Replication of large area nanoimprint stamp with small critical dimension loss

In this paper,the replication process of large area nanoimprint stamp with small critical dimension(CD) loss was investigated,using the thin residual layer nanoimprint lithography(NIL) technology.The residual layer thickness was optimized by changing the spin-coated resist thickness.The dependences of the residual layer etching rate on gas flow,chamber pressure,and RF power were investigated,and the optimized process conditions were established.By means of the thin residual layer NIL technique and optimized residual layer etching process,large area stamp with small CD loss and multi-orientation patterns was successfully replicated on 2-inch SiO2/Si wafer.The CD loss was controlled within 5 nm.The replicated stamp showed high performance in the patterning with thermal NIL.The replication process reported in this work could also be used to fabricate large area nanostructures with small CD loss.

Effects of Dimensional Tolerances on the Friction Power Loss of Hydrodynamic Journal Bearing System

According to the dimensional tolerances on hydrodynamic journal bearing system, a nonlinear oil film force model was established,and the Reynolds' equation was solved by adopting finite difference method. In order to fulfill different dimensional tolerances in the system,adopting 2kfactor design and using the eccentricity ratio corresponding to the stability critical curve,the effects of the friction power loss brought by the dimensional tolerances of the dynamic viscosity,bearing width,bearing diameter and journal diameter were analyzed. The effect on dynamic characteristics of the hydrodynamic journal bearing system was quantitatively analyzed,and the nonlinear dynamic analysis, modeling and calculation methods were studied while considering the manufacturing tolerances. The results show that in contrast to the impacts of the tolerances in journal diameter,dynamic viscosity and bearing width,the bearing diameter tolerance would lead to the rise in the power loss, and the dimensional tolerances have different degrees of impacts on the journal bearing system. The friction power loss decreased as the eccentricity ratio increased, and when the eccentricity ratio was 0. 695 the power loss came to the minimum.The investigation would find the best solution and reduce energy consumption,then control varieties of nonlinear dynamical behavior effectively,and provide a theoretical basis for hydrodynamic journal bearing system in parameter design.

Study on the Effect of Magnetic Pole Clipping Angle on the Eddy Current Loss of Permanent Magnets in the Rotor of Magnetically Levitated Permanent Magnet Synchronous High-Speed Electric Spindles

Aiming at the problem of high temperature and even demagnetization failure of permanent magnet (PM) due to PM eddy current loss in PM synchronous high-speed motors, this paper proposes a technique to lessen PM eddy current loss by cutting the angle of PM poles to change the shape of PM structure. Firstly, an analysis is conducted on the mechanism of PM synchronous high-speed motor eddy current loss production, the theoretical analytical model of PM eddy current loss is deduced, and it is theoretically proved that the magnetic pole shaving angle can reduce PM eddy current loss. Then, a 25 KW surface-type PM synchronous high-speed motor as an object, using two-dimensional time-step finite element method (FEM) to model and analyze PM eddy current loss. The results show that the smaller the PM pole shaving angle, the less its eddy current loss will be, it is possible to minimize the pole shaving angle of eddy current loss by 9.8% compared to the unshaved angle. Finally, the temperature field of the PM is calculated using a finite element method, and the outcomes demonstrate that the maximum temperature of the PM with a magnetic pole shaving angle can be reduced by about 5% compared with the unshaved angle.

The anatomic determinants of conductive hearing loss secondary to tympanic membrane perforation

Objectives: Recent studies have introduced middle ear volume(MEV) as a novel determinant of perforation-induced conductive hearing loss(CHL) in a mechanism driven by trans-tympanic membrane pressure differences. The primary aims of this preliminary report are to: 1) correlate CHL with perforation size; 2) describe the relationship between CHL and MEV; and 3) compare CHL across a range of cholesteatoma involvement.Design: A retrospective pilot study was performed in 31 subjects with audiometry indicative of conductive hearing loss, temporal bone CT scans,and no prior middle ear surgery. Perforation size and MEV were analyzed with respect to CHL in a cohort of 10 perforated ears with no cholesteatoma. CHLs were compared in 3 groups defined by extent of cholesteatoma involvement.Results: Ears with large and small perforations showed mean ABG values of 32.0 ± 15.7 dB and 16.0 ± 16.4 dB, respectively. A direct relationship was observed between MEV and CHL for ears with large perforations across all frequencies, whereas this relationship for small perforations was frequency-dependent. Finally, a statistically significant increase in CHL was found across ears with increasing cholesteatoma involvement at 1000 Hz(X^2(2) = 9.786, p = 0.008),2000 Hz(x^2(2) = 8.455, p = 0.015),and 4000 Hz(x^2(2)= 8.253, p = 0.016).Conclusions: These pilot data suggest that greater perforation-induced conductive hearing losses may be associated with larger perforation sizes and cholesteatoma. The correlation between MEV and CHL may require additional study.

