Archive

25 June 1995, Volume 12 Issue 2 Previous Issue    Next Issue

---

For Selected:

Download Citations EndNote Ris BibTeX

Toggle Thumbnails

Select

ELECTRON TRANSPORT PROCESS IN CATHODE SHEATH OF HELIUM DC GLOW DISCHARGE PLASMA Wei Helin, Liu Zuli 1995, 12(2): 145-152.  Abstract ( )   PDF (521KB) ( )   The electron transport process in cathode sheath of helium dc glow discharge is studied by using Monte Carlo simulation. The electron is accelerated between collisions by. the non-uniform electric field. The electron free flying time between collisions is determined by electron-neutral collision frequency. The proposed method may reduce complicated calculation and save much computer time. The experimental data of collision cross section are fitted with semi-experience expressions. The electron mean energy, the energy spatial distribution, and other parameters distribution are studied as well.

Select

FOURIER SERIES SOLUTION OF THICK/THIN BEAM LAID ON ELASTIC FOUNDATIONS AND TENSIONLESS ELASTIC FOUNDATIONS Bu Xiaoming 1995, 12(2): 153-156.  Abstract ( )   PDF (234KB) ( )   An elastic foundation model is proposed, and the general form of equilibrium equations.for the thick/thin beam on elastic foundations is established. Their solutions are presented in terms of Fourier series.

Select

THE GAUSSIAN FUNCTION EXPANSION METHOD OF ATOMIC SELF-CONSISTENT-FIELD COMPUTATIONS IN TWO DIMENSIONS Zhao Yongfang, P. Pyykkö 1995, 12(2): 157-163.  Abstract ( )   PDF (406KB) ( )   Self-consistent-field (SCF) method using signle electron wave functions expanded in terms of Gaussian functions is reported for atoms in two dimensions. The energy expressions are derived for closed-shell and open-shell configurations comprising, s, p, and d electrons and Hartree-Fock-Roothaan two-dimensions equation for closed-shell atoms. Finally, main results of SCF numerical computations are given and discussed.

Select

A FAST AND RECURSIVE METHOD FOR SOVLING THE HIGH MULTIDIMENSIONAL PATH INTEGRALS Bao Jingdong 1995, 12(2): 164-168.  Abstract ( )   PDF (296KB) ( )   A recursive Monte Carlo method to calculate quantum partition function expressed by N-multidimensional path integrals is presented. The stability of numerical results dependent wpon the number of path discretization and total pathe number is discussed as well.

Select

MONTE CARLO SIMULATION OF LOW ENERGY ELECTRON SCATTERING IN ALUMINUM TARGET Tan Zhenyu, He Yancai 1995, 12(2): 169-173.  Abstract ( )   PDF (321KB) ( )   Based on the privious paper (1) and using the Mott cross-section by solving relativistic Dirac equation for elastic scattering, Monte Carlo simulation of low energy electron scattering in Al bulk target and film have been performed. The calculated backscattering coefficients, the energy distribution of backscattered electrons and electron transmission coefficients are in good agreement with experimental results. The spatial distribution of low-loss backscattered electrons is also calculated and exhibits a high resolution.

Select

NUMERICAL SIMULATION OF NONLINEAR CHROMATOGRAPHY PROCESS USING TVD AND MmB SCHEMES Lin Bingchang, Wang Jida, Yang Shuli 1995, 12(2): 174-178.  Abstract ( )   PDF (291KB) ( )   The nonlinear chromatography process is simulated with high precision and resolution schemes to catch the shock, which are TVD (Total Variation Decreas) scheme and MmB (Locally Maximum-minimum Bounds preserving) scheme. Through the simulation, the shock effect is determined in higher resolution and the retention times are calculated in higher resolution and the retention times are calculated in higher Precision. The comparison among results of theoretical analyse, experiments and the presend numerical simulation shows that the simulation results can betalcenas a fair approximation for nonlinear chromatography process.

Select

THE CALCULATIONS OF FREE-FREE GAUNT FACTOR IN HOT PLASMAS Sun Yongsheng, Yuan Jianhui, Meng Xujun, Zheng Shaotang 1995, 12(2): 179-184.  Abstract ( )   PDF (359KB) ( )   This paper is based on the relativistic self-consistent ion-sphere model. The densi-ty and the continuum wave function are determined consistently with the model. The free-free Gaunt factor is calculated by using partial wave phase-shift method, and the results are compared with the predictions of the Born-Elwert approximation.

Select

ON RESEARCHING OF THE INTEGRALS J(n,X) WITH SERIOUSLY OSCILLATORY WAVEFUNCTIONS IN ATOMIC PHYSICS (I) Shen Zhijun, Chen Guoxin, Fang Quanyu 1995, 12(2): 185-190.  Abstract ( )   PDF (342KB) ( )   In electron-atom (or ion) collisions, it is unadvoidable to compute a large number of integrals with seriously oscillatory wavefunctions. Since direct numerical integration are very exhausted and often may fail to reach the required accuracy, analytical methods are necessary. In this paper an improved method is presented for calculating integrals with seriously osillatory in tegrands. The accuracy of phase calculation is improved for the case of wavenumber being different. The analytical expressions of integrals are given for the case of wavenumber being equal. The accury of simulation results are very satisfactory.

