Loading...

Archive

    25 September 1991, Volume 8 Issue 3 Previous Issue    Next Issue

    For Selected: Toggle Thumbnails
    FR METHOD OF NUMERICAL INVERSION OF ACOUSTIC WAVE EQUATIONS
    Zhang Wenfei
    1991, 8(3): 225-229. 
    Abstract ( )   PDF (312KB) ( )  
    This paper presents a new method, FR method, to solve the inverse problems of acoustic wave equations. FR method is a general numerical inversion algorithm of acoustic wave equations. It can be used to one, two or three dimensional inverse problems with any different boundary conditions. The basic principle and formulae are derived carefully in the paper, and numerical examples of one and two dimensional inverse problems are given. The numerical results show the accuracy and stability of the method.
    DOUBLE-TIME-SCALE CIC ALGORITHMS FOR SIMULATION OF PLASMAS
    Nie Xiaobo, Zhang Jiatai, Xu Linbao
    1991, 8(3): 230-234. 
    Abstract ( )   PDF (300KB) ( )  
    In this paper we present a double-time-scale cloud-in-cell (CIC) algorithm for simulation of plasmas which reduce by half-quarter computational cost of standard CIC methods. The algorithm is integrating electrons by small time steps and integrating ions by large steps. The algorithm have been used in one and a half-dimensional CIC scheme. The results of theoretic analysis and simulation support the stability and correctness of the algorithm.
    AN ADAPTIVE GRID EMBEDDING TECHNIQUE WITH HIGH RESOLUTION FOR NUMERICAL SIMULATION OF INVISCID SUPERSONIC JET/EXTERNAL-FLOW INTERACTIONS
    Wu Songping, Chun-Hian Lee
    1991, 8(3): 235-242. 
    Abstract ( )   PDF (775KB) ( )  
    A procedure based on an adaptive grid embedding technique is proposed in the present paper to simulate the complex flow fields induced by the interactions between supersonic jets and external flows. The proposed procedure utilizes the high-resolution TVD scheme for solving the local flow field where flow parameters are found to vary drastically, and any conventional second-order scheme can be used to solve the problem in the remaining region. To serve the purpose, a single-dimensionally adapting technique is developed and applied successively, in conjunction with the Strang-type splitting scheme, to solve the multi-dimensional gas dynamic equations.As a demonstration, solutions of the Euler equations for a flow of supersonic jet interacting with an external flow computed using the adaptive grid technique developed in this work, as well as the conventional TVD procedure, are presented in this paper. The numerical results reveal that the present technique is capable of providing visualizations of complex flow fields with high resolution. Yet, through comparisons, it is also found to be highly efficient.
    THE ENERGY LOSS AND DISTRIBUTION OF THE ESCAPED ALPHAS ON THE FIRST WALL IN A TOKAMAK REACTOR
    Qi Changwei
    1991, 8(3): 243-248. 
    Abstract ( )   PDF (380KB) ( )  
    In this paper.using Monte-Carlo methods,we have calculated the loss and distribution of the energetic alphas on. the first wall for four groups of plasma parameter. We get conclusions as follows: The collisional diffusion energy loss in slowing-down process is much larger than the drift loss for a parabolic plasma; the energy loss caused by the ripple is proportional to the square root of the ripple; the point of the maximum of the loss energy along toroidal angle on the first wall deviates from the minimum point of the toroidal magnetic field due to the toroidal magnetic field and the direction of the deviation is decided by the direction of the toroidal current; whether the escaped alpha particles bomb at the upper part or the lower part of the first wall depends on the direction of the toroidal magnetic field.
    EDDY SEPARATED METHOD FOR SOLVING CONVECTION-DIFFUSION EQUATION
    Zhang Shixiong
    1991, 8(3): 249-256. 