基于改进YOLOX的棉花姿态品级识别及其定位研究

【目的】旨在解决高质量采棉要求下,采棉机对不同姿态和品级的棉花进行精确识别与定位的问题,提出了1种基于改进YOLOX的棉花检测方法YOLOX-Cotton。【方法】YOLOX-Cotton使用YOLOX模型作为主体框架,包含识别模块和定位模块,并引入了CA(coordinate attention)模块和SIoU损失函数,以多种姿态、品级的棉花图片作为数据集,对其进行训练并测试。【结果】YOLOX-Cotton模型的识别模块能够识别不同姿态和品级的棉花,且模型精确率、召回率和平均精度均值达到92.9%、86.8%和92.4%,与原YOLOX模型相比分别提升了5.2、5.5和6.1百分点。该模型的定位模块能够准确获得棉花的位置,测量结果均在田间试验验证结果的阈值范围内,所有样本的标准偏差均小于0.01。【结论】YOLOX-Cotton能够有效解决采棉机在高质量采棉要求下对棉花的识别与定位问题,将为实现高质量采棉提供了有力的技术支撑。

周期阻尼结构在驾驶室降噪铝钢板隔声特性上的应用

针对驾驶室降噪用大块阻尼铺设的缺陷,提出了二周阻尼铝-钢板敷设方式,通过建立周期阻尼结构振动数学模型,仿真计算研究了周期阻尼结构模态分析,并分析了铝-钢板密度、阻尼块密度和形状三个主要参数对隔声量的影响,最后对比周期阻尼和大块阻尼隔声效果。结果表明:模态振幅减小的变化趋势随着频率的增大越来越明显,说明具有周期阻尼的更有利于降低铝-钢板结构的振动;隔声效果随着铝-钢板密度和阻尼块密度的增加而增大,在250~1000 Hz的中高频段,隔声量增加得更为明显。

Effects of Water Weight-Loss Walking Training on Lower Limb Motor Function and Gait in Stroke Patients

Background: Water weight-loss walking training is an emerging physical therapy technique, which provides new ideas for improving the motor function of stroke patients and improving the quality of life of patients. However, the rehabilitation effect of water weight-loss training in stroke patients is currently unclear. Objective: To analyze the effect of water weight loss walking training in stroke patients. Methods: A total of 180 stroke patients admitted to our hospital from January 2019 to December 2021 were selected and randomly divided into two groups. The control group received routine walking training, and the research group performed weight loss walking training in water on this basis. The lower limb motor function, muscle tone grade, daily living ability, gait and balance ability were compared between the two groups before and after treatment. Results: Compared with the control group, the FMA-LE score (Fugl-Meyer motor assessment of Lower Extremity), MBI score (Modified Barthel Index) and BBS score (berg balance scale) of the study group were higher after treatment, and the muscle tone was lower (P Conclusion: Water weight loss walking training can enhance patients’ muscle tension, correct patients’ abnormal gait, improve patients’ balance and walking ability, and contribute to patients’ motor function recovery and self-care ability improvement.

