基于熵条件二阶差分格式的嵌套网格分区算法

以带熵条件的二阶TVD格式为基本格式,设计了三维曲线坐标下的Euler方程数值计算方法.对多个物体使用分区嵌套网格设计了相应的分区算法.所建立的算法适用于处理物体相互分离以及相互接触的定常与非定常流动问题.给出了绕两圆柱的超声速流动在两圆柱处于不同位置时的详细结果,并模拟了装配鸭舵火箭弹在超声速旋转飞行时的气动特性.     

Numerical schemes with high order of accuracy for the computation of shock waves

ＩｎｔｒｏｄｕｃｔｉｏｎＴｈｅｃｏｍｐｕｔａｔｉｏｎｏｆｆｌｏｗｆｉｅｌｄｗｉｔｈｓｈｏｃｋｗａｖｅｓｈａｓｂｅｅｎｔｈｅｓｕｂｊｅｃｔｏｆｒｅｓｅａｒｃｈｆｏｒｍａｎｙｙｅａｒｓ.Ｔｈｅｒｅａｒｅｂａｓｉｃａｌｌｙｔｗｏｃａｔｅｇｏｒｉｅｓｏｆｍｅｔｈｏｄｓ,ｎａｍｅｌｙ,ｓｈｏｃｋｆｉｔｔｉｎｇｍｅｔｈｏｄａｎｄｓｈｏｃｋｃａｐｔｕｒｉｎｇｍｅｔｈｏｄ.Ｔｈｅｆｏｒｍｅｒｄｉｖｉｄｅｓｔｈｅｃｏｍｐｕｔａｔｉｏｎａｌｄｏｍａｉｎｉｎｔｏｓｕｂ＿ｄｏｍａｉｎｓｂｙｔｈｅｓｈｏｃｋｗａｖｅｓ.Ｉｎｅ…     

叶栅可压缩流场的数值分析

将一维ＡＵＳＭ＾＋格式与三阶精度ＭＵＳＣＬ格式融合,给出了其在任意风线坐标下的二维形式,采用有限体积法,将ＡＵＳＭ＾＋格式与Ｒｕｎｇｅ－Ｋｕｔｔａ格式结合,对叶栅可压缩流场进行了数值模拟,典型算例的计算结果与有关献结果相符很好,表明ＲＫ－ＡＵＳＭ＾＋混合格式对叶栅可压缩流场进行数值模拟是可行的。     

一种适用于板料成形过程分析的自适应有限元方法

本文主要讨论在板料成形过程的动态显式有限元分析中,有限元网格的丙分割技术。单元为四节点壳单元,板料的材料特性假定满足Ｈｉｌｌ各向异性的弹塑性准则,采用自适应ｈ－方案,通过细化和聚合单元调整网格的疏密,模具被假定为由刚性小平面组成,与同一单元的节点发生接触处的模具表面的法线之间的夹角作为单元再分的判据。通过对方表盒冲压成形过程的计算,说明该方法是有效的。     

Shift operator generated by trigonometric systems

We study the isometric operator generated by a trigonometric system. This operator is used in problems with directional derivatives and in theory of the Fourier series. We prove that this isometric operator is a pure shift and present application of this fact to the inverse problem of magnetoresistense. Translated fromMatematicheskie Zametki, Vol. 67, No. 4, pp. 539–548, April, 2000.     

Symmetric Powers of Galois Modules on Dedekind Schemes

We prove a certain Riemann–Roch-type formula for symmetric powers of Galois modules on Dedekind schemes which, in the number field or function field case, specializes to a formula of Burns and Chinburg for Cassou–Noguès–Taylor operations.     

LINEAR ISOMETRIC NON-ANTICIPATIVE TRANSFORMATIONS OF WIENER PROCESS

ThispaperisanoutcomeofnumerousdiscusionsandexchangeofideaswithlatePredragPerunicic,whosedeepandprofoundmathematicalknowledgeandhumanvaluesgaveusaninspirationandmotivationtofinalizeourjointideas.1.Wereferto[3,4]forthenotionofspectralmultiplicitytheoryintheseparableHilbertspace.LetHbeacyclicHilbertspacewiththeresolutionoftheidentity{p(t)}ofthemaximalspectraltypelIP(dt)ll'~dt-ordinaryLebesguemeasure.Anon-anticipativelineartransformationsisdefinedbyVolterrakernelg(t,u),uSt,ajsf'g(t)u)p(du),t>0…     

数值求解Euler方程的UCGVC差分格式的注记

１．引言 近年来高精度差分格式的研究引起国内外的普遍重视,目的是更准确地模拟复杂流场的流动．众所周知,传统的二阶ＴＶＤ类格式虽然能较好地捕捉激波,但却存在局部极值点降阶的问题,而且由于一些格式的数值粘性过大,当用该格式计算粘性流特别是高雷诺数问题时,格式本身的数值粘性可能掩盖了流场的物理粘性,从而降低了格式对边界层的分辨率,因而无法正确计算热流值。文献［３］指出,采用高精度格式可适当放松对网格雷诺数的要求,因此发展三阶或三阶以上的格式是需要的。近年来,人们已经发展了一些无伪振荡的高阶格式,如ＥＮ…     

铁磁链Landau-Lifshitz方程的显式差分法

正如在研究流体动力学时,Ｎａｖｉｅｒ－Ｓｔｏｋｅｓ方程起着十分重要的作用一样,在对于非平衡态磁学的研究中,描述连续铁磁体自旋场发展过程的 Ｌａｎｄａｎ－Ｌｉｆｓｈｉｔｚ方程［１］起着十分重要的作用［２］．一九九三年,美国和印度签署了一个大约 ２８０万美元的合作研究计划,在三年的时间里,对Ｌａｎｄａｕ－Ｌｉｆｓｈｉｔｚ铁磁链方程进行研究．在无阻尼的情况下,它为一完全可积的孤立子系统［３,４,５］。很多物理学家研究了它的孤立子解的存在性、逆散射方法以及相互碰撞［３,４,５］．关于解的存在性, Ａｌｏｎ…     

An Explicit Pseudo-Spectral Scheme with Almost Unconditional Stability for the Cahn-Hilliard Equation

In this paper, an explicit fully discrete three-level pseudo-spectral scheme with almost unconditional stability for the Cahn-Hilliard equation is proposed. Stability and convergence of the scheme are proved by Sobolev's inequalities and the bounded extensive method of the nonlinear function (B.N. Lu (1995)). The scheme possesses the almost same stable condition and convergent accuracy as the Creak-Nicloson scheme but it is an explicit scheme. Thus the iterative method to solve nonlinear algebraic system is avoided. Moreover, the linear stability of the critical point $u_0$ is investigated and the linear dispersive relation is obtained. Finally, the numerical results are supplied, which check the theoretical results.