Archive

25 September 1993, Volume 10 Issue 3 Previous Issue    Next Issue

---

For Selected:

Download Citations EndNote Ris BibTeX

Toggle Thumbnails

Select

DESIGN OF FOCUS DEFLECTION SYSTEM FOR ELECTROSTATIC DEFLECTION TYPE CAMERA TUBES (2) ELECTRON TRAJECTORIES CALCULATION Yang Guowei, Guan Yonggui 1993, 10(3): 257-265.  Abstract ( )   PDF (487KB) ( )   In the previnus paper, the field calculatin for focus deflection system of a cam-era tube with electrostatic deflection yokes were deseribed. Now that the potential field is known we can treat the trajectories described by the electrons in this field. The electron trajectory calcula-tion in the focus-deflection system with the electrostatic deflection differs from usual trial solution determined by the discrete methods in the rotationally system field. Because poteotial fuoction of the focus-deflection field determined by the analytic method are analytic fuoctions of the variations r, θ, z, in the tracing process of the electron trajectories the potential and it's deriva-tives at an arbitrary point may be given by the aoalytic fuoctions.The purpose of the paper is to give the method of solving geoeral equation of motion for an electron in mixed field which produces a combined electromagnetic focus field and the electrostatic deflection field.

Select

ELECTROMAGNETIC WAVE MODE IN ONE DIMENSION PERIODIC LOSSY MEDIA AND ITS NUMERICAL ANALYSIS Li Chengzu 1993, 10(3): 266-272.  Abstract ( )   PDF (445KB) ( )   Electromagnetic wave eigenmode equation in one dimension periodic lossy medium is derived. It has been found that the solutions of the equation can be classified as a series of conduction and forbidden bands. Substituion of the real Re(K2) for complex K to symbolise modes shows the band structure clearly and facilitates classifing and comparing of the modes. The figures of the lowest two bands of TE wave and TM wave for a group of parameters are given by using numerical method. They are compared with electron energy band in periodic potential field.

Select

FULL FLOW SIMULATION IN A THREE-DIMENSIONAL SQUARE ENCLOSURE WITH SPACERS Yu Liren, Chen Dawei 1993, 10(3): 273-278.  Abstract ( )   PDF (429KB) ( )   The block-up coefficient technique and the finite volume approach are used to solve numerically the complete N-S equations in a square enclosure with spacers. A pressure-velocity correction algorithm (SIMPLE-C, Semi-Implicit Method for Pressure Linked Equation-Consistent) is adopted in the direct calculation of all the physical variables. The applicability, feasibility and efficiency of the block-up coefficient technique applied to simulate three-dimensional viscous flows are investigated.The primary results have shown that the block-up coefficient technique is indeed a simple and convenient numerical approach, with encouraging prospects, for computing the velocity, pressure and other physical quantities for the flows in a channel with complex boundary shape.

Select

PARALLEL θ-BLOCK PREDICTOR-CORRECTOR METHODS FOR ORDINARY DIFFERENTIAL EQUATIONS Li Linzhong, Xue Xiaohui 1993, 10(3): 279-289.  Abstract ( )   PDF (562KB) ( )   This paper gives a class of parallel block predictor-corrector methods with a parameter θ for solving initial value problems in ordinary differential equations. The stability and convergence on iterative processes are discussed. It is proven that the methods not only have large stability intervals but also have large convergent regions, thus they are very suitable for par allel integration of ordinary differential equations.

Select

THE EVALUATION OF SINGULAR INTEGRALS IN, POTENTIAL FIELDS AND THE SLOPE VARIABLE Bian Shaofeng, Wu Xiaoping, Liu Quanwei 1993, 10(3): 290-296.  Abstract ( )   PDF (359KB) ( )   Definiting the slope y/x as a new variable z, the variable transformation from (x, y) to (x, z) is presented. It is found that such a set of new variable (x,z) almost makes the singular integrals dependent on the distance reciprocal r in potential fields become non-singular. By using the new variable, analytical expressions concerning common singular integrals in potential fields are derived as well. Numerical results show that the numerical integration con verges very fast with the method.