血-迷路屏障通透性与突聋发病机制及预后的关系:在3D real IR的发现

目的分析突聋患者的内耳钆造影MRI三维真实重建反转恢复(three dimensional real inversion recovery,3D real IR)成像上的表现,探讨血-迷路屏障的通透性与突聋发病机制及其预后的关系。方法对41例单侧突聋患者行内耳钆造影MRI,测量患耳和健耳的耳蜗信号强度,并测量延髓信号强度,分别计算出耳蜗/延髓比值(cochlear/medulla ratio,CM ratio),以CM比值作为血-迷路屏障通透性的标志物,分析突聋患者患耳、健耳CM比值的不对称程度与疗效之间的关系。结果41例患者中,33例(80.48%)患耳的CM比值高于健耳,差异有统计学意义(P<0.05);患耳CM比值为健耳的1.5倍以下者18例,治疗有效率为77.78%(14/18);患侧CM比值不高于健侧者8例,治疗有效率为100%;达到健耳的1.5倍至1.75倍之间者7例,治疗有效率为100%(7/7);达到健耳的1.75倍至2.0倍之间者2例,治疗有效率为50%(1/2);达到健耳的2.0倍以上者14例,治疗有效率为14.28%(12/14);差异有统计学意义(P<0.05)。结论内耳3D Real IR可显示突聋患者血-迷路屏障通透性的改变,80.48%的突聋患者患侧耳蜗出现高信号,患耳CM比值达健耳的1.75倍以上者多数预后不良。

南海大陆坡海底对海洋环境噪声的影响

通过位于大陆坡上和深海处的两站点相同深度噪声实验数据,分析了大陆坡海域不同海底环境对海洋环境噪声的影响,并利用三维射线模型计算了不同海底条件下的噪声谱级和垂直指向性,进而研究了海底环境对环境噪声场的影响规律,数值仿真的噪声谱级和实验结果在绝对幅值和随频率变化趋势方面均符合较好。实验和仿真结果表明,由于大陆坡上坡方向的浅海环境不利于航船噪声远距离传播,因此大陆坡上站点的低频(300 Hz以下)噪声谱级较深海处更低;由于大陆坡海底底质较硬,海面高频风生噪声除了直达接收器外还可经海底反射后到达,因此大陆坡上站点的高频(1 kHz以上)噪声谱级较深海处高且随频率下降缓慢,导致大陆坡上和深海处的环境噪声谱级曲线随着频率增加会存在交叉现象。

基于改进YOLOv5算法的车辆目标检测

近年来,车辆目标检测已经逐渐成为智能交通领域的研究热点,然而,现实道路场景的复杂性常常给目标检测带来了诸多挑战,尤其是频繁出现的目标遮挡、重叠以及小目标检测问题,该文提出一种基于改进YOLOv5算法的车辆目标检测模型。首先将CA(coordinate attention,坐标注意力)机制集成到检测网络中,以增强在高密度目标环境中对车辆的识别精度;接着,采用Focal-EIOU Loss作为替代损失函数,以实现更精确的定位精度和更快的收敛速度;最后,引入ODConv(omni-dimensional dynamic convolution,全维度动态卷积),通过结合多个卷积核以及执行多维特征关注,提升车辆特征提取的效果。在自制车辆数据集上的实验表明,与原始算法相比,改进算法的平均精度提高了1.5%,证实了其在车辆目标检测方面的有效性和优越性。

多层异构复合材料的制备及吸波性能

通过模板法与物理混合法相结合,成功制备了三维网络结构的BaTiO_(3)(BTO)+Fe_(3)O_(4)三聚氰胺泡沫异质结构复合材料,采用扫描电镜和X射线衍射对复合材料样品的表面形貌和晶体结构进行了表征,并使用矢量网络分析仪测试了该样品在2~18 GHz频率范围内的复介电常数和复磁导率,并根据测量数据计算了反射损耗值。随后,使用COMSOL多物理场仿真软件进行有限元分析,研究了该复合体系的吸收机制和吸波性能。结果表明:成功引入了BTO和Fe_(3)O_(4)形成的大量异质界面到三维网络结构碳中,构建了异质结构。BTO和Fe_(3)O_(4)优化了三维网络结构的阻抗匹配,从而显著提高了多层异构复合材料的有效吸收带宽。此外,BTO、Fe_(3)O_(4)和三维网络碳结构构筑成了高气孔率的三维网络,导致电磁波在材料内部发生多次反射和散射,增强了电磁波与材料之间的相互作用,实现了对电磁波的吸收,提升了吸波效率。该复合材料展示了出色的吸波性能,当复合材料的厚度为2.9 mm时,最强的反射损耗值达到−54.76 dB,而当复合材料的厚度为2.7 mm时,有效吸收带宽提升至7.92 GHz。

