scispace - formally typeset
Search or ask a question

Showing papers in "Chinese Journal of Computational Physics in 2015"


Journal Article
TL;DR: In this paper, the spectral element model with atmospheric near space resolved (SEMANS) is integrated for 10 years under configuration of 81 local elements in each projected face of cubed sphere with spectral approximation of 8th degree Gauss-Lobatto-Legendre interpolation polynomial and 66 vertical layers with top layer pressure 4.5 × 10-6 hPa.
Abstract: A numerical model,which solves equations with spectral element method and resolves atmospheric near space,is developed. Model validations are performed. The spectral element model with atmospheric near space resolved( SEMANS) is integrated for 10 years under configuration of 81 local elements in each projected face of cubed sphere with spectral approximation of 8th degree Gauss-Lobatto-Legendre interpolation polynomial and 66 vertical layers with top layer pressure 4. 5 × 10- 6 hPa. Through verification against ERA-Interim reanalysis dataset from ECMWF( European Center for Medium-Range Weather Forecasts) and COSPAR( Committee on Space Research) international reference atmosphere 1986,it is found that SEMANS reproduces features of 2 wave-lengths pattern along zonal circle in the northern hemisphere and 1 wave-length pattern along zonal circle in the southern hemisphere at 30hPa; SEMANS reproduces zones of low temperature at about 100 hPa,0.001 hPa and high temperature at about 1 hPa,above 0. 0001 hPa; SEMANS also reproduces main features of zonal mean zonal wind in January and July below 0. 001 hPa level.

2 citations


Journal Article
TL;DR: In this paper, a numerical model constrained by fine separated layer water injection was constructed and a linear system consisting of a seven-diagonal sparse coefficient matrix with attached columns and rows was derived, which was solved by partitioned matrix multiplication combined with preliminary treatment of conjugate gradient.
Abstract: Wellbore constrained equations were deduced considering separated layer water injection in a heavy oil model. A numerical model constrained by fine separated layer water injection was constructed. A linear system consisted of a seven-diagonal sparse coefficient matrix with attached columns and rows was derived,which was solved by partitioned matrix multiplication combined with preliminary treatment of conjugate gradient. Development effects of separated layer water injection and generalized water injection were compared and influences of segmentation and injection allocation models were assessed.

2 citations


Journal Article
Zhou Yunlon1
TL;DR: In this paper, a state feedback controller is proposed to control Hopf bifurcation in a fluid-flow model for congestion control in wireless networks, and it is proved that with the controller, one can change critical value of bifurbcation and enlarge stable region.
Abstract: Controlling Hopf bifurcation in a fluid-flow model for congestion control in wireless networks,we propose a state feedback controller. Choosing communication delay as a bifurcation parameter,we proved that with the controller,one can change critical value of bifurcation and enlarge stable region. Hopf bifurcation of the system is delayed. Moreover,by choosing appropriate parameters,we change easily stability and direction of bifurcating periodic solutions. Theoretical analysis and numerical results verifies that the method controls Hopf bifurcation effectively.

1 citations


Journal Article
TL;DR: A double-single precision algorithm implemented with single-precision floating-point numbers is employed to reduce numerical errors and shows that the improved algorithm sustained a highest GPU efficiency of 89.
Abstract: A boundary element method( BEM) for large-scale acoustic analysis is accelerated efficiently and precisely with Graphics Processing Units( GPUs) Based on Burton-Miller boundary integral equation,an implementation scheme that can be handled efficiently in GPU is derived and applied to accelerate conventional BEM Data caching techniques in GPU are introduced to improve efficiency of the prototype algorithm A double-single precision algorithm implemented with single-precision floating-point numbers is employed to reduce numerical errors It shows that the improved algorithm sustained a highest GPU efficiency of 89 8% for large-scale problems,and its accuracy was almost the same as that with double-precision numerals directly while costing only 1 /28 in time and half in GPU memory consumption of the latter The largest problem size up to 3 million unknowns was solved rapidly on a desktop PC( 8GB RAM,NVIDIA Ge Force 660 Ti) by the method Its performance was better than the fast BEM algorithms in both time and memory consumption

1 citations


Journal Article
Sun Chen-yan1
TL;DR: In this article, a velocity-stress-pressure formulation instead of velocity-vorticitypressure formulation was developed to solve unsteady laminar flow problems, and the result of stress formulation fits experiment better and has higher accuracy than vorticity formulation.
Abstract: To solve unsteady laminar flow problems,a method of velocity-stress-pressure formulation instead of velocity-vorticitypressure formulation is developed. With Newton's linearized method to linearize convective terms and preconditioned conjugate gradient method to solve equations,unsteady stress formulation of Navier-Stokes equations is solved. Comparison between numerical and experimental results of cavity laminar flow shows that result of stress formulation fits experiment better and has higher accuracy than vorticity formulation. The stress formulation can deal with subgrid stress with least squares finite element method. Comparison with experimental results of cavity turbulent flow reveals feasibility of the method. It lays a firm foundation for large eddy simulation computation.

1 citations


Journal Article
TL;DR: Prefomance testing shows that MRT model runs as fast as LBGK model on new generation of GPU cards and overall speedup reaches two orders of magnitude compared with serial code.
Abstract: A sparse lattice representation lattice Boltzmann method algorithm is implemented on Graphics Processing Units( GPU) to accelerate pore scale flow simuation. Prefomance testing shows that sparse lattice representation approach grately reduces memory requirement and maintains performance under low porosity compared with basic algorithm. Overall speedup reaches two orders of magnitude compared with serial code. Various factors including collision model,float number precision,and GPU that affect computing speed of the algorithm are invesgated independently. It indicates that MRT model runs as fast as LBGK model on new generation of GPU cards. While on old GPU cards,MRT model's computing speed matchs LBGK only when using single precision float.

1 citations


Journal Article
TL;DR: Development and validation of the programs used to simulate warhead verification technologies of passive neutron, passive γ ray, active neutron, active high-energy photon and delayed neutron methods are introduced, including theory and program flow.
Abstract: We set up a numerical experiment platform. It consolidates codes used to simulate warhead verification technologies of passive neutron,passive γ ray,active neutron,active high-energy photon and delayed neutron methods. Two functions on simulating time-dependent coincidence and neutron multiplicity counter measurements were added to the platform. They are carried out by DTB code and NMC code. This paper introduces development and validation of the programs,including theory and program flow. Two numerical experiments are designed to validate the program. The platform provides systematic data to support statistical analysis on nuclear warhead verification technologies

1 citations