文件列表:

src (0, 2011-09-24)
src\bcfarf.f (5803, 2001-01-05)
src\bcinflo.f (3399, 2001-01-05)
src\bcond.f (3234, 2001-01-05)
src\bcoutflo.f (2967, 2001-01-05)
src\bnodes.f (4447, 2001-01-05)
src\checkmet.f (4080, 2001-01-05)
src\control.inc (1484, 2001-01-05)
src\conver.f (3299, 2001-01-05)
src\depvars.f (3785, 2001-01-05)
src\edgesfin.f (3168, 2001-01-05)
src\edgesini.f (3502, 2001-01-05)
src\files.inc (1537, 2001-01-05)
src\fluxc1.f (4087, 2001-01-05)
src\fluxc2.f (4984, 2001-01-05)
src\fluxroe1.f (4702, 2001-01-05)
src\fluxroe2.f (5383, 2001-01-05)
src\forces.f (3105, 2001-01-05)
src\fvecsymm.f (2628, 2001-01-05)
src\geometry.inc (1121, 2001-01-05)
src\gradient.f (7143, 2001-01-05)
src\initcon.f (2107, 2001-01-05)
src\initflow.f (3900, 2001-01-05)
src\initmet.f (4028, 2001-01-05)
src\initmetb.f (6081, 2001-01-05)
src\input_n (2982, 2001-01-05)
src\irsmoo.f (4650, 2001-01-05)
src\length.f (1635, 2001-01-05)
src\license (18349, 2001-01-05)
src\limiter.f (6520, 2001-01-05)
src\limitini.f (4378, 2001-01-05)
src\limitref.f (2458, 2001-01-05)
src\main.f (11760, 2001-01-05)
src\Makefile (1660, 2010-10-13)
src\massflo.f (2857, 2001-01-05)
src\n0012.ugr (557062, 2001-01-05)
src\numerics.inc (1448, 2001-01-05)
src\periodic.f (2162, 2001-01-05)
src\physics.inc (1517, 2001-01-05)
src\pltflow.f (5770, 2001-01-05)
... ...

******************************************************************************** E U L 2 D U (c) Jiri Blazek, September 1997 ******************************************************************************** Program for the solution of 2-D Euler equations on unstructured triangular grids. Median-dual finite-volume scheme is employed for the spatial discretisation. The equations are integrated in time using an explicit multistage time-stepping scheme. Further details are provided in Subsection 12.7 of the book. This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. -------------------------------------------------------------------------------- Files -------------------------------------------------------------------------------- LICENSE - terms and conditions Makefile - compilation options README - what you read now *.f - source files ("main.f" contains the main program) *.inc - include files The grid files stored in the directory "../run" are: channel.ugr - grid file for channel with 10% circular bump n0012.ugr - grid file for NACA 0012 airfoil vki1.ugr - grid file for VKI-1 transonic turbine cascade There are the following input and result files in the directory "../run": input_c - input file for transonic channel flow input_n - input file for transonic flow past NACA 0012 airfoil input_v - input file for flow past VKI-1 turbine cascade conver_c - convergence history for channel flow conver_n - convergence history for airfoil flow conver_v - convergence history for cascade flow iso_c - data for isoplots of channel flow iso_n - data for isoplots of airfoil flow iso_v - data for isoplots of cascade flow output_c - screen dump for channel flow output_n - screen dump for airfoil flow output_v - screen dump for cascade flow plot_c.ps - PostScript file with results for channel flow plot_n.ps - PostScript file with results for airfoil flow plot_v.ps - PostScript file with results for cascade flow surf_c - flow quantities at wall surface of the channel surf_n - flow quantities at wall surface of the airfoil surf_v - flow quantities at wall surface of the cascade The grids were generated by triangulation of the corresponding structured grids in the subdirectory "struct2d/run" using the program from "grid2du". The files "plot_", "surf_", "output_" and "conver_" were generated using the correspon- ding "input_" files. -------------------------------------------------------------------------------- How to compile "eul2du" -------------------------------------------------------------------------------- Edit first the flags (SRC, EXEC, O, LD, FC, FFLAGS, and LDFLAGS) in "Makefile" as appropriate for your computer, compiler and/or operating system. There are examples of settings for GNU g77, IBM, SGI and SUN f77 compilers. Type "make" at the prompt to compile the program. -------------------------------------------------------------------------------- How to run "eul2du" -------------------------------------------------------------------------------- The program reads a number of user parameters from the standard input. The user parameters are stored in a plain ASCII file. An example is provided in the file "input". In order to run "eul2du", you have to type at the prompt: % eul2du < [input file] where [input file] is the path and name of