基于多维场景划分的台区线损率异常研判及关联用户精准追踪方法

针对台区线损率异常辨识困难,关联用户难以精准追踪的问题,提出一种基于多维场景划分的台区线损率异常研判及关联用户精准追踪方法。首先,提出了一种基于改进完备集合经验的模态分解(improved complete EEMD with adaptive noise,ICEEMDAN)算法的数据预处理策略,通过分析台区历史线损率曲线的变化趋势,构建台区线损率粗放场景集,并在此基础上采用支持向量聚类算法进一步对场景集进行二次精细划分,实现对台区多维场景集的建立。其次,提出一种基于簇类个案数目的区间动态平移策略,用于确定不同场景下线损率标准库的区间范围,并采用区间重叠率策略对划分中的孤立区间进行合并,实现对台区线损率标准库的完备划分。最后,给出一种基于关联分析的线损异常追踪方法,通过灰色关联分析法确定线损异常强相关因素,并基于改进Adtributor算法定量分析各因素和台区用户间的内在关联度,提高对台区线损率异常用户的追踪精度。算例采用台区真实数据进行仿真分析,结果表明了所提方法的有效性和实用性。

微咸水条件下微压过滤冲洗池的水头损失试验与预测模型

为探究进水流量、矿化度、滤网孔径对微灌用微压过滤冲洗池水头损失的影响,开展微压过滤冲洗池运行特性的物理模型试验,建立了微压过滤冲洗池在微咸水条件下的水头损失预测模型。结果表明,在方差分析中仅流量对微压过滤冲洗池的水头损失影响显著,确定的影响因素排序由大到小为进水流量、矿化度、滤网孔径;该模型的决定系数R^(2)为0.955,均方根误差为0.0523,平均相对误差为3.20%。建立的模型可准确预测微咸水条件下的水头损失,研究成果可为泵前过滤设备的实际应用提供技术支撑。

复杂海洋环境水下航行体低频声传播损失计算方法综述

本文对国内外水下航行体低频声学特征传播损失计算的传统方法和改进方法进行综述,并针对复杂海洋边界条件、三维海洋环境等问题的传播损失仿真计算的适用性进行分析,提炼并总结了复杂海洋环境下传播损失计算方法的研究现状。未来,水下航行体声传播损失研究将更加聚焦于低频声传播特性、声场时-空-频相干特性以及两极冰区环境下的声传播规律等问题,进一步提高传播损失计算方法在复杂海洋环境下的准确性和稳定性是重点研究方向。

分叉角与回转半径对双向流道灌水器水力性能影响模拟研究

为研究双向流道灌水器水力性能,将其结构单元作为研究对象,以分叉角和回转半径两个结构参数为影响因素,采用正交试验设计16种结构参数组合方案,利用计算流体动力学软件Ansysfluent对流道内部流体的流动状态进行数值模拟,通过量纲分析建立流道结构参数与流道单元局部水头损失的回归模型。结果表明:双向流道单元局部水头损失与分叉角、回转半径和流道流速有关,分叉角与局部水头损失呈负相关关系,回转半径和速度与局部水头损失呈正相关关系;当入口流速为固定值时,分叉角为32.4°、回转半径为1.72 mm,流道单元局部水头损失最大;当设置10个流道单元时,交叉排列方式组合的灌水器流态指数较优。

基于多维度注意力机制的行人重识别算法研究

针对有监督行人重识别算法泛化能力较差的问题,提出一种基于多维度注意力机制的行人重识别算法。该算法引入了一个多维度注意力模块,通过学习行人图像在多个维度上的特征信息,提高行人图像泛化能力。此外,还提出了一个优化损失模块,加速身份损失的收敛,从而提高模型的训练速度。通过将这2个模块结合起来,本文算法在公开数据集Market—1501和DukeMTMC⁃ReID上进行了大量实验。实验结果表明:所提出的算法能够有效地解决有监督行人重识别算法泛化能力较差的问题。