    Abstract ( )   PDF (561KB) ( )  
    The convection-diffusion equation is a fundamental equation in many fluid flow problems. The numerical solution of this equation will lead to an algebraic equation set of high order, the solving of which by conventional methods needs large amount of computer space and computing time. This paper proposes an eddy separated method. Taking the advantage of the upwind nature of the convection-diffusion equation, this method separates the eddy regions from the convention-dominated region, then establishes and solves several smaller algebraic equation sets for the flow in the eddy regions. In the convection-dominated region, on the other hand, a series of explicit schemes can be simply used. This method keeps high accuracy.The results are as same as those obtained with previous methods. Because the order and bandwidth of these algebraic equation sets are much smaller than those formulated with previous methods, both computer space and computing time can be greatly reduced. This method is especially efficient for flows of higher Reynolds numbers with local eddy regions.
    SECOND ORDER ACCURATE MmB SCHEMES FOR 2-D NONLINEAR CONSE RVATION LAWS ON REGULAR TRIANGULAR MESHES
    Yang Shuli
    1991, 8(3): 257-263. 
    Abstract ( )   PDF (367KB) ( )  
    In this paper, a class of second order accurate MmB (locally Maximum_minimumBounds preserving) schemes is constructed for initial value problems of 2_D nonlinearconservation laws on regular triangular meshes, and the numerical solutions for Riemann problems of 2_D inviscid Bergers eqution are given by using these schemes. The numerical results showthat the schemes have high resolution and nonoscillatory properties.
    VIBRATION ANALYSIS OF UNIFORM COLUMNS WITH ARBITRARILY SHAPED CROSS SECTIONS PARTIALLY SUB MERGED IN WATER
    Zhou Ding
    1991, 8(3): 264-270. 
    Abstract ( )   PDF (401KB) ( )  
    This paper studies the problem of free bending vibration of uniform columns with arbitrarily shaped cross sections partially submerged in water. A general, exact, analytical solution based on Fourier collocation method is presented to solve the dynamic characteristics of coupled vibration of columns with water. Finally, some numerical results are given for elliptical columns submerged in water.
    A MODIFIED NEWTON'S METHOD FOR FULLY IMPLICIT DIFFERENCE SCHEMES OF GAS RESERVIORS SIMULATION AND NUMERICAL COMPARISON OF RESULTS
    Lin gang, Shi Jimin, Lin Zhengbao, Lü Tao, Lin Aimin
    1991, 8(3): 271-278. 
    Abstract ( )   PDF (445KB) ( )  
    The numerical simulation of oil (gas) reservoirs can be described by two-phase penetration dynamic equations, which are high-dimensional, non-linear, singular and time-dependent partial differential equations, subjected to some appropriate initial and boundary conditions.The partial differential equations are discretized by differencing schemes (such as Fully Implicit method, 1MPES,etc.)and produce different non-linear algebraic systems.To solve a non-linear system Newton's method is the most popular method. But it requires calculation of a Jacobi matrix and solving a linear system in each iteration.These will consume huge computing CPU time.For saving the CPU time we have modified the Newton's method with Samanskii's idea and controled the number of iterations in solving the linear systems. The programming is quite simple. We tested our algorithm in a set of ideal data.The results are as accurate as those obtained from the Newton's method and the efficiency is better.
    KACZMARZ ITERATION AND COMPUTATION OF FLOWS
    Shi Jinsong, Zhuang Weihua, Zhang Weijun
    1991, 8(3): 279-286. 
    Abstract ( )   PDF (423KB) ( )  
    In this paper a numerical method, which is composed of the finite element method and Kaczmarz iteration, solving for two-dimensional steady flows is presented. The numerical example using the method is given.
    UPWIND COMPACT SCHEME WITH DISPERSION CONTROLLING FOR THE AERODYNAMIC EQUATIONS
    Ma Yanwen, Fu Dexun
    1991, 8(3): 287-294. 
    Abstract ( )   PDF (480KB) ( )  
    A upwind compact scheme with dispersion regulator is developed and discussed. The scheme developed is dissipative and the dispersion term in the truncation error can be controlled. According to the behaviour of the flow parameters near the shock the scheme with specially chosen control function gives high resolution. The scheme is simple, efficient and the shock can be captured well. The scheme is used to compute the shock reflection problem, and the computed results are satisfactory.
