Decomposition of the equation of state of asymmetric nuclear matter into different spin-isospin channels

We investigate the equation of state of asymmetric nuclear matter and its isospin dependence in various spin-isospin ST channels within the framework of the Brueckner-Hartree-Fock approach extended to include a microscopic three-body force（TBF） . It is shown that the potential energy per nucleon in the isospinsinglet T = 0 channel is mainly determined by the contribution from the tensor SD coupled channel. At high densities,the TBF effect on the isospin-triplet T = 1 channel contribution turns out to be much larger than that on the T =0 channel contribution. At low densities around and below the normal nuclear matter density,the isospin dependence is found to come essentially from the isospin-singlet SD channel and the isospin-triplet T = 1 component is almost independent of isospin asymmetry. As the density increases,the T = 1 channel contribution becomes sensitive to the isospin asymmetry and at high enough densities its isospin dependence may even become more pronounced than that of the T = 0 contribution. The present results may provide some microscopic constraints for improving effective nucleon-nucleon interactions in a nuclear medium and for constructing new functionals of effective nucleon-nucleon interaction based on microscopic many-body theories.

Single particle potentials of asymmetric nuclear matter in different spin-isospin channels

We investigate the neutron and proton single particle （s.p.） potentials of asymmetric nuclear matter and their isospin dependence in various spin-isospin ST channels within the framework of the BruecknerHartree-Fock approach. It is shown that in symmetric nuclear matter, the s.p. potentials in both the isospinsinglet T = 0 channel and isospin-triplet T = 1 channel are essentially attractive, and the magnitudes in the two different channels are roughly the same. In neutron-rich nuclear matter, the isospin-splitting of the proton and neutron s.p. potentials turns out to be mainly determined by the isospin-singlet T = 0 channel contribution which becomes more attractive for the proton and more repulsive for the neutron at higher asymmetries.

Miniature quad-channel spin-exchange relaxation-free magnetometer for magnetoencephalography

A miniature quad-channel optically pumped atomic magnetometer(OPM) has been developed based on the spinexchange relaxation-free(SERF) mechanism. With a vapor cell of size 8 mm×8 mm×8 mm, we have incorporated four SERF magnetometer channels, which provides sufficient spatial resolution for magnetoencephalography(MEG). The four channels share the same laser beam for the best cancellation of common mode noise due to laser fluctuations. With gradient measurement, the sensitivities of the four sensors are better than 6 fT/Hz^(1/2), which is also good enough for MEG measurement. The vapor cell is heated to 160℃ by a novel nonmagnetic current-heating structure. Our sensor with high spatial resolution and compact size is particularly suitable for MEG systems.

Study on the Model of Thermal Channel Spinning Process for PET Polymer

A mathematical model to describe the TCS process inPET high-speed melt-spinning is established.The de-velopment of crystallinity,molecular orientation andmorphological features of high-speed-spun PET fiberin TCS process is investigated at take-up speeds rangingfrom 3600-4400 m/min and thermal channel tempera-tures ranging from 160-200℃ by simulation.The pro-files of tensile force,running velocity,temperature,bi-refringence,crystallinity and diameter in the TCS spin-line for various process conditions are obtained.Thepredicted results using this model are compared with themeasured crystallinty,diameter and birefringence.Theresults indicate that they are in fairly good agreement.The"necking point"in the TCS spinline can be predictedby this model,and its position changes with the corre-sponding process parameters such as take-up speed andthermal channel temperature.It is found that the neck-ing point moves towards the spinneret when the take-upspeed or the thermal channel temperature increases.

Field-assisted electron transport through a symmetric double-well structure with spin-orbit coupling and the Fano-resonance induced spin filtering

We have investigated theoretically the field-driven electron-transport through a double-quantum-well semiconductor-heterostructure with spin-orbit coupling. The numerical results demonstrate that the transmission spectra are divided into two sets due to the bound-state level-splitting and each set contains two asymmetric resonance peaks which may be selectively suppressed by changing the difference in phase between two driving fields. When the phase difference changes from 0 to π, the dip of asymmetric resonance shifts from one side of resonance peak to the other side and the asymmetric Fano resonance degenerates into the symmetric Breit-Wigner resonance at a critical value of phase difference. Within a given range of incident electron energy, the spin polarization of transmission current is completely governed by the phase difference which may be used to realize the tunable spin filtering.

