A Smart Deep Convolutional Neural Network for Real-Time Surface Inspection

A proper detection and classification of defects in steel sheets in real time have become a requirement for manufacturing these products, largely used in many industrial sectors. However, computers used in the production line of small to medium size companies, in general, lack performance to attend real-time inspection with high processing demands. In this paper, a smart deep convolutional neural network for using in real-time surface inspection of steel rolling sheets is proposed. The architecture is based on the state-of-the-art SqueezeNet approach, which was originally developed for usage with autonomous vehicles. The main features of the proposed model are: small size and low computational burden. The model is 10 to 20 times smaller when compared to other networks designed for the same task, and more than 700 times smaller than general networks. Also, the number of floating-point operations for a prediction is about 50 times lower than the ones used for similar tasks. Despite its small size, the proposed model achieved near-perfect accuracy on a public dataset of 1800 images of six types of steel rolling defects.     

铌对取向硅钢冷轧组织织构的影响

采用光学显微镜(OM)、电子背散射衍射(EBSD)等研究了铌对取向硅钢冷轧过程中的组织与织构的影响。结果表明:含铌正火板和不含铌正火板经冷轧后都是条带状组织,但含铌冷轧板在宽度方向更窄。铌元素的存在对织构也具有优化作用,经过中间退火后,含铌冷轧板二次冷轧织构得到明显改善,相较于不含铌二次冷轧板拥有强度较低的旋转立方织构以及强度较高的{111}<112>织构,其取向密度最高可达12.460。     

基于烟丝特性和卷制原理的卷烟机最佳回丝量计算模型

为确定卷烟机最佳回丝量,根据烟支卷制原理,结合烟支重量、规格、密度和平准盘凹槽规格、卷烟机吸丝道导轨规格等因素建立了卷烟机最佳回丝量计算模型。通过实例验证,该模型推算出的卷烟机回丝量是合适的,不仅能提供满足烟支质量要求的烟丝量,而且能最大程度减少因回丝量过大造成的烟丝造碎,降低烟支含末率,提升产品质量。该模型确定卷烟机最佳回丝量的方法实施方便,且不受卷烟机机型、烟支规格、原辅材料等因素限制,具有一般适用性。     

基于SAOR的Massive MIMO系统信号检测算法

大规模多输入多输出(Massive multiple input multiple output, Massive MIMO)系统采用最小均方误差(Minimum mean square error, MMSE)接收检测方法时存在矩阵求逆复杂度高的问题，已有较多降低复杂度的研究。在降低检测算法复杂度的同时，如何提高算法收敛速度和检测性能一直是人们关注的焦点。本文将对称加速超松弛(Symmetric accelerated over-relaxation， SAOR)迭代算法应用于Massive MIMO系统信号检测中，避免了复杂的矩阵求逆计算，实现了复杂度较最小均方误差算法降低了一个数量级。仿真结果表明，基于SAOR的检测方法通过较少的迭代次数就能逼近最小均方误差（Minimum mean square error, MMSE）算法的检测性能，为Massive MIMO系统中接收信号的快速检测提供了较好的实现方法。     

中压进汽腔的流场数值模拟比较

通过数值模拟计算,对中压对称进汽和切向进汽两种结构的流场进行了分析比较,结果表明,单一切向进汽腔的总压损失更小,出口汽流角的周向分布均匀度更好。更进一步,为整体评估中压进汽腔的流场以及对叶片级的流动影响,对中压进汽腔及第1级叶片的整体流体域流场情况进行了分析比较,结果表明,采用大几何角静叶的切向进汽腔气动性能最优;当采取切向进汽腔时,需合理选择第1级静叶几何角并耦合计算,才能实现进汽腔的气动优化。     

基于改进奇异值分解滤波和谱峭度的滚动轴承故障诊断

针对含噪信号的有效奇异值个数难以确定的问题，提出了一种改进的奇异值分解降噪方法--奇异值累积法。该方法通过计算奇异值的实际下降值与奇异值平均下降速度累积量的差值，并取该差值最大值点的位置作为有效奇异值的分界点来确定有效奇异值的个数。在此基础上，提出了一种基于奇异值累积法与快速谱峭度的滚动轴承故障诊断方法。采用奇异值累积法对原信号进行降噪处理，然后利用快速谱峭度确定滤波器中心频率及带宽，通过分析频段包络谱中明显的频率成分来诊断故障。该方法可以有效去除信号中的噪声，使得到的峭度值所反映的故障冲击更接近实际情况。对含内圈、外圈故障的滚动轴承实验数据进行分析，实验结果表明，相比快速谱峭度的故障诊断方法，该方法具有更好的故障识别效果。     