    THE SOLUTION OF THE EIGENVECTORS IN ONE DIMENSIONAL DISORDERED SYSTEM
    Xu Hui
    1991, 8(3): 295-304. 
    Abstract ( )   PDF (604KB) ( )  
    A method for solving directly the eigenvectors of a Hermitian five diagonal matrix is developed. It does not use a block tridiagonal matrix to solve. The results of calculation demonstrate that this method with high accuracy and low storage is very suitable to the disordered sys tem. It can be used to calculate high order matrix in order to meet the needs of disordered system.
    A METHOD OF IRREGULAR DIFFERENCE MESHES FOR 2-DIMENSIONAL RESERVOIR SIMULATIONS
    Ma Jun, Lei Guangyao
    1991, 8(3): 305-311. 
    Abstract ( )   PDF (429KB) ( )  
    Using a difference method of irregular meshes, a numerical model for 2-dimensional oil-water reservoir simulations is presented and its numerical results of two examples have been given when IMPES method is adopted. The numerical results of example 1 are in good agreement with the analytic solution. The number of difference meshes used for example 1 is less than that used in Petrosa's model of hybrid grid for the same example and the method given in this paper is more efficient and convenient. It is difficult to solve the problem of example 2 for the difference mehtod of rectangular meshes. However, satisfying results are also obtained when the method given here is used for the numerical solution of the example 2. The results show that this method of irregular difference meshes can be applied flexibly to probems of reservoir simulations with a complex boundary and a complex structure of stratum.
    HOURGLASS CONTROL IN THE QUADRILATERAL ELEMENT
    Wang Xiaojun, Hu Xiuzhang, Liu Shengqiu
    1991, 8(3): 312-318. 
    Abstract ( )   PDF (435KB) ( )  
    The hourglass deformation in the quadrilateral element is discussed in this paper. It is shown that the hourglass modes can be controlled in the computation of large elastic plastic deformation by introducing the hourglass strains and the corresponding stresses. Numerical examples that a high velocity rod impacts a plate target are used to illustrate the effectiveness of the method by the aid of a two dimensional Lagrange program. The results show that the hourglass control proposed here is feasible.
    CALCULATION OF ONE DIMENSIONAL LANDSLIP PROBLEM WITH TVD SCHEME
    Guo Wenhai
    1991, 8(3): 319-324. 
    Abstract ( )   PDF (324KB) ( )  
    One dimensional landslip problem is calculated with TVD scheme. The numerical result basically coincides with the actual result. TVD scheme has been proved to be very efiective in solving wide aerodynamic problems. The present research also shows that TVD scheme can be used to slove more physical problems in which the discontinuity occurs.
    THE TRANSPLANTATION, AUGMENTATION AND APPLICATION OF ORIGEN-2 CODE
    Li Xingyue, Gui Shouzao
    1991, 8(3): 325-329. 
    Abstract ( )   PDF (327KB) ( )  
    ORIGEN-2 is a versatile point depletion and decay computer code. Its functions, major charactaristics, and theoreticahmodels are briefly introduced in here. The transplantation of the code on IBM 4300 series computers and its augmentation of output photon data and application in calculating gamma-ray spectra of fission products are described.
    A MONTE CARLO CODE SIMULATING THE EXPOSURE TO A MAN BY PHOTONS
    Li Xingyue, Cai Jianhua
    1991, 8(3): 330-336. 
    Abstract ( )   PDF (461KB) ( )  
    A Monte Carlo code applied to radiation protection and medical radiology is presented. It simulates the irradiation caused by photons and can give the effective dose equivalent (He)and absorbed dose equivalents of 20 organs (tissues) in a reference man. A special method which divides all photon histories into several kinds and traces each kind separately was adopted to improve the precision of the calculated results.The code can be also used to other calculations of doses in systems with complex geometry.
Share: