地点:福大旗山校区数学与计算机科学学院4号楼229室
时间:2016年11月25日
(1)时间:9:00-9:35;
报告人: 高卫国(复旦大学数学科学学院);
报告题目:Eigenvalue Problems in Computational Physical Sciencesand Data Science;
报告摘要:In this talk, we will introduce a few linear and nonlinear eigenvalue problems arising from physical sciences. Numerical methods for solving them and some existing issues will also be discussed.
(2)时间:9:35-10:10;
报告人:丁宁(北京应用物理与计算数学研究所);
报告题目:Z箍缩聚变研究数值模拟将面临的挑战;
报告摘要:快Z箍缩过程可以产生实验室中最强的X光辐射源,具有驱动惯性约束聚变的重要应用前景。丝阵Z箍缩过程可以分成三个主要阶段:初始的金属丝等离子体的形成和丝阵消融阶段、随后的内爆动力学和磁瑞利-泰勒(MRT)不稳定发展的阶段、最后在中心轴附近的停滞和辐射阶段。数值模拟是研究丝阵Z箍缩内爆的必要手段,对不同的物理过程需要研制使用不同的程序。针对丝消融和内爆过程,我们改进了Zuse-2D程序,进行二维(r,z)磁流体力学(MHD)模拟,获得了磁场和先驱电流的二维时空分布,分析了早期丝阵等离子体融合成壳的条件。针对内爆动力学过程,研制了二维(r, z)三温辐射MHD程序-MARED,研究了MRT不稳定性的时间演化。针对等离子体加速到轴滞止产生强X射线辐射的阶段,我们改进了一维非平衡辐射MHD程序-CRMHA,研究了丝阵内爆产生的辐射能谱。此外,利用全电路模拟研究了负载和驱动器的能量耦合过程,分析了驱动器参数对电磁脉冲输出的影响,采用合适的电路、电磁场以及MHD模型,描述电磁脉冲逐级压缩和功率放大过程。尽管,经过十几年的努力,我们已获得了上述Z箍缩物理过程的认识,具备了一定的模拟能力,并得到了Z箍缩实验的部分验证。但面向聚变研究,目前,我们关注Z箍缩驱动惯性约束聚变相关物理问题的数值模拟,如,准球形电磁内爆、动态黑腔和靶丸内爆等,Z箍缩聚变研究中数值模拟将面临诸多新的挑战。
(3)时间:10:30-11:05;
报告人:倪国喜(北京应用物理与计算数学研究所);
报告题目:多相反应流的数值模拟方法;
报告摘要:此次报告介绍本团队关于多相反应流的数值模拟方法所取得的结果。对于非理想流体,主要考虑了两种情况,刚性状态方程和JWL(Jones-Wilkins-Lee)型的状态方程.,针对这些非理想反应流体,本文提出了一种单元物理量重构法.它由单元格内的已知量出发,应用单元格内不同物质的物理量之间的关系,混合密度与各自密度之间的关系、以及混合内能与各物质内能之间的关系,建立关于多个变量的方程组, 再由温度平衡与压强平衡条件得到相应的方程,通过“移动跟踪法”求解该方程,重构出单元格内不同物质的物理量,这样就可以求解各个单元边界数值通量,从而得到一种高效的数值方法。
数值结果表明,这种算法既能清晰地捕捉一维和二维的爆轰波的结构和细节特征,也能比较准确地分辨多波相互作用的情形,胞格边界清楚、排列有序,三波点的特征明显.这些都验证了该算法的有效性和可靠性。
(4)时间:11:05-11:40;
报告人:赖惠林(福建师范大学数学与计算机科学学院);
报告题目:可压流体Rayleigh-Taylor不稳定性的离散Boltzmann建模与模拟;
报告摘要:The effects of compressibility on Rayleigh-Taylor instability (RTI) are investigated by inspecting the interplay between thermodynamic and hydrodynamic nonequilibrium phenomena (TNE, HNE, respectively) via a discrete Boltzmann model. Two effective approaches are presented, one tracking the evolution of the local TNE effects and the other focusing on the evolution of the mean temperature of the fluid, to track the complex interfaces separating the bubble and the spike regions of the flow. It is found that both the compressibility effects and the global TNE intensity show opposite trends in the initial and the later stages of the RTI. Compressibility delays the initial stage of RTI and accelerates the later stage. Meanwhile, the TNE characteristics are generally enhanced by the compressibility, especially in the later stage. The global or mean thermodynamic nonequilibrium indicators provide physical criteria to discriminate between the two stages of the RTI.
