SHINE装置的准直器设计

本文主要介绍了上海硬X射线自由电子激光装置的准直器设计。由于异常束损的功率很高,选用低原子序数的钛合金作为挡块材料,同时为了缩短挡块长度,采取了钛合金和铜沿束流方向上下游排布的双段式结构。FLUKA模拟表明电子在挡块内的电磁级联过程产生大量的次级粒子,对挡块的有限元分析表明异常束损会导致挡块内最高温度达到550℃左右,最高应力不超过材料的屈服强度。     

An improved U-Net for cell confluence estimation

Cell confluence is an important metric to determine the growth and the best harvest time of adherent cells. At present, the evaluation of cell confluence mainly relies on experienced labor, and thus it is not conducive to the automated cell culture. In this paper, we proposed an improved U-Net algorithm (called DU-Net) for the segmentation of adherent cells. First, the general convolution was replaced by the dilated convolution to expand the receptive fields for feature extraction. Then, the convolutional layers were combined with the batch normalization layers to reduce the dependence of the network on initialization. As a result, the segmentation accuracy and F1-score of the proposed DU-Net for adherent cells with low confluence (<50%) reached 96.94% and 93.87%, respectively, and for those with high confluence (≥50%), they reached 98.63% and 98.98%, respectively. Further, the paired t-test results showed that the proposed DU-Net was statistically superior to the traditional U-Net algorithm.     

On measuring network robustness for weighted networks

Knowledge and Information Systems - Network robustness measures how well network structure is strong and healthy when it is under attack, such as vertices joining and leaving. It has been widely...     

基于自注意力对抗的深度子空间聚类

子空间聚类(Subspace clustering)是一种当前较为流行的基于谱聚类的高维数据聚类框架.近年来,由于深度神经网络能够有效地挖掘出数据深层特征,其研究倍受各国学者的关注.深度子空间聚类旨在通过深度网络学习原始数据的低维特征表示,计算出数据集的相似度矩阵,然后利用谱聚类获得数据的最终聚类结果.然而,现实数据存在维度过高、数据结构复杂等问题,如何获得更鲁棒的数据表示,改善聚类性能,仍是一个挑战.因此,本文提出基于自注意力对抗的深度子空间聚类算法(SAADSC).利用自注意力对抗网络在自动编码器的特征学习中施加一个先验分布约束,引导所学习的特征表示更具有鲁棒性,从而提高聚类精度.通过在多个数据集上的实验,结果表明本文算法在精确率(ACC)、标准互信息(NMI)等指标上都优于目前最好的方法.     

基于电抗器开断过电压特征分析的匝间绝缘短路故障在线辨识

干式空心电抗器内部包封固化不易拆卸,当出现匝间绝缘短路故障时难以发现,停电检修维修成本较大周期较长,为了能利用运行中的信号在线检测出匝间绝缘故障,提出一种基于电抗器开断过电压特征分析的匝间绝缘故障辨识方法：在电抗器匝间绝缘短路故障分布式参数模型基础上,建立了带有匝间绝缘短路故障的电抗器开断过电压模型,分析匝间绝缘短路故障对电抗器开断过电压波形的影响作用,并利用变分模态分解算法对过电压波形进行频谱分析,最后搭建计及匝间绝缘故障开断过电压仿真模型,对存在不同程度的绝缘短路的电抗器进行开断模拟,获取电抗器开断电压波形及其振荡频率等数据,对波形数据进行分析和特征提取,量化开断过电压最大幅值及振荡频率与绝缘之间的差值,可以有效辨识出匝间短路故障。     

基于激光点云数据的三维运动图像重构技术北大核心CSCD

为了解决三维运动图重构时存在的重构测量距离与实际距离误差大、激光点云数据数量多和重构图像清晰度对比低的问题,提出了基于激光点云数据的三维运动图像重构技术,通过配准多帧激光点云数据,从中获取激光点云数据集,再采用平面拟合方法对激光点云数据集实行去噪处理,最后利用曲面重构法完成对曲面模型的拟合,实现三维运动图像重构。实验结果表明,通过对三维运动图像重构进行测量距离与实际距离的对比、激光点云数据数量的对比和测量图像与实际图像清晰度的对比测试,验证了三维运动图像重构技术的实用性高。     

基于哈特曼原理的两相流气相参数研究北大核心CSCD

本文提出哈特曼光线追踪的方法来测量气液两相流的气相参数,并进行了理论和实验研究。首先,采用哈特曼模板获得阵列光束,通过模拟追踪光线在气液两相流中的传播过程,研究出射光线与两相流中气泡尺寸、位置等参数之间的关系,其次,建立BP神经网络模型来实现气泡参数的准确反演,仿真结果表明,气泡定位相对误差在7%以内,气泡粒径的相对误差在±4%以内。在此基础上搭建了实验系统,以已知大小的透明颗粒在垂直管道沉降模拟单个气相通过管道的过程,进行了实验研究,结果表明气泡粒径的相对误差可控制在±6%以内。     

抛锚状态下的船载地波雷达目标自适应检测方法

船载地波雷达可利用船载平台机动灵活的特点,扩展地波雷达探测范围,但也会由于平台的运动,影响目标探测性能. 针对船载平台抛锚运动引起艏向变化和目标回波展宽范围不规则导致目标检测性能下降的问题,提出了一种适用于平台抛锚状态时的船载地波雷达目标自适应检测方法. 首先,从平台抛锚时船载地波雷达的目标回波特性和检测背景统计特性两个方面来分析目标检测中的难题. 然后,基于理论仿真与实测数据的分析结果,提出目标自适应检测方法. 该方法先通过前端数据的预处理消除艏向变化带来的影响,再将处理结果进行检测背景的自适应选取并作统计特性转化处理,经过恒虚警率检测后得到检测结果. 最后,利用实测数据对本文提出的方法进行验证. 结果表明:本文提出的船载地波雷达目标自适应检测方法能够有效解决船载平台抛锚时的目标检测难题,可显著提高抛锚状态下船载地波雷达的目标探测能力.     

Joint power control and user scheduling for backbone-assisted industrial wireless networks with successive interference cancellation

In a backbone-assisted industrial wireless network (BAIWN), the technology of successive interference cancellation (SIC) based non-orthogonal multiple access (NOMA) provides potential solutions for improving the delay performance. Previous work emphasizes minimizing the transmission delay by user scheduling without considering power control. However, power control is beneficial for SIC-based NOMA to exploit the power domain and manage co-channel interference to simultaneously serve multiple user nodes with the high spectral and time resource utilization characteristics. In this paper, we consider joint power control and user scheduling to study the scheduling time minimization problem (STMP) with given traffic demands in BAIWNs. Specifically, STMP is formulated as an integer programming problem, which is NP-hard. To tackle the NP-hard problem, we propose a conflict graph-based greedy algorithm, to obtain a sub-optimal solution with low complexity. As a good feature, the decisions of power control and user scheduling can be made by the proposed algorithm only according to the channel state information and traffic demands. The experimental results show that compared with the other methods, the proposed method effectively improves the delay performance regardless of the channel states or the network scales.