Grain refinement caused by intensified cavitation within narrow channel and its improvement to ultrasonically soldered Al joint property

In this work, grain refinement of the aluminum soldered joint was obtained by applying cavitation within narrow channels and the possible grain refinement mechanism was proposed. Aluminum sheets with different channel widths were ultrasonically soldered by pure Sn in air. An ultrasonic system with a TC4 sonotrode, was operated at a frequency of 20 kHz and power of 1000 W during soldering. The effect of channel width on grain size, element distribution and strength of the soldered joint was studied. Results showed that the grain size decreased from 2.62 to 1.04 µm and the element ratio of Al in solder increased from 0.93 to 4.86% when the channel width decreased from 0.8 to 0.2 mm. Instant solidification of Sn grains was readily observed in the joint before cooling due to the large undercooling induced by the intensified cavitation inside the narrow channels. The random cavitation induced nucleation of Sn was believed to be mainly responsible for the grain refinement of the soldered joint. The shear strength of the joint increased from 29.5 to 48.8 MPa and the hardness increased from 16.5 to 25.2 HV due to the grain refinement of Sn and the presence of Al transferred from the substrates.     

基于动态阈值匹配追踪的主动声呐直达波抑制方法

双基地声呐中的直达波干扰可以通过重构抵消的方式进行抑制。这种方法需要对直达波信道进行估计，传统的正交匹配追踪算法的收敛速度慢，分步正交匹配追踪等算法在提高收敛速度时牺牲了信道估计精度，导致回波检测能力下降。本文提出一种动态阈值匹配追踪算法估计直达波的信道响应，在提高收敛速度的同时兼顾了信道估计精度。在仿真环境中，达到同样的直达波抑制效果，所提算法与传统正交匹配追踪算法相比，收敛速度显著提升，检测输出的回波强度比分步正交匹配追踪算法高4dB；海试数据处理结果中，所提方法迭代收敛速度较正交匹配追踪算法提升4倍；输出的回波强度比分步正交匹配追踪算法高2dB。     

混合板式换热器板通道的数值模拟研究

本文采用二维非稳态数学模型、周期性充分发展边界条件对混合板式换热器中的板通道内的流动和换热进行数值模拟。在Re=4407～22035工况范围内,计算了半圆型、半椭圆Ⅰ型和半椭圆Ⅱ型三种不同纵截面形状波纹板通道内的流动和换热性能。数值模拟结果分析表明,波纹通道内的流动会因为流体从凸面流向凹面时产生回流而形成涡,即产生扰动,强化了后面的换热,但同时增大了阻力。并得出半圆型通道换热最强,Nu为椭圆Ⅱ型的1.4倍;但其阻力最大,压降为椭圆Ⅱ型的1.9倍。     

Effect of pulse magnetic field stimulation on calcium channel current

This study aimed to investigate the effect of low frequency and high amplitude pulse magnetic field (PMF) on Calcium ion channel current of cells. Measurements were done on the Human Embryonic Kidney 293 cells (HEK 293), which have only Calcium ion channels functioning. The whole cell current was measured by patch clamp method, with the clamped voltage ramping from −90 mV to +50 mV across the cell membrane. A PMF was generated by a 400-turn coil connected to a pulse current generator. The frequency of the pulse was 7 Hz, the width of the pulse was 3 ms, and the amplitude of the pulse, or the flux density, was ranging from 6 to 25 mT. The results showed that the profile of the whole cell Calcium channel current could be modified by the PMF. With the PMF applied, the phase shifting occurred: the onset of the channel opening took place several mili-seconds earlier than that without the PWF and correspondingly, the whole cell current reached its maximum earlier, and the current returned back to zero earlier as well. When the PWF was stopped, these effects persisted for a period of time, and then the current profile “recovered” to its original appearance. The decrease of the onset time and peak current time could be due to the local electric potential induced by the PWF and the direct interaction between PMF and ion channels/ions. The exact mechanisms of the observed effects of PMF on the cell are still unknown and need to be further studied.     