the file with parameters (e.g., "input_c"). The program generates four output files - convergence history, data for isoplots, distribution of flow quantities at wall(s) and a restart file (in binary format). In the case of external flows, the file with convergence history contains the quantities: step - iteration number resid - density based convergence measure (see Eq. (12.1)) resmax - maximum difference between new and old density [kg/m^3] i-res - index of the grid node where resmax is found cl - lift coefficient cd - drag coefficient cm - moment coefficient Lift, drag and moment coefficients are replaced by: m_flow - average mass flow rate [kg/s] m_ratio - ratio of mass flow at outlet to mass flow at inlet for internal flows. The file for the generation of isoplots always contains the grid coordinates (denoted as x, y). Additionally, the following quantities are written out for each grid point if specified in the input file: rho - density [kg/m^3] u - x-component of the velocity [m/s] v - y-component of the velocity [m/s] p - static pressure [Pa] p-tot - total pressure [Pa] T - static temperature [K] T-tot - total temperature [K] M - Mach number M-isen - isentropic Mach number pt-loss - total pressure loss The file with the surface distribution contains the same quantities. The variables in the convergence and in the plot files are stored column-wise. The file for the isoplots contains in addition the indices of the nodes of the triangles. Please note that all physical quantities are defined in SI units. In order to run "eul2du", a grid file must be provided. The grid file contains the dimensions of the grid, the description of the boundary faces, the Carte- sian coordinates of the nodes and the nodes of the triangles. The format of the grid file is as follows: (1) dimensions (one line): - number of grid nodes (NO dummy nodes) - number of triangular cells, - number of boundary conditions (2) for each boundary condition (two lines): - type 100-199 = inflow 200-299 = outflow 300-399 = viscous wall (treated here as inviscid wall) 400-499 = inviscid wall 500-599 = symmetry line 600-699 = farfield 700-799 = periodic boundary - index of the last boundary face (NOT used for periodic boundaries) - index of the last boundary point - name (on new line) (3) list of boundary faces (2 nodes of a face, arbitrary oriented) for all boundary conditions; if the corresponding boundary is periodic, the list contains the two related periodic nodes instead (4) Cartesian coordinates of the nodes (first column is x, second is y) (5) indices of three nodes defining a triangle; the orientation is arbitrary. Only ONE of the two related periodic boundaries has to be included in the boundary conditions (point 2 above). The nodes of the "shadow" periodic boundary are obtained from the list of boundary nodes (point 3). The program contained in "grid2du" provides a good example of how the grid file can be generated. -------------------------------------------------------------------------------- Main variables -------------------------------------------------------------------------------- Grid dimensions: ~~~~~~~~~~~~~~~~ mxnodes = max. number of grid nodes (including dummy nodes) mxcells = max. number of grid cells mxedges = max. number of edges (including edges between boundary and dummy nodes) nnodes = no. of grid nodes (including dummy nodes) nndint = no. of physical grid nodes (nnodes - dummy nodes) nedges = no. of edges (including edges between boundary and dummy nodes) nedint = no. of edges excluding those to dummy nodes Dummy nodes are defined for inlet, outlet and farfield boundaries only. The variables cv(), dv(), x(), y(), edge() and sij() are divided into two parts. The first part is related to the physical grid (dimensions nndint, nedint). The second part refers to the dummy nodes. This approach makes it easier to loop either over the physical grid nodes (edges) only or over all nodes (edges) using the same variable vector. Conservative & dependent variables: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ cv() = conservative variables: cv(in,1) = density cv(in,2) = density * u cv(in,3) = density * v cv(in,4) = density * E ... dv() = dependent variables: dv(in,1) = static pressure dv(in,2) = static temperature dv(in,3) = speed of sound dv(in,4) = ratio of specific heats dv(in,5) = specific heat coefficient at constant pressure ... mxcv = max. no. of conservative variables (4 + additional ones) mxdv = max. no. of dependent variables (5 + additional ones) nconv = number of conservative variables ndepv = number of dependent variables Geometrical