Select

THE PARTIAL SEPARATION OF VARIABLES-FRACTIONAL STEPS DIFFERENCE METHOD FOR SOLVING 3-DIMENSIONAL ELASTIC WAVE FIELD IN CYLINDRICAL COORDINATE SYSTEM He Bairong, WangZhanbiao, Beijing Feng, Nie Yongan, Jin Jiange 1993, 10(3): 297-308.  Abstract ( )   PDF (559KB) ( )   An analytical-numerical method is developed for studying the diffraction of elastic wave by an abnormal cylinder embeded in layered media. The method is based on a combination of partial separation of variables and fractional steps difference techniques in cylin-drical coordinate system. For examining the efficiency of this method a calculation example is given.

Select

THE NUMERICAL SIMULATION FOR SUPERSONIC INVISCID STEADY FLOW PAST THREE-DIMENSIONAL BODY Chen Rangfu 1993, 10(3): 309-317.  Abstract ( )   PDF (529KB) ( )   An implicit upwind finite volume scheme wihch combines a kind of very simple flux splitting with a modified numerical flux method of second order accurate TVD scheme is presented in this paper using this method, we performed numerical simulation for the supersonic inviscid steady flow past an asymmetric three-dimensional body based on unsteady gasdynamics Euler's equations. The numerical results show this method has high resolution for discontinuity and is time-saving.

Select

MOLECULAR DYNAMICS RESEARCH OF EJECTION AND DAMAGE OF METALS INDUCED BY REFLECTION OF SHOCK WAVE AT FREE SURFACE Zhang Jinglin, Wang Jihai, Yang Shuxia 1993, 10(3): 318-324.  Abstract ( )   PDF (584KB) ( )   The two-dimensional molecular dynamics calculation is used to simulate ejection and damage phenomena. The intermolecular force is described by Morse potential. Molecules are initially located in equilibrium position and spaced so that the free surface has an angle shape for simulating real machined surface. Shock wave may be produced by motion of first layer of molecules (piston) or by impact of flyer of molecules. When the disturbance arrives at free surface, the boundary molecules start to run away from it with large velocity and the inside damage begins. The ratio of ejection velocity to the surface one changes from 1 to 3 and it is dependent on the shock strenth, surface angle and material property. When the semiangle is bigger than 60°, the boundary molecules can not run away from others, i.e. the ejection disappears. It also disappears when shock strength is lower than a limit value dependent ont he surface angle and material. In the both ejection and jetless cases the calculation shows the process of microvoid formation and growth.

Select

A NEW SPACE MARCHING METHOD FOR SUPERSONIC-VISCOUS FLOW OVER BLUNT-BODY Lin Mingsen, Fu Dexun 1993, 10(3): 325-330.  Abstract ( )   PDF (324KB) ( )   A new space marching method is given to compute supersonic flow fields surrounding a Blunt-Body body. The computations have been performed for flow conditions convesponding to an exensive series availabe of experimental surface pressures and heat-transfers.

Select

REFLECTION ADN TRANSMISSION OF WAVES IN DIMENSION-CHANGABLE 2/R FRACTAL CANTOR LAYERS Chen Lujun, Li Cunzhi, Ma Yonggen 1993, 10(3): 331-335.  Abstract ( )   PDF (273KB) ( )   The wave reflection and transmission properties in the dimension-Changable 2/R cantor fructal layers are analysed. The formulae determining the reflection and transmission spectra with a self-similar procedure are given. This new physical reflection spectra with a quasi-self-similarity is significant for certain applications.

Select

INTERPOLATING MATRIX METHOD FOR MULTI-POINT BVPS AND ITS ERROR ANALYSIS Niu Zhongrong 1993, 10(3): 336-344.  Abstract ( )   PDF (436KB) ( )   Interpolating matrix method is a numerical method for solving multi-point boundary value problems. This paper provides the error analysis of the method, proving the same precision for the solutions y(x), y'(x), ..., y(m)(x)obtained by the method. Author gives the convergence order estimate and proves good stability property of the method the second order scalar equations.

Select

AN IMPROVEMENT OF G-P ALGORITHM AND ITS APPLICATION Wang Fuquan, Luo Chaosheng, Chen Guoxian 1993, 10(3): 345-351.  Abstract ( )   PDF (426KB) ( )   In this paper, we give the two improved formulas of distance, which can be used in the G-P algorithm, and the corresponding recurrence formulas, which improves the G-P algorithm,We present a rule and a mathematical model to determine no-scale interval. Such a algorithm to solve the end points of the interval can be applied to the signal processing of the seismic prospecting. The no-scale interval and the fractaldimension of time data series of the seismic tunnel are computed as well in our work.

