铝电解槽内电-磁-流场的耦合仿真方法及应用

基于电磁流体力学的基本理论,应用等效电阻网络法与有限元相结合计算铝电解槽电流场,标量电位法和双标量磁位法计算铝电解槽磁场,k-ε双方程模型计算电磁力驱动下的铝液流动,并与工业测试结果进行比较.对200 kA侧部四点进电的铝电解槽进行计算与测试,其结果吻合良好;水平磁场形成一个顺时针的漩涡,垂直磁场沿长轴和短轴呈反对分布;铝液流动呈现出沿长轴方向排列的4个漩涡.仿真磁场误差小于10.0%,流场误差不超过5.0%.     

大型预焙铝电解槽电、热场的有限元计算

采用加权余量的伽辽金法推导了铝电解槽电、热场计算的有限元方程.利用ANSYS有限元软件具有的多重单元、多重属性及其能耦合求解电、热场的特点,建立了铝电解槽阳极和熔体大面切片的有限元模型.在合适的边界条件的假定下,对160kA预焙槽的电、热场进行了仿真计算,分析了槽内的温度分布和电压、电流分布.结果表明:所建有限元模型的仿真结果与设计值吻合较好,证实了采用ANSYS软件优化铝电解槽设计和开发新型铝电解槽的可行性与准确性.     

铝电解槽内湍流流动与界面波动的数值模拟

建立了模拟铝电解槽内电解液和铝熔液湍流流及两层熔液界面波动的数学模型，利用交错网格上的一种基于ＳＩＭＰＬＥ算法的改形格式，计算了２８０ｋＡ电解槽内两层熔液的流动状况与界面形状。     

基于BP网络和专家系统的铝电解槽分层故障诊断

针对铝电解槽故障种类繁多和不易诊断的问题,设计了基于BP网络和专家系统的分层故障诊断系统,包括前层分类和后层预报;通过对槽电阻信号的频谱分析,提取了故障特征信息,并对故障进行分类;建立了基于BP网络的前层分类器,用于诊断特征显著的故障;制定了故障诊断和控制规则,完善了专家系统的知识库,根据前层分类结果对余下故障进行诊断;通过制定规则,将前层分类和后层预报相结合,实现了故障诊断系统的整体设计;仿真结果及理论分析表明,该系统可有效预报单一及复合故障,提高故障诊断的准确率,保证铝电解槽工作状况的稳定。     

泡沫铝柱壳对药柱水下爆炸压力场影响的数值研究

基于流体弹塑性模型,建立了描述泡沫铝在爆炸载荷下的冲击响应方程;采用Lagrange差分格式,在均匀网格上对方程进行了离散;编写数值计算程序,进行了药柱水下爆炸的一维数值计算,重点考虑了泡沫铝对水中爆炸冲击波分布的影响。结果表明,数值计算结果与实验数据、LS-DYNA模拟结果基本吻合,证明所建立的泡沫铝流体弹塑性本构方程可以用来描述泡沫铝的冲击响应行为,且泡沫铝对水中爆炸冲击波压力场的影响显著。     

轴对称场的电阻网格模拟

本文介绍利用电阻网格模拟轴对称场的原理和方法。并把模拟实验结果与数值计算结果对比，证明模拟实验的精确性比较好，克服了用电解槽法模拟所特有的误差。     

基于最大-最小蚂蚁系统优化ELM的电解槽故障诊断

针对铝电解槽故障特征种类繁多,难以快速准确的实现故障类型诊断,设计了一种基于最大-最小蚂蚁系统(MMAS)优化的极限学习机(ELM)故障诊断方法。介绍了电解槽常见的故障类型及其对槽电压的影响,对采集到的故障情况下的槽电压信号进行降噪处理,根据对降噪后故障信号的局域均值分解(LMD)结果得到故障特征。采用ELM算法辨识故障类型,针对ELM算法存在的参数问题,采用MMAS对ELM隐含层参数寻优。结果表明,MMAS优化的ELM既保证了较快的训练速度,同时获得了更高的故障测试正确率。     

磁驱动准等熵压缩铝物理量的数值分析

根据一维磁流体力学计算程序SSS/MHD的控制方程组,建立了磁驱动准等熵压缩台阶铝的计算模型,并对铝中的物理量变化过程进行了数值计算。在计算得到的电流和样品后界面粒子速度历程与实验测试结果相符合的基础上,分析了负载构型(样品)的动态电感变化、磁扩散规律,以及温度、密度、压力和速度等物理量的分布和变化规律。研究结果对深入认识磁驱动准等熵压缩物理过程和改进实验设计具有重要意义。     

铝基碱土金属二元合金形成热的经验理论计算

运用Miedema经验理论计算铝基碱土金属二元系统的整个成分范围内的固态和液态合金形成焓以及金属间化合物的形成焓 ,并将计算结果同已有的实验结果进行了比较 ,结果表明 ,计算值与实验符合较好 ,同时讨论了原子尺寸因素、电负性、电子轨道能量等对合金形成焓的影响。     

水溶液体系铝硅酸钠微结构的拉曼光谱和量子化学从头计算研究

选取了代表性水溶液体系铝硅酸钠的典型试样:铝酸钠溶液、硅酸钠溶液、铝硅酸钠溶液,分别测定了它们的拉曼光谱。构建了系列铝硅酸钠溶液微结构模型,应用量子化学从头计算方法优化了空间几何构型,并计算了其拉曼光谱,计算结果分别与相应试样的实验拉曼光谱进行对比。结果表明:在高浓度条件下,铝酸钠溶液中铝物种的主要存在形式是Al(OH)-4(Q0)和Al2O(OH)2-6(2Q1),低浓度时只存在Q0结构的铝酸根离子,以4Q0聚集态的四水合分子形式存在;在2.0M Na2SiO3溶液中,硅酸钠团簇结构可能的存在形式是四水合Na2H2SiO4;在铝硅酸钠溶液中,铝硅酸根离子的形态是由Si-O-Si和Al-O-Si键以不同铝硅原子配比相互连接形成的环状或链状结构。     

计算顺磁-铁磁相变临界磁场的一种方法

利用线性自旋波方法并结合Hellmann-Feynman定理引入一种计算量子磁性系统临界磁场的方法.计算两种特殊模型的临界磁场,并将理论结果与数值结果进行比较,理论值与数值结果高度吻合,验证了理论方法的准确性.     