Select

THE DIFFERENCE SCHEMES OF CONVECTIONAL EQUATION Zhang Tiande, Sun Chuanzhuo 1995, 12(2): 191-195.  Abstract ( )   PDF (255KB) ( )   New difference schemes of convective equation are constitnted by the method of undetermined coefficients.

Select

THE ENTRANCE OF THE BLOOD CELL INTO A THIN PIPE Sun Suohong, Chen Yaosong 1995, 12(2): 196-202.  Abstract ( )   PDF (438KB) ( )   The motion of the film separating two fluid phases is studied when it begin to be attracted into a thin pipe. Finite element method, especially the similar element is used. The elements attaching to the film are changed step by step in the calculation. The computed entrancing length variation is consistent with experiments. The computation offers the whole detail of the now.

Select

A HIGH-SPEED NEURAL NETWORKS ALGORITHM ON CLASSIFICATION-PROBLEM Cheng Guoxin, Zhang Chengfu 1995, 12(2): 203-206.  Abstract ( )   PDF (277KB) ( )   Artificial-Neural Network (ANN) is a complex network system composed of many simple elements connected extensively each other and can be regarded as a simulation or abstraction of biological neural system. Multi-Layer Network (MLN) is one of the most important ANN. A essential algorithm to train MLN is BP (Back-Propagation) algorithm. Based on BP algorithm, a hgih-speed algorithm-train MLN layer-by-layer algorithm is proposed with main points:4) train MLN layer-by-layer instead of all together; b) giving instruction to hidden units; c) giving appropriate "energy function' according to specific Problems; d) preserving the merits of BP algorithm. The simulation results increase by orders of magnitude in training-speed. This algorithm applies also some research on running mechanism of MLN.

Select

A TVD SCHEME WITH LESS NUMERICAL DISSIPATION Yang Dinghui, Wang Ruquan 1995, 12(2): 207-214.  Abstract ( )   PDF (442KB) ( )   In order to reduce the numerical viscosities of TVD schemes and improve their resolution, this paper suggests a new TVD scheme of Harten form with less numerical dissipation. Numerical experiments show that such a TVD scheme is better than Harten TVD scheme[1] and modified Harten scheme (Harten-Yee TVD scheme[2]) in simulating boundary layer and vortices.

Select

THE GRAND CANONICAL QUANTUM MONTE CARLO METHOD AND THE TWO-DIMENSIONAL HUBBARD MODEL Chen Feng, Ying Heping, Xu Tiefeng, Li Wenzhu 1995, 12(2): 215-218.  Abstract ( )   PDF (271KB) ( )   The work discussed algorithm of grand canonical quantum Monte Carlo method for the one-band two-dimensional Hubbard model, and calculated local magnetic moment, magnetic susceptibility, staggered magnetic susceptibility and the internal energy of the model. Results show that grand canonical quantum Monte Carlo method is very effective in dealing with Hubbard model from weak to strong correlation rigim, and also in dealing with many——body problems with strong correlation.

Select

THE VECTORIZABLE PE ITERATIVE METHODS Liu Xingping, Hu Jiagan 1995, 12(2): 219-226.  Abstract ( )   PDF (412KB) ( )   The algorithms of Vectorizable PE Method for linear systems of the form Ax=f are proposed, when A is block tridiagonal matrix. The convergence of these iterative methods is analysed, when A is an M matrix or H matrix. The resulting VPE method has been tested on YH-1 computer. Numericla examples indicate that the new method is very efficient, since the vectorial computation can be applied.

Select

Charge Transfer in Aq+ + H Collisions Qiu Yubo, Long Yanqiu 1995, 12(2): 227-233.  Abstract ( )   PDF (373KB) ( )   Charge transfer cross sections in Cq+ +H, Nq+ + HO q+ + H collisions are calculated with the CTMC method. The energy range of incident projectile is from 10keV/amu to 250keV/amu. The ions are partially stripped o rfully. Hamilton's equations of motion are given by the 12 coupled equations. Six pseudo-random parameters are required for the determination of the trajectories. A model potential between Aq+ ion and the active electron is used. The other two interactions are pure Coulombic. In our calculation, more than 2000 trajectories are required. Introducing reduced variables, σtr=σtr/q, E=E/q1/2, a q-scaled electron capture cross section curve is obtained, all cross section data are grouped around a single curve. Numerical results are compared with other theoretical calculations. It shows that the resuts are in reasonable agreement with other Ones.

Select

charge transfer ionization cross sections Monte Carlomethod classical trajectory Gao Yanfeng, Wang Ruihong, Zhang Xiaoze, Ling Yude 1995, 12(2): 234-240.  Abstract ( )   PDF (506KB) ( )   Some successful applications of MCNP code are introduced for the nuclear critical safety, techneque research for nuclear safeguards, and neutron slowing-down install simulation. Developments of this code for these applications are desovibed. A brief discussion is made about necessary development for future applications as well.