整体壁板离散填料滚弯成形有限元模拟

具有一定曲率外形的铝合金整体壁板属于典型的多筋零部件,传统弯曲成形工艺成形困难,容易产生多种缺陷。提出了离散填料辅助滚弯成形整体2A12T4铝合金壁板滚弯工艺,并通过实验与整体填料滚变成形进行了对比,同时采用有限元模拟技术,探究了离散填料辅助整体壁板滚弯成形中的影响因素。研究结果表明:填料数量、上下垫板厚度等因素对成形性能影响较大,填料在3~6块时成形稳定,可以有效地避免筋条屈曲与失稳;随垫板厚度增加,壁板表面圆弧更加光滑、连续,但是回弹性增强;上下垫板分别在10~20 mm厚度时,综合性能较好。通过离散填料辅助滚弯技术可以成形出质量较高的整体壁板。     

食品速冻保鲜过程冰晶生长的相场模拟

针对食品速冻保鲜工艺因大冰晶破坏细胞结构造成品质下降的问题,将细胞溶液视为水和溶质二元系统,采用五重对称强界面能各向异性相场模型,再现冰晶生长演化过程,探讨了冷冻时间对冰晶生长行为的影响。结果表明:冰晶生长模拟结果与实际形貌一致,两者主枝中部具有发达的侧向分支,主枝根部出现缩颈,主枝和侧枝尖端具有棱角。在主枝和侧枝之间扩散通道不畅通,形成半封闭液相区其溶质浓度达到1.76%,冰晶生长受到抑制。固液界面区域的无量纲温度在冰晶主枝根部变化平缓,而在尖端变化急剧。随着冷冻时间延长,无量纲温度和溶质浓度先快速增加后逐渐趋于稳定,尖端生长速率和曲率半径先急剧减小后逐渐趋于稳定,冰晶从初期近似五边形经过5主枝结构向五重对称多分支小平面结构转变。     

Hydrogen storage in MgH2LaNi5 composites prepared by cold rolling under inert atmosphere

Mg-AB₅ composites are promising systems for hydrogen storage applications, due to their possibility of hydrogen cycling at relatively low temperatures. Traditionally, these composites are mainly processed by high-energy ball milling (HEBM) techniques employing longer processing times. In this study, cold rolling was applied to prepare MgH₂LaNi₅ composites and the hydrogen storage properties were investigated. The materials were processed using a vertical rolling mill under argon atmosphere, leading to a good homogeneity and no contamination at shorter processing times. The mixture of MgH₂-1.50 mol.% LaNi₅ showed the best hydrogen storage properties at 200 °C and 100 °C and the lowest desorption temperature even when compared to cold rolled MgH₂. The results indicate that the composite MgH₂LaNi₅ is transformed into a mixture of three phases MgH₂, Mg₂NiH₄ and LaH₃ upon hydrogen absorption/desorption cycles. The synergetic effect among these phases when in appropriate proportion in the sample seems to play a crucial role in the acceleration of hydrogen absorption/desorption kinetics at lower temperatures in comparison to MgH₂.     

Casting methods for the production of rotationally symmetric copper bimetals

ABSTRACT

Multiple-material products are characterised by a complex property profile which is achieved by combining the particular advantages of at least two different materials. Bimetal casting is an energy- and material-efficient technology for the production of multi-metallic objects. This paper describes the development of a semi-continuous casting process for the formation of a rotationally symmetric bimetal with a cohesive bonding character at the interface of a copper–tin alloy (CuSn6) and pure copper (Cu99.5). Initial experiments are conducted by static casting to evaluate the thermal process window. Based on the results of the initial experiments, a vertical semi-continuous compound casting process is developed. A stable cohesive bond between the joining partners is accomplished by forming a solid solution at the interface.

This paper is part of a Thematic Issue on Copper and its Alloys.