Angular dependence of the coercivity and remanence of ordered arrays of Co nanowires

The angular dependence of the coercivity and remanence of ordered hexagonal arrays of Co nanowires prepared using anodic aluminum oxide templates was investigated. The experimental evolution of coercivity as a function of the angle, in which the external field is applied, is interpreted considering micromagnetic simulations. Depending on the angle between the axis of the wire and the applied magnetic field direction our results show that the magnetization reversal mode changes from vortex to a transverse domain wall. Besides, we observed that the dipolar interactions cause a reduction in coercive fields, mainly in the direction of easy magnetization of the nanowires. Good agreement between numerical and experimental data is obtained.     

Synthesis of magnetic gradients for NMR tomography

A numerical method is applied to calculate an optimal distribution of currents in air which generate the magnetic field gradients required to spatially encode the radiofrequency signal in a NMR tomographic experiment. We compare the performances of the gradient circuits for the whole body air-core electromagnet described by Bangert and Mansfield (J. Phys. E 15; 1982) with the results of our optimization.     

Effect of a uniform magnetic field on dielectric two-phase bubbly flows using the level set method

In this study, the behavior of a single bubble in a dielectric viscous fluid under a uniform magnetic field has been simulated numerically using the Level Set method in two-phase bubbly flow. The two-phase bubbly flow was considered to be laminar and homogeneous. Deformation of the bubble was considered to be due to buoyancy and magnetic forces induced from the external applied magnetic field. A computer code was developed to solve the problem using the flow field, the interface of two phases, and the magnetic field. The Finite Volume method was applied using the SIMPLE algorithm to discretize the governing equations. Using this algorithm enables us to calculate the pressure parameter, which has been eliminated by previous researchers because of the complexity of the two-phase flow. The finite difference method was used to solve the magnetic field equation. The results outlined in the present study agree well with the existing experimental data and numerical results. These results show that the magnetic field affects and controls the shape, size, velocity, and location of the bubble.     

Analytical sensitivity and reflected power through a D-shape optical fibre sensor

In this paper, the reflection properties and sensitivity of a D-shape optical fibre sensor are investigated theoretically and numerically with the emphasis on the metal layer [gold (Au), silver (Ag), copper (Cu), and aluminum (Al)]. Maxwell’s equations are used to determine the electric and magnetic fields of the incident waves at each layer. Snell’s law is applied and the boundary conditions are imposed at each layer interface to calculate the reflected power and sensitivity of the sensor. In the numerical results, the mentioned power is computed and illustrated as a function of wavelength, angle of incidence, metal layer thickness and refractive index of the external medium when the metal layer changes. The variation of sensitivity with the wavelength of the incident radiations is also proposed for some of the given metals.     

双周期加速结构腔间耦合系数的模拟计算

在直线加速器加速管的设计中, 腔间耦合系数是一个关键参量, 它的误差是影响场分布误差的主要因素. 准确计算耦合系数，对于腔间耦合孔几何参数的设计十分重要. 目前, 采用电磁场计算软件进行数值模拟是计算耦合系数的主要 方法. 文献调研发现, 使用MAFIA程序计算耦合系数, 对于单周期的盘荷波导结构, 根据MAFIA程序内部电(磁)边界条件和周期性边界条件, 对应两者有行波模拟、驻波模拟两种方法计算色散曲线. 本文结合这两种方法, 对计算双周期加速结构耦合系数的方法进行了探索, 它们可得到相近的结果, 与实验测量值的差别小于15%.     

A discussion on the magnetization calculation in polycrystalline antiferromagnetic system: Application to EuTiO3

In this work a numerical methodology is theoretically proposed to calculate the magnetization of a polycrystalline system, considering a microscopic Hamiltonian model, which describes a magnetic system consisting of two sublattices of different magnetic ions coupled by exchange interactions and with parallel and perpendicular direction of the magnetic field. The influence of changing the applied magnetic field direction on an antiferromagnetic system was systematically analyzed. The proposed numerical method was applied to the cubic perovskite EuTiO₃ and a good agreement with the experimental data was obtained.     

磁场引起的近藤单态退相干机制

磁场作为一个环境能够诱导近藤单态的退相干。我们采用格林函数方法,计算磁场下量子点耦合Aharonov-Bohm环系统的退相干特性,数值结果显示磁场引起的近藤单态的退相干是一个突然的过程。     

The deuteron and its magnetic moment the effects of non-local nuclear forces and of unitary transformations

The deuteron theory is renewed and completed for non-local Hamiltonians. A method to calculate deuteron wave functions is given. The formula for the current density and the magnetic dipole moment of a non-local potential is derived, and applied to the GCT, GT, and HJ-potential. The effects of a unitary transformation on the deuteron data — electric quadrupole moment, magnetic dipole moment, and radius — are shown. For the numerical calculations the Ristig-Kistler transformation and the HJ-potential were used. It is shown that an overall fit cannot be reached.