(5)时间:13:40-14:15;
报告人:尹丽(北京应用物理与计算数学研究所);
报告题目:Posteriori Element Residual Error Estimations for the CellFunctional Minimization Scheme;
报告摘要:In this talk, I presents a posteriori element residual error estimator for the cell functional minimization discrete scheme, which is a kind of hybrid.
(6)时间:14:15-14:50;
报告人:姚昌辉(深圳大学数学与统计学院);
报告题目:A Second Order and Three order BDF Numerical Scheme for Nonlinear Maxwell's Equations;
报告摘要:In this talk, we introduce a second order numerical scheme for Maxwell's equa-
tions with nonlinear conductivity, using the N\\'edelec Finite Element Method (FEM). A purely explicit treatment of the nonlinear term greatly simplifies the computational effort, since we only need to solve a constant-coefficient linear system at each time step. The curl-conforming nature of the N\\'edelec element assures its divergence-free property. In turn, we present the linearized stability analysis for the numerical error function to obtain an optimal $L^2$ error estimate. In more details, an $O (\au^2 + h^s)$ error estimate in the $L^2$ norm yields the maximum norm bound of the numerical solution, so that the convergence analysis could be carried out at the next time step. A few numerical examples in the transverse electric case (TE) in two dimensional spaces are also presented, which demonstrate the efficiency and accuracy of the proposed numerical scheme.
Similarly, we also present three BDF scheme for this problem, where an asymmetric scheme is proposed and the error estimates is analyzed with the convergent order O(\au^3+h^s). An important telescope formula is employed to prove the convergence. We also give the error estimated for the initial data.
(7)时间:14:50-15:25;
报告人:李娴娟(福州大学数学与计算机科学学院);
报告题目:LBBConstantofTriangularSpectralMethodforStokesEqs ontheInf-SupConstantofaTriangularSpectralmethodfortheStokesequations;
报告摘要:The Ladyzenskaja-Babuska-Brezzi (LBB) condition is a necessary condition for the well-posedness of discrete saddle point problems stemming from discretizing the Stokes equations. In this paper, we prove LBB condition and provide the (optimal) lower bound for this condition for the triangular spectral method proposed by L. Chen, J. Shen, and C. Xu. Then this lower bound is used to derive an error stimate for the pressure. Some numerical examples are provided to confirm the theoretical estimates.
(8)时间:15:45-16:20;
报告人:王靖岳(福州大学数学与计算机科学学院);
报告题目:On the approximation of discrete total variation operator by the N-neighbour anisotropic TV;
报告摘要:Functions with bounded total variation (TV) are widely seen in problems dealing with free discontinuities, such as image processing, mean curvature flows, front tracking, and others. We are concerned with the relationship between isotropic and anisotropic discrete total variation operators specifically used in the Rudin-Osher-Fatemi image denoising model. Isotropic TV operators yields boundary shapes that tend to be circles, while anisotropic TV prefers shapes that are compatible with the Wulff shape associated with its anisotropic function. It has been found that minimization problems utilizing a special type of anisotropic TV operator, N-neighbour anisotropic TV, can be computed with a very fast algorithm called graph-cut method. Unfortunately this algorithm can not be used on isotropic TV due to the loss of coarea formula for isotropic discrete TVs. It has also been observed that N-neighbour anisotropic TV minimization yields similar result to isotropic TV minimization as the number of neighbourpoints increases. We describe the general formula to construct N-neighbour anisotropic TV operators and prove the error between the anisotropic and isotropic TV minimizations. We also do some numerical experiments.
(9)时间:16:20-16:55;
报告人: 王美清(福州大学数学与计算机科学学院);
报告题目:偏微分方程在大数据处理中的应用-图像处理与计算金融;
报告摘要:实际生活中的很多问题可以表述为极值问题,然后通过Euler-Lagrange方法转化为偏微分方程。本文介绍偏微分方程在图像分割(图像处理)和期权定价(计算金融)中的应用,以及遇到的数值挑战。
"},"user":{"isNewRecord":true,"name":"系统管理员