tolerance of rice (Oryza sativa L.), when maintained in artificially Cd spiked regimes. A pot experiment was conducted under controlled conditions for 146 days, by using natural soil, mixed with different levels of CdCl₂ [0 (no Cd), 0.5 and 1 mg/kg] together with the exogenous application of FeSO₄ at [0 (no Fe), 1.5 and 3 mg/kg] levels to monitor different growth, gaseous exchange characteristics, oxidative stress, antioxidative responses, minerals accumulation, organic acid exudation patterns of O. sativa. Our results depicted that addition of Cd to the soil significantly (P < 0.05) decreased plant growth and biomass, gaseous exchange parameters, mineral uptake by the plants, sugars (soluble, reducing, and non-reducing sugar) and altered the ultrastructure of chloroplasts, plastoglobuli, mitochondria, and many other cellular organelles in Cd-stressed O. sativa compared to those plants which were grown without the addition of Cd in the soil. However, Cd toxicity boosted the production of reactive oxygen species (ROS) by increasing the contents of malondialdehyde (MDA), which is the indication of oxidative stress in O. sativa and was also manifested by hydrogen peroxide (H₂O₂) contents and electrolyte leakage to the membrane bounded organelles. Although, activities of various antioxidative enzymes like superoxidase dismutase (SOD), peroxidase (POD), catalase (CAT) and ascorbate peroxidase (APX) and non-enzymatic antioxidants like phenolics, flavonoid, ascorbic acid, anthocyanin and proline contents increased up to a Cd level of 0.5 mg/kg in the soil but were significantly diminished at the highest Cd level of 1 mg/kg in the soil compared to those plants which were grown without the addition of Cd in the soil. The negative impacts of Cd injury were reduced by the application of FeSO₄ which increased plant growth and biomass, improved photosynthetic apparatus, antioxidant enzymes, minerals uptake together with diminished exudation of organic acids as well as oxidative stress indicators in roots and shoots of O. sativa by decreasing Cd retention in different plant parts. These results shed light on the effectiveness of FeSO₄ in improving the growth and upregulation of antioxidant enzyme activities of O. sativa in response to Cd stress. However, further studies at field levels are required to explore the mechanisms of FeSO₄-mediated reduction of the toxicity of not only Cd, but possibly also other heavy metals in plants.     

紫色马铃薯酒发酵工艺优化及主发酵期花青素与色泽的动态变化

以紫色马铃薯为原料，在单因素的基础上，通过响应面法对紫色马铃薯酒的发酵工艺参数进行优化。最佳发酵工艺参数为:马铃薯与面粉比1:0.6、料水比1:1 (g/mL)、α-淀粉酶添加量8.5 U/g、酒曲添加量0.45%。此工艺在28 ℃下发酵14 d，得到的紫色马铃薯酒酒精度13.5%vol，花青素含量166.34 mg/mL，酒体澄清透明呈紫红色，酸甜适中，有独特酒香，综合感官评定得分90分。测定主发酵期花青素和色泽的动态变化，分析表明，紫色马铃薯酒在发酵过程中花青素呈现先上升后下降的趋势，色度呈现先下降后趋于平缓的趋势，色调则与之相反。聚合色度和褐变指数总体均呈现逐渐上升的趋势。试验结果为工业化生产紫色马铃薯酒提供理论依据。     

一种强声警告与驱离设备的设计与实现

介绍一种基于D类功放设计并实现强声警告与驱离设备的方法.该设计应用在水域及港口防护中,配合主动型小目标探测声呐,能有效地提升对水下入侵目标的防御和处置能力.     

煤立磨喂料锁风装置的改造

以5?000t/d水泥熟料生产线煤立磨三道锁风阀改造为例，针对MPF2217煤立磨喂料下料管锁风装置频繁堵料，故障率高，尤其冬季生产时三道锁风阀堵料严重的问题，通过调研学习，并总结多次改造治理的经验，实施了拆除三道锁风阀、增加导料槽和增加密封钢屋等措施，改造后解决了三道闸板堵料问题，降低了职工的劳动强度，改善了现场环境，提高了设备运转率。     

导管架平台上部组块双船拆除

为研究复杂海域环境中海洋平台双船拆除方法的安全性与可操作性,采用水动力数值计算与水池模型试验相结合的方法进行研究。将双船水动力耦合计算与水池模型试验进行对比,在相对理想海况下,结合与实际工况接近的规则波作用,进行驳船水动力性能变化与运动响应的幅值分析。在受力分析方面,对甲板耦合装置进行改良设计,对比试验数据与数值计算数据,得到受力在合理范围内的结论。研究内容为实际施工的合理性提出一定的指导性意见。     

上升气泡内气溶胶沉降机理研究

气溶胶池洗过滤是反应堆严重事故中去除放射性源项的重要手段。本文以严重事故条件下上升气泡中气溶胶的滞留过程为背景，设计搭建了可视化单气泡鼓泡实验装置。通过该装置研究了气溶胶在上升气泡中的沉降效率，并与MELCOR中的气溶胶沉降模型计算结果进行了对比。结果表明，气溶胶沉降效率对气泡尺寸的变化较为敏感，当气体流量大于0.1 L/min时，气泡等效直径迅速增加，相应的气溶胶沉降效率快速降低；与MELCOR模型计算结果的对比表明，两者在总体趋势上呈现出较好的一致性，但计算结果低估了液相对气溶胶的实际去除能力，导致这种偏差的主要原因是气泡在上升过程中存在无规则的晃动以及气液界面的波动。     

地面驱动螺杆泵井故障诊断方法

为提高地面驱动螺杆泵井的故障诊断效率,提出了一种将小波包与概率神经网络相结合的故障诊断方法.运用小波包技术提取有功功率信号的能量特征,与动液面、瞬时流量、井口回压共同作为特征参数,输入到概率神经网络进行故障诊断.为证明所提诊断方法的优越性,与BP神经网络、径向基神经网络进行对比,结果表明所提方法诊断效率更高、速度更快,更加满足故障诊断实时性的要求;故障诊断实例结果也进一步证明了该方法在地面驱动螺杆泵井故障诊断的可行性.     

一种基于深度学习的智能威胁评估算法

在现代信息化战争中依据战场态势变化进行合理威胁评估，对实现指挥系统自动化具有重要意义。为了克服传统规则算法评估能力的局限，设计了一种基于深度学习的新型智能威胁评估算法。首先设定了防空威胁评估指标选取原则，在此基础上从多个不同角度出发建立了反映空中目标威胁的指标体系，然后构建深度神经网络模型予以解决。新算法的优点在于威胁评估快速可信、易于实现，且对于不断变化的空情威胁可进行持续更新优化。最后，通过典型作战场景的仿真数据验证了模型的有效性和可行性。