基于全裕度SA的高读可靠性STT-MRAM两位量化器设计

受隧道磁阻比(Tunnel Magneto Resistance,TMR)下降、工艺偏差和热波动等因素影响,先进工艺节点(亚25 nm)下的自旋转移力矩磁随机存储器(Spin Transfer Torque Magnetoresistive Random Access Memory,STT-MRAM)的读取裕度降低,读写正确率下降。纠错码(Error Correction Codes,ECC)技术方案能有效提升STT-MRAM读取可靠性。但是,常见的一位硬判决量化器无法发挥多位量化软判决ECC编解码算法的优势,而基于模数转换器的软判决量化器在实现上又要以牺牲面积和功耗为代价。提出的面向STT-MRAM的两位量化器采用全裕度灵敏放大器(Sensitive Amplifier,SA)和电阻量化判决门限,显著降低了量化器的电路复杂度和读写错误率。结果表明,在TMR和低读取裕度条件下,基于提出的两位量化器和极化码(Polar)编码的ECC算法能实现优于基于一位量化器的里德-索洛蒙博斯-乔赫里-霍克文黑姆码(Bose,Chaudhuri&Hocquenghem,BCH)编解码方案的输出帧错误率(Frame Error Rate,FER)。此外,本文提出在STT-MRAM级联信道模型中考虑实际SA引入的读判据错误,用于准确评估读取裕度对读正确率造成的影响。结果表明,基于全裕度SA的两位电阻量化方案能够显著提升判决器的有效读取裕度,减小SA引入的读判据错误,在TMR≤90%时,实现比采用一位硬判决量化器的BCH码平均低47%的输出FER。

Electron resonance-transmission through a driven quantum well with spin-orbit coupling

We have studied the spin-dependent electron transmission through a quantum well driven by both dipole-type and homogeneous oscillating fields. The numerical evaluations show that Dresselhaus spin-orbit coupling induces the splitting of asymmetric Fano-type resonance peaks in the conductivity, in which the dipole modulation and the homogeneous modulation are equivalent. Therefore, we predict that the dipole-type oscillation, which is more practical in the experimental setup, can be used to realize the tunable spin filters by adjusting the field oscillation-frequency and the amplitude as well.

Field-assisted spin-polarized electron transport through a single quantum well with spin-orbit coupling

We have investigated theoretically the field-driven electron transport through a single-quantum-well semiconductor heterostructure with spin-orbit coupling. The splitting of the asymmetric Fano-type resonance peaks due to the Dresselhaus spin-orbit coupling is found to be highly sensitive to the direction of the incident electron. The splitting of the Fano-type resonance induces the spin-polarization dependent electron current. The location and the line shape of the Fano-type resonance can be controlled by adjusting the energy and the direction of the incident electron, the oscillation frequency, and the amplitude of the external field. These interesting features may be used to devise tunable spin filters and realize pure spin transmission currents.

基于Friele模型的转杯纺黏胶混色针织物测配色系统

以红、黄、蓝3色原液着色黏胶短纤维为原料,采用多通道转杯纺技术,以质量比为10%的梯度纺制66种混色纱线,制备针织物样品,利用分光光度计测试其颜色特征。对Friele模型的关键参数σ在[0,1]内进行赋值以预测样品色彩,计算得出σ=0.134时双组分混色样品平均色差最小;取该值构建基于Friele模型的多通道转杯纺黏胶织物配色系统,利用全光谱配色法和所建模型预测3组分混色样品的配方比例,所得预测比例与实际比例间的平均比例误差为10.21%,平均色差为0.4561。研究结果验证了所建Friele模型的准确性,并为完善多通道转杯纺黏胶混色纺织品智能配色系统提供了依据。