Select

USING DT METHOD TO CALCULATE COMBINED RADIATION-CONDUCTION HEAT TRANSFER IN THREE DIMENSIONAL CYLINDRICAL SEMI-TRANSPARENT MEDIUM Tan Heping, Xia Xinlin, Li Lin, Bao Yiling, and Yu Qizheng 1995, 12(2): 241-247.  Abstract ( )   PDF (469KB) ( )   Combined with spectral band model, control volume method and discrete transfer (DT) method are employed to investigate numerical calculation dealing with the transient combined radiation-conduction heat transfer in three dimensional cylindrical semi-transparent medium. The first and the third nonlinear boundary conditions are both considered. Compared with Zone method and improved Monte-Carlo approach, the calculated results by this method are reliable and more precise.

Select

COMPUTER SIMULATION STUDY OF THE PROPERTIES OF THREE PHOTON ANNIHILATION OF POSITRONS AND ELECTRONS Cheng Jinrong, Zhou Qin, Gao Yongchun, Yang Detian, and Liu Hong 1995, 12(2): 248-254.  Abstract ( )   PDF (399KB) ( )   This paper presents a method for sampling six random variables which indicate three emission angles of gamma rays and three gamma rays energies in the simulation test of three photon annihilation of positrons and electrons From the subsamples produced by the sampling mentioned above, frontier distribution and joint distribution about angles and about energies are obtained. Comparing the present results concerned with the known quantum electrodynamics (QED) prediction and experimental data, it is found that the three results are quite in agreement.

Select

AN OPEN BOUNDARY CONDITION FOR N.S. EQATIONS Chen Yaosong, Jiang Tao 1995, 12(2): 255-260.  Abstract ( )   PDF (373KB) ( )   The design of open boundary condition (OBC) is one of the most challenging problems in the field of nummerical computation for N. S. equation. In this article the perturbation solution of N.S. equation in the region approaching to infinity is taken as the basis for designing the OBC. A testing example proposed by M.philip has been calculated. The results show that the present OBC is superior to all others existing in literature.

Select

A SIMPLE CONFIGURATION-INTERACTION THEORY IN DIVALENT ATOMS Zeng Minyong, He Zhixian 1995, 12(2): 261-269.  Abstract ( )   PDF (520KB) ( )   A simple non-variational configuration-interaction theory is presented for the divalent atomic system with two electrons outside a closed shell core, such as an alkaline earth atom and its corresponding isoelectronic sequence. The theory is used for exact solution of the multiconfiguration state wave function and its corresponding energy eigenvalue in the system, and for quantitative analysis of configuration interaction which reflects the exchange and correlative interaction between electron-electron. Discussion focuses on the use of a nearly complete finite basis set construeted from B-spline which is capable of taking into account effectively the continum contribution from positive-energy one-electron orbitals confined in a finite volume. Numerical example is for Be atom of the divalent atomic system and the obtained result conforms excellently with experiment.

Select

INVESTIGATIONS OF ITERATIVE METHOD OF THE TWO-PHASE COUPLING IN NUMERICAL COMPUTATION OF GAS-SOLID FLOW AND COMBUSTION Yan Weiping, Xu Tongmo, Xu Jinyuan 1995, 12(2): 270-274.  Abstract ( )   PDF (367KB) ( )   Numerical computation to predict the pulverized-coal diffusion flame using Lagrangian treatment of the gas-solid flow and combustion is implemented. The authors analyse and discuss the gas-particle momemtum, mass and energy transfer processes. A modified scheme for existing computation model is proposed by the authors, which is effective in promoting two phase coupling process and can overcome divergency in computation for high pulverized-coal concentration combustion.

Select

ESTABLISHMENT OF EVENT GERENATOR FOR HIGH ENERGY PHYSICS PROCESS Zhang Ziping, Ye Jingbo 1995, 12(2): 275-278.  Abstract ( )   PDF (280KB) ( )   An event simulation generator from differential cross section formula is described certain high energy physics process, which.includes the integral of final state phase space, and the generation of final particles' four-momentum. For the commonly used stratified and importance sampling method, their individual characteristics, application domain, operating procedure and skills are discussed in detail, their advantages and disadvantages as well. Finally a Bhabha event generator established for LEP/SLC energy range is introduced.

Select

A PRACTICAL MONTE CARLO SAMPLING LIBRARY ON PC Li Weiping, Zhang Xiaoze, and Xu Shuyan 1995, 12(2): 279-288.  Abstract ( )   PDF (608KB) ( )   A Practical Monte Carlo Sampling Library is described. The library provides pesudorandom number generators using the best algorithms, excellent random variable generators of many important probability distributions such as spatial position, energy and motion direction distributions often used in particle transport problems. In addition a special test system for sampling library is designed and applied. The library runs on PC/386 (486). Its installation and usage are very easy. It can be applied in various fields and easy portable.