Archive

25 December 1988, Volume 5 Issue 4 Previous Issue    Next Issue

---

For Selected:

Download Citations EndNote Ris BibTeX

Toggle Thumbnails

Select

NUMERICAL SIMULATION OF UNSTEADY DETONATION BY METHOD OF CHARACTERISTICS Wang Zi-xiu, Li Hua 1988, 5(4): 383-393.  Abstract ( )   PDF (603KB) ( )   A phenomenological model of shock initiation and detonation of heterogeneous explosives is implemented in the one-dimensional hydrodynamic code based on the method of characteristics. Since there is no artificial viscous term to smear the shock, we can get a clear picture of the simulated structure of detonation wave. We have calculated Unsteady detonation process of PBX-9404 and the results are quite well consistent with the experiment data.

Select

THE NUMERICAL SOLUTIONS OF THE SEPARATION FLOW USING THE IMPROVED FINITE ANALYTIC METHOD Jiang Zong-lin 1988, 5(4): 394-402.  Abstract ( )   PDF (547KB) ( )   In this paper, the improved finite analytic method has been proposed for calculating the fluid flow with the large separation region at the high Reno-Ids number. By this algorism, the complex operations caused by exponential functions and series sum can be removed and a lot of the computational time can be saved. The Renolds number of the flow which can be resolved with this method is about twenty times as much as that of the flow with the or iginal one on the same computational grids. Numerical experiments show that reults obtained with both methods at the same boudary conditions and the same Renolds number are extremely similar and the solutions of the flow at the high Renolds number are satisfactorily.

Select

ELECTRON IMPACT EXCITATION OF ATOMS (IONS) BY HIGH-ENERGY ELECTRON Fang Quan-yu, Ji Wen-gui, Cai Wei, Qiu Yu-bo 1988, 5(4): 403-419.  Abstract ( )   PDF (867KB) ( )   In astrophysics, plasma, quantum chemistry and advanced technique, an increasing demand for collision data is evident. Among those data,cross section of electron impact excitation is important and fundamental data. In this paper, we present the electron impact excitation cross section for some ions at high-energy incident electron. In our calculation, the LS coupling antisymmetric single configuration wavefuction of the target has been used. The plane wave BORN approximation has been used in treating the collision problem. Angular part of the transition matrix element of complex atom is treated carefully. We have computed single electron radial wavefunction of atom by means of HARTREE-FOCK-SLATER self consis-tent field method. We call our computer program as PBA(9). By thiscomputer progrom, we can compute excitation cross sections, collision strengths, generalized oscillator strengths and optical dipole oscillator strengths. In fact, we have computed widely from H-like to Ne-like iso-electronic sequences. It is made to compare our results with the other ones available. It shows that our results are satisfying for high energy electron. For example, the values of the excitation cross section for C and O fall into the accurate region of recommend data given by Japan's center NAGOYA of atamic data. We conclude that the method used in this paper is reliable and the results are correct. It can provide us with electron-impact excitation useful informantion of light atoms (ions)by high energy electron.

Select

HEAT TRANSFER AND NATURAL CONVECTION COMPUTATION FOR THE SOLIDIFICATION PROCESS OF CASTINGS Zhao Yong, Liu Zhuang 1988, 5(4): 420-429.  Abstract ( )   PDF (688KB) ( )   A mathematical model using stream function, vorticity and cemperature as basic varibles is developed to represent the heat transfer and natural convection in the solidification prooess of castings. The solutions of the governing equations are carried out by means of a semi-implicit finite difference scheme based on fixed grid methodology. Consideration is given to the latent heat evolution and velocity conditiion in the liquid-solid phase zone(inte-rface).A specific enthalpy method associated with the alloy equilibrium phase diagram is employed to account for the latent heat of phase change. A new revision approach according to solid fraction is proposed for the treatment of the zero velocity condition as liquid region turns to solid. Moreover, the possibility of applying the present model to the problems of irregular geometries is discussed. In the last part of the paper, a tpyical freezing problem is solved, and the influence of natural convection on the solidification of castings is analysed in detail.

Select

A STRANGE ATTRACTOR IN HOMOGENEOUS TURBULENCE Fong Ching-Zhen, Gu Zhi-fu, Chen Wei-yan 1988, 5(4): 430-436.  Abstract ( )   PDF (481KB) ( )   In this paper we calculate the correlation dimensions, entropies, and the largest Lyapunov exponents of homogeneous furbulence signals by making use of the technique of phase space reconstruction with delay coordinates. The results show that this is a kind of deterministic chaos perturbed by the random noise, its underlying attractor is a strange attractor with the background of random noise.

Select

AB INITIO CALCULATION OF THE POTENTIAL CURVES FETWEEN ATOMS (Li-Li, F-F) Cui Zuo-lin, Pan shou-fu 1988, 5(4): 437-442.  Abstract ( )   PDF (360KB) ( )   Using ab initio program MONSTERGAUSS and various basis sets, we calculated the potential curves between Li-Li and F-F atoms at wider nuclear distances. We also calculated the equilibrium nuclear distances of ground state molecules Li2 (1Σg+) and F2(1Σg+) and election corselation erergy with configuration interaction method(CI). The calculated results were discussed and compared with other's calculated and experimental results.