结合Lab空间和单尺度Retinex的自适应图像去雾算法

针对暗通道先验算法在大片天空区域透射率估计过小及景深突变处出现Halo效应的问题，提出一种结合Lab空间和单尺度Retinex的图像去雾算法。将RGB图像转换至Lab空间提取出亮度分量，利用Canny算子对亮度分量提取边缘信息，丰富恢复图像细节；利用单尺度Retinex对非边缘区域进行高斯自适应滤波估计出优化后的亮度分量，获得“伪”去雾图像，得到粗略的透射率；利用交叉双边滤波优化透射率消除Halo效应；最后根据大气散射模型恢复出无雾图像。实验结果表明，该方法恢复出的图像细节明显，整体平滑，且对含大片天空区域的图像也有较好的恢复效果。     

Intermittency,Moments, and Friction Coefficient during the Subcritical Transition of Channel Flow

The intermittent distribution of localized turbulent structures is a key feature of the subcritical transitions in channel flows, which are studied in this paper with a wind channel and theoretical modeling. Entrance disturbances are introduced by small beads, and localized turbulent patches can be triggered at low Reynolds numbers (Re). High turbulence intensity represents strong ability of perturbation spread, and a maximum turbulence intensity is found for every test case as Re ≥ 950, where the turbulence fraction increases abruptly with Re. Skewness can reflect the velocity defects of localized turbulent patches and is revealed to become negative when Re is as low as about 660. It is shown that the third-order moments of the midplane streamwise velocities have minima, while the corresponding forth-order moments have maxima during the transition. These kinematic extremes and different variation scenarios of the friction coefficient during the transition are explained with an intermittent structure model, where the robust localized turbulent structure is simplified as a turbulence unit, a structure whose statistical properties are only weak functions of the Reynolds number.     

Determination of the best carrier frequency based on the system and environmental parameters in underwater acoustic systems

One of the most challenging features of underwater acoustic (UWA) channel in comparison with its terrestrial radio frequency counterpart is highly frequency-dependent path loss. Thus, utilizing efficient carrier frequencies in UWA systems can considerably reduce the path loss. In this context, this paper presents an approximate formula for determining the best carrier frequency based on both the system and environmental parameters. To achieve this goal, this research first addresses a simple algorithm including general steps for tuning the parameters of Francois and Garrison (FG) formula in the frequency range of 10 to 100 kHz based on the appropriate experimental data which can be acquired from any interest region. Second, for a more accurate modeling of path loss, this paper considers the loss due to the reflections of sound from both the rough surface and bottom of the sea by employing the ray theory. Third, this study presents a general algorithm for modification of the power spectral density (PSD) of ambient noise based on Wenz formula in the frequency range of 10 to 100 kHz and the required experimental measurements which can be simply collected from any interest channel. Moreover, it is mathematically demonstrated that the ambient noise in the frequency range of 10 to 100 kHz, can be generally approximated with a strict sense stationary (SSS) colored normal stochastic process which is ergodic not only in mean and covariance but also in distribution. Finally, an approximate formula for the best carrier frequency is derived by maximizing the sound to noise intensity ratio (SNR). To verify the validity of simplifications and approximations utilized in this study and to assess the performance of our proposed algorithms and formulas, experimental results obtained in the Strait of Hormuz (SoH) are compared with the original, simplified, and modified models under different scenarios.     

基于模糊标签和深度学习的TOF-MRA影像脑动脉瘤自动检测

脑动脉瘤破裂造成的蛛网膜下腔出血致死致残率极高，借助深度学习网络辅助医生实现高效筛查具有重要意义．为提高基于时间飞跃法磁共振血管造影（Time of Flight-Magnetic Resonance Angiography，TOF-MRA）的脑动脉瘤自动检测的精度，本文基于模糊标签方式，提出一种基于变体3D U-Net和双分支通道注意力（Dual-branch Channel Attention，DCA）的深度神经网络DCAU-Net，DCA模块可以自适应地调整通道特征的响应，提高特征提取能力．首先对260例病例的TOF-MRA影像预处理，将数据集分为174例训练集、43例验证集和43例测试集，然后使用处理后的数据训练和验证DCAU-Net，测试集实验结果表明DCAU-Net可以达到90.69%的敏感度，0.83个/例的假阳性计数和0.52的阳性预测值，有望为动脉瘤筛查提供参考．