Select

TWO-DIMENSIONAL APPROXIMATE FINITE ANALYTIC COEFFICIENTS Zhang Shixiong 1993, 10(3): 352-358.  Abstract ( )   PDF (449KB) ( )   The Finite Analytic Method (FA)is an efficient numerical method to solve the fluid flow problems . However, the calculation of Finite Analytic Coefficients by the conventional method is bored and time consuming . Moreover,in some extreme cases the resulted coefficients may become totally erroneous. This paper presents an approximate formula to determine FA coefficients . which is convenient to use. The error caused by the approximaion is less then 0.003.

Select

THE NUMERICAL SOLUTION OF INTERNAL AND EXTERNAL FIELD OF INVISCID AND VISCOUS FLOW THROUGH CONVERGENT NOZZLES Shao Jifeng, Wen Gongbi, Wu Wangyi, Dong Songye 1993, 10(3): 359-363.  Abstract ( )   PDF (308KB) ( )   The numerical solution of inviscid and viscous flow through convergent nozzles is presented from Euler equation and N-S equation. MacCormack explicit difference scheme is used with an improved artificial viscous term. The method employs the split of zone to overcome the instability near exit of the nozzeles. The results of a typical calculation are in good agreement with experimental data.

Select

Research on the Covariance Matrix of the Evaluated Data Wang Zisheng, Cai Dunjiu 1993, 10(3): 364-368.  Abstract ( )   PDF (241KB) ( )   By comparing the weighted mean values for experimental data with the recommended ones in a variety of small energy interval for the sets of reaction cross sections, a system of linear equations is proposed for determining the adjusted factors generated by corrections for the experimental data, which may be used to solve the covariance problems on the evaluated data with all physical corrections. The covariance matrices of some dosimetry reaction cross sections are calculated by this method.

Select

THE INVESTIGATION ON UNSTEADY AERODYNAMICAL CHARACTERISTICS OF AN OSCILLATING TRAIL-EDGE FLAP Wang Baoyu, Li Feng, Wang Yiyun, Cui Erjie 1993, 10(3): 369-372.  Abstract ( )   PDF (300KB) ( )   Some aerodynamic characteristics of an oscillating trail-edge flap were investigated by numerical simulation, such as the unsteady effect of flow field and flow field comparison between different frequences or amplitudes. The Euler equations were solved by LU-ADI algorithm, the grid is C type grid. The numerical results showed that: (1) with the oscillation of trail-edge flap, the shock strength and its location are nonlinear to the motion of the trail-edge flap, (2) the higher the reduced frequence, the less the influence to variable pressure jump, (3) with the increase of the amplitude, the variations of shock strength and its location become more seriously.

Select

PARALLEL DIRECT INTEGRATION ALGORITHM IN DYNAMIC ANALYSIS OF STRUCTURES ON YH-1 COMPUTER Zhou Shuquan, Liang Weitai 1993, 10(3): 373-378.  Abstract ( )   PDF (324KB) ( )   An efficient parallel direct integration algorithm is presented. Computational results for a model in YH-1 computer show that for the model of degree of freedom of 1666, the speed of the parallel algorithm is 30 times faster than of traditional serial algorithm, and if the features of YH-1 computer considered and the assemble language used, it is 50 times faster in the same model case.

Select

DISCRETE NEWTON REGULARIZATION METHOD AND ITS APPLICATION Han Bo, Liu Jiaqi 1993, 10(3): 379-384.  Abstract ( )   PDF (304KB) ( )   In this paper, Discrete Newton Method for solving nonlinear operator equations and Tikhonov's regularization method for solving linear ill-posed problems are combined into a stable iterative scheme for solving nonlinear operator equations with hard computed and ill-conditioned derivatives. The results of convergence analyses and error estimates are ob-tained. The application of this method for solving the inverse problem on remote sensing of the thermal conductivity in the nonhomogeneous material is demonstrated. It is shown by numerical simulations that this method is feasible and effective.