Select

AN OPTIMAL CONTROL METHODS OF THE NUMERICAL COMPUTATION Zhang Suo-chun 1988, 5(4): 443-454.  Abstract ( )   PDF (628KB) ( )   This paper is in order to solve the numerical computation of the periodic orbits of nonlinear autonomous differential equations. We design an optimal control method of the numerical computation. The solution method is basically a continue shooting method,under the guardianship of nonlinear least squaves that takes advantage of an optimal Central formulation, in which attached to the contral techniques of the penalty function.In this method,we use the Crank-Nicolson discretization schemes and Conjugate Gradient methods.This method has been applied to the solution of nonlinear differential equations modelling chemical reactions,the corresponding numerical results are presented here.

Select

APPLICATION OF THE SOLUTIONS OF THE HYPERGEOMETRIC EQUATION TO FUNCTION EXPANSIONS Zhu Jia-lin 1988, 5(4): 455-465.  Abstract ( )   PDF (543KB) ( )   Using the Solutions of the hypergeometric equation at the rauge around a singular point, the expansion form of a function related with hypergeometric function are obtained at the same range.For the elliptic complete infegrals of the first and second kind,we show not only the forms of the expausions but also the numerical results of them.

Select

THE FIRST REREARCH FOR REZONING TECHNICH Zhou Lian-sheng, Bei Xin-yuan, Zhang Zhi-jie 1988, 5(4): 466-472.  Abstract ( )   PDF (439KB) ( )   In this paper, the rezoning technique for Lagrangian meshes is discussed with equidistant principle. A Rezoning threshold value η and fourstep rezoning methed are applied. According to volume ratio, the concerned parameters of meshes between new and old one are redistributed conservatively. The numerical tests show that in order to improve the effect of rezone, not only the position of inner mesh points but also the position of bourdary points are needed to be readjusted.

Select

A HIGHER ORDER ACCURACY NUMERICAL QUADRATURE RULE Wu Xin-yuan 1988, 5(4): 473-477.  Abstract ( )   PDF (261KB) ( )   This paper discusses the following quadrature rule with higher degree of accuracy ∫abf(x)dx)=(b-a)(7f(a)+16f((a+b)/(2))+7f(b)+(b-a)(f'(a)-f'(b)))/30+E[f] Where E[f]=((b-a)7/(604800))f(6)(ξ),a<ξ<b andits Compositrule are presented aswell:∫ab(f(x)dx)=(b-a)((7f(a+2ih)+7???19880410-2???f(a+2ih)+16f(a+2i-1h)+(b-a)(f'(a)-f(b))/2n)/30n+En[f] Where En[f]=((b-a)7)/(604800n6)f(6)(η),a< η< b h=(b-a)/2n which possess the all advantages of simpsons rule, but the degree of accuracy isincreased by two order than Simpson's rule.The numerical tests that the quadrature formula of this paper is very efficient.

Select

A HIGH ORDER FINITE ELEMENT BOUNDARY INTEGRAL EQUATION METHOD Jiang Yu Guo, Kuan-Hang 1988, 5(4): 478-483.  Abstract ( )   PDF (298KB) ( )   The finite element boundary integral equation method to solve heat conduction problems was presented in the paper[1]. A high order element that bases upon the reference[1] is introduced in this paper.The formulas of high order element method are deduced in detail.If the element numbers are same,the results of seueral examples show that the high order element can get much better results than the linear element.

Select

PROBING INTO MECHANISM OF NUCLEAR FISSION BY HYPERNUCLEI A238U Zheng Guo-tong, Chen Zhe-qin, Qiu Zhi-hong 1988, 5(4): 484-492.  Abstract ( )   PDF (510KB) ( )   The final populated probabilities of A after A238U asymmetric fission have been calculated for vanious scission process such as constant speed or accelerated speed of scission approached from some proper rising potential barries.The results show these two processes of sciessien are distinguishable by observing the final populated probabilities of A.

Select

NUMERICAL SIMULATION METHOD FOR HEAT TRANSFER OF THE COOL INGOTS HEATED IN STEEL MILL SOAKING PITS Chen Bai-ren 1988, 5(4): 493-500.  Abstract ( )   PDF (465KB) ( )   In this paper, based on[1], the numerical simulation method of the circular tube mathematical model for cool ingots heated in soaking pits is derived further, This method has been applied successfully in computer control system of soaking pits in the blooming rolling mill of Pan Zhi-Hua Iron and Steel Company.

Select

A ASYMPTOTIC NUMERICAL METHOD FOR THE STEADY-STATE CONVECTION DIFFUSION EQUATION Wu Qi-guang 1988, 5(4): 501-506.  Abstract ( )   PDF (318KB) ( )   In this paper,A asymptotic numerical method for the steady-state Convection diffusion equation is proposed,which need not take very fine mesh size in the neighbourhood of the boundary layer.Numerical computation for model problem show that we can obtain the numerical solution in the boundary layer with moderate step size.