提高细纱车间温湿度均匀性的方法与实践

为了有效改善细纱车间温湿度的不均匀性和稳定纱线品质,通过分析细纱车间机器设备散热特点和空调送排风系统存在的问题,从空调系统设计和运行管理两方面入手,提出采用合理设计车尾吸风沟道结构、优化空调送风、回(排)风系统、增加气流分隔和区域局部加湿等方法,减小细纱车间区域温湿度差异。工程改造实践表明:通过在细纱车间工艺排风口加装45°导向弯头和细纱机车尾通道处增加高压微雾局部加湿装置等措施,使车间中部区域细纱机车尾箱热空气外溢现象大为改善,车尾和车头、车中区域温度差异降至1℃以内,相对湿度差异降至2个百分点以内,工程改造基本上达到了车间工艺生产温湿度均匀性要求,纱线品质及一致性得到保证。

基于深度学习的对抗样本防御方法研究

受对抗样本自身可迁移属性的影响,传统对抗样本防御方法的防御效果存在不稳定的情况,为此,提出基于深度学习的对抗样本防御方法。文章借助深度学习算法,构建了对抗样本伊辛模型,设置模型的初始状态为神经网络的输入数据,采用自旋状态表示每一个神经元值与对抗样本伊辛模型的格点,并利用神经网络中卷积运算的特征,消解势场中预先给定的外部磁化作用,以最大限减少低对抗样本伊辛模型在能量作用下的局部自旋问题。在对抗样本防御阶段,利用对抗样本伊辛模型的通道相关性,生成重要性掩码对通道的激活进行调整,并结合对抗样本伊辛模型通道梯度累积值的实际情况设置了差异化的重要性掩码生成函数。在应用测试过程中,为验证防御效果,在快速梯度下降法(Fast Gradient Sign Method,FGSM)、Deepfool、C&W(Carlini and Wagner)攻击算法、投影梯度下降(Projected Gradient Descent,PFD)、集成对抗检测器(Energy-Aware Data-centric,EAD)共5种对抗策略下设计了对抗样本防御方法,对比不同对抗样本防御方法的性能,发现文章提出的基于深度学习的对抗样本防御方法的曲线下的面积(Area Under the Curve,AUC)值稳定在0.95以上,说明对抗样本防御方法具有较好的防御性能。

段彩纱双通道集聚纺吸风槽形状的优选

为了提高两组分集聚纺段彩纱成纱质量,对双通道集聚纺吸风槽的形状、结构参数及负压设置进行了研究。设计了Y形、V形、倾斜Y形及不对称Y形4种双通道集聚吸风槽,采用有限元气流场模拟并分析流场特征,对初步选定的Y形槽与V形槽的双槽间距、集聚点单槽高度、槽口宽度进行结构参数优化;通过纺制不同混纺比两组分段彩纱,对槽形及集聚负压进一步进行优选和试验验证。结果表明:Y形槽和V形槽在流场压力和速度分布上有明显优势,有助于纤维须条的集聚;双槽间距为16 mm、槽口宽度为1.3 mm、总高度为19.2 mm的V形槽流场分布更有利,所纺纱线强伸性能、毛羽指标改善,综合性能优于其他槽形,与数值模拟结果一致;集聚负压过小和过大都不利于纤维须条的汇聚,负压为2700 Pa时,可有效集聚纤维须条,且能耗水平较低。

热管纺丝工艺的模型研究

利用ＰＥＴ聚合物的本构方程，与热管和侧向冷却风有关的热传递机理以及纺丝运动学和非等温结晶动力学等建立了热管纺丝过程的数学模型。运用该数学模型进行了模拟计算，并得到了ＰＥＴ纤维热管纺丝线上的纺丝线张力、运行速度、温度、双折射、结晶度和直径沿纺程的梯度分布。于不同的工艺条件下在德国Ｂａｒｍａｇ公司生产的热管纺丝设备上纺制了ＴＣＳ－ＰＥＴ纤维，并分别测定了其卷绕丝的直径、双折射和结晶度。模拟计算得到的卷绕丝的双折射、结晶度和直径与实测结果吻合良好。

旋转导弹的控制信号分析

首先阐述了旋转导弹的单通道控制机理 ,指出在实际应用中由于初始条件的变化并不总是能够得到理想的等幅不等宽的脉冲调宽信号 ,因而传统理论分析得出的线性化信号频率与弹旋频率比并不总是最优的。从相位误差和幅值线性度这两个方面考虑 ,根据仿真计算结果 ,得出最优的频率比。