基于多维特征的电网海量日线损数据异常识别研究

鉴于现有线损数据异常识别方法无法判断异常线损原因,且查全率低等问题,提出了基于多维特征的电网海量日线损数据异常识别方法。融合各类电力运行数据,组建多维特征日线损数据异常溯源模型,使用皮尔逊相关系数计算不同线路间变压器与电压的关联性,明确线损异常原因;归一化处理日线损数据,引入时间离散度理念评估线损异常程度,利用神经网络学习获得负荷变化下线损数据异常计算模型,输入不同节点负荷值完成日线损数据异常识别操作。仿真实验结果表明,提出方法可以判断异常线损原因,并且异常识别查全率高达94%,提升了电网海量日线损数据异常识别准确性,解决了线损数据异常计算的偏差问题。

PAW-YOLOv7:河道微小漂浮物检测算法

河道漂浮物检测对于船舶自动驾驶以及河道清理有着重大意义,但现有的方法在针对河道漂浮物目标尺寸小且互相遮挡、特征信息少时出现检测精度低等问题。为解决这些问题,本文基于YOLOv7,提出了一种改进模型PAWYOLOv7。首先,为了提高网络模型对小目标的特征表达能力,构建了小目标物体检测层,并将自注意力和卷积混合模块(ACmix)集成应用于新构建的小目标检测层;其次,为了减少复杂背景的干扰,采用全维动态卷积(ODConv)代替颈部的卷积模块,使网络具有捕获全局上下文信息能力;最后,将PConv(partial convolution)模块融入主干网络,替换部分标准卷积,同时采用WIoU(Wise-IoU)损失函数取代CIoU,实现网络模型计算量的降低,提高网络检测速度,同时增加对低质量锚框的聚焦能力,加快模型收敛速度。实验结果表明,PAW-YOLOv7算法在本文利用数据扩展技术改进的FloW-Img数据集上的检测精度达到89.7%,较原YOLOv7提升了9.8%,且检测速度达到54帧/秒(FPS),在自建的稀疏漂浮物数据集上的检测精度比YOLOv7提高了3.7%,能快速准确地检测河道微小漂浮物,同时也具有较好的实时检测性能。

Dynamic-YOLOX:复杂背景下的苹果叶片病害检测模型

针对目前苹果叶片数据集的叶片病害种类不全以及图片背景单一等问题,构建了复杂背景下包括苹果叶部六种常见病害的苹果叶片病害数据集。针对目前主流苹果叶片病害检测模型检测精度不高、模型复杂和不满足实时监测等问题,提出了一种基于YOLOX-S(you only look once X-S)改进得到的复杂背景下的苹果叶片病害自适应检测模型Dynamic-YOLOX。设计并使用ECA-SPPFCSPC模块(efficient channel attention cross-stage partial fast spatial pyramid pooling module)更换YOLOX-S模型主干网络尾部Dark5中的空间金字塔池化(SPP)以及跨阶段局部网络(CSPNet)模块来增强模型关注深层语义特征、抑制无用信息的能力,并减少硬件内存开销。设计了动态跨阶段局部网络(ODCSP)模块,并用其更换YOLOX-S模型主干网络中Dark2、Dark3、Dark4部分以及颈部网络中所有的CSPNet模块,使得模型在面对不同输入特征时有更强的自适应性,在减少模型的参数量和计算量的同时提高了模型的平均检测精度。引入Varifocal Loss更换模型中分类置信度损失的BCEWithLogits Loss来提升模型对苹果叶片中密集小目标病害的检测精度。在自制数据集上Dynamic-YOLOX相对原始YOLOX-S模型的mAP提升了4.54个百分点,达到84.63%,同时模型的参数量和计算量分别下降了11.97%和13.45%,检测速度达到44.07 FPS。对比主流苹果叶片病害检测模型,Dynamic-YOLOX具有一定优越性。

基于CFD方法的离心风扇叶片优化研究

通过优化叶片入口角度、叶片弧半径、叶片数量以及离心风扇厚度,分析了内外径一定条件下的最优离心风扇,以达到降低电机损耗提高效率的目的。通过大量CFD仿真研究得到结果为:叶片入口角为20°、叶片圆弧半径600 mm、叶片数量20片时,风扇风压大50 Pa左右,风扇损耗降低4 kW左右;离心风扇厚度减半后,虽然降低了风摩损耗,但离心风扇的流量有效阈值范围变小,不建议改变离心风扇厚度。