变径热管纺丝机理的研究

变径热管纺丝是一种新开发的纺丝工艺。通过建立纺丝动力学模型,对不同的工艺条件进行模拟,计算出纤维沿纺程上的温度、速度及应力分布,找出纺丝工艺参数的调节规律。

聚醚醚酮熔纺工艺的研究

选用进口聚醚醚酮树脂,系统研究了热甬道温度、喷丝头拉伸比对所得初生丝结构、最大牵伸倍数及所得纤维物理机械性能的影响.结果表明:热甬道可降低初生丝的取向度,从而提高可牵伸倍数,使纤维的物理机械性能更优;随着喷丝头拉伸比的增大,初生丝取向度增大,最大牵伸倍数降低,所得纤维的机械性能下降。

输棉通道几何参数对转杯纺气流场影响的数值研究

输棉通道的几何形状以及输棉通道与转杯的空间位置关系对通道和转杯内部气流流动有很大的影响,进而影响纤维在输棉通道内的伸直状态。本文通过数值计算的方法研究了输棉通道和转杯内部气流流动特征,并讨论了输棉通道特征数和空间位置角对通道和转杯内部气流流动特性及纤维伸直状态的影响。模拟结果表明,梳棉通道负压从出口到入口呈现先平稳增加,后急剧增加的趋势,特征数越大,该趋势越明显;输棉通道内部速度差随着特征数增大而增大;空间位置角α增加,输棉通道轴向负压先减小后增大,而输棉通道轴向速度差增大;空间位置角β对输棉通道静压和速度的分布影响较小,它主要改变纤维与转杯的初始接触位置。

TCS法生产异形细旦阳离子染料可染涤纶

用 TCS法生产异形细旦阳离子涤纶 FDY,探讨切片含水率、纺丝温度、冷却成形、上油率、热管温度等工艺参数对产品质量的影响。结果表明 ,采用合适的干燥工艺可获得含水率小于 2 0μg/g、粘度降小于 0 .0 0 5 d L/g的干切片 ,纺丝系统采用两套联苯加热保温 ,纺丝温度 2 99～ 30 1℃ ,降低侧吹风速度 ,提高侧吹风温度 ,上油率控制在 1%～ 1.4% ,热管温度为 172～ 174℃ ,卷绕速度为 4.5～ 4.6km /m in。

三通道数码细纱机纺制竹节纱的原理及其性能

为探究竹节倍率、竹节长度及竹节间距对多彩竹节纱性能的影响规律,提出一种基于三通道数码环锭细纱机利用单通道、双通道和三通道3种喂入方式纺制竹节纱的纺纱方法,分别采用4种竹节倍率(1.5、2.0、2.5、3.0)、3种竹节长度(5.1、7.2、9.3 cm)、相同竹节间距,纺制基纱线密度为27.8 tex的纯棉多彩竹节纱,并对纱线的外观参数和强伸性进行测试与分析。结果表明:竹节纱的竹节实际长度大于理论值,竹节实际间距和竹节实际倍率小于理论值;竹节倍率不变,随着竹节长度的增加,竹节实际长度与理论长度的偏差量不变,且竹节纱的强力逐渐减小;竹节长度不变,随着竹节倍率的增大,竹节长度接近理论值,且竹节纱的强力逐渐减小。

基于图像处理的家蚕吐丝管轮廓检测方法研究

作为自然微系统,家蚕在吐丝过程中表现出良好的微流动特性,其吐丝管结构与吐丝机理分析是研制仿生微通道和微流体系统的关键。以五龄家蚕为实验对象,通过石蜡切片和显微观察技术获得家蚕吐丝管断面切片及其图像;在Matlab和UG等软件中,运用图像处理技术对家蚕吐丝管切片图像进行滤波、分割、边缘提取、轮廓拟合与修复等操作,提取五龄家蚕吐丝管断面轮廓的显微图像;利用Image-Pro Plus6.0软件进行家蚕吐丝管轮廓尺寸测量。结果表明,图像处理技术可用于家蚕吐丝管轮廓检测,能够满足精度要求,为家蚕吐丝管的三维重构和内部微流动特性检测等后续工作奠定基础。