quantities: ~~~~~~~~~~~~~~~~~~~~~~~ x() = x-coordinate of a grid point y() = y-coordinate of a grid point tria() = node indices of a triangle sij() = x-, y-component of the face vector (n*dS); sij(ie,_) represents for ie > nedint an average face vector for a boundary node (face between boundary and dummy node); sij() always points from node i to node j (see Fig. 5.9) vol() = control volume (shaded area in Fig. 5.8) cref = reference body length (airfoil chord) xref = x-coordinate of a reference point (used in "bcfarf.f", "forces.f") yref = y-coordinate of a reference point Other global variables: ~~~~~~~~~~~~~~~~~~~~~~~ cvold() = conservative variables from previous time step diss() = dissipation term of Roe's upwind scheme edge() = edge list (node i, node j); edge(ie,_) represents for ie > nedint an edge from a boundary node (i) to a dummy node (inlet, outlet, farfield) gradx() = gradient of density, u, v, and pressure wrp. to x-coordinate grady() = gradient of density, u, v, and pressure wrp. to y-coordinate lim() = value of the limiter function (density, u, v, pressure) rhs() = residual (right-hand side) tstep() = time step (without the CFL-number) Variables related to boundary conditions: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ mxbfaces = max. number of boundary faces mxbnodes = max. number of boundary nodes mxbounds = max. number of boundary conditions (set in "control.inc") nbfaces = no. of boundary faces nbnodes = no. of boundary nodes nbounds = no. of boundary conditions bface() = indices of two nodes defining a face (NOT for periodic boundaries) bname() = name of boundary condition (used for plot data) bnode() = boundary node related data: (_,1) = index of the node itself (_,2) = index of the related dummy node (inlet, outlet, farfield) or the second periodic node, otherwise = -99 (_,3) = index of edge to dummy node (edge(ie,_), ie > nedint); bnode() is set in "bnodes.f", "edgesfin.f", and "rbfaces.f" btype() = type of boundary condition: 100-199 = inflow 200-299 = outflow 300-399 = viscous wall (treated here as inviscid wall) 400-499 = inviscid wall 500-599 = symmetry line 600-699 = farfield 700-799 = periodic boundary ibound() = pointer from boundary condition to boundary faces and nodes: (_,1) = last index in bface() (_,2) = last index in bnode() sbf() = normal vector of boundary face (outward pointing, size equal to edge length); defined for all boundaries except periodic ones External flow: machinf = Mach-number at infinity alpha = angle of attack pinf = static pressure at infinity tinf = static temperature at infinity rhoinf = density at infinity uinf = u-component of velocity vector at infinity vinf = v-component of velocity vector at infinity qinf = total velocity (=SQRT(uinf**2+vinf**2)) Internal flow: ptinl = total pressure at inlet ttinl = total temperature at inlet betainl = flow angle at inlet (with x-axis, positive in clock-wise direction) pout = static pressure at outlet betaout = approximate outlet angle (needed for initial guess) p12rat = ratio of inlet to outlet static pressure (needed for initial guess) User input parameters & related quantities: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ if_grid, fn_grid = grid data if_gtop, fn_gtop = grid topology & boundary conditions if_flow, fn_flow = flow field if_surf, fn_surf = quantities along wall surface(s) if_conv, fn_conv = convergence history if_rsti, fn_rsti = restart solution - input if_rsto, fn_rsto = restart solution - output kflow = type of flow ("E"=external, "I"=internal) gamma = ratio of specific heat coefficients gam1 = gamma - 1 ggm1 = gamma/gam1 cpgas = specific heat coefficient at constant pressure rgas = gas constant maxiter = max. number of iterations iter = actual iteration count lrest = use previous solution for restart ("Y"=yes, "N"=no) convtol = convergence criterion (2-norm of density change for which the iteration process is stopped) cfl = CFL-number epsirs = coefficient of implicit residual smoothing nitirs = no. of Jacobi iterations (implicit residual smoothing) iorder = order of spatial discretisation (1 or 2) ktimst = switch between local (="L") / global (="G") time-stepping limfac = limiter coefficient (used only for 2nd-order Roe scheme, i.e., iorder=2) epsentr = entropy correction coefficient (Harten's correction, Eq. (4.92)) lvort = far field vortex correction ("Y"=yes, "N"=no) nrk = number of stages (Runge-Kutta) - max. 5 ark() = stage coefficients betrk() = dissipation blending coefficients ldiss() = dissipation evaluation (0=no, 1=yes) pi = 3.14... rad = 180./pi Generation of edge list: ~~~~~~~~~~~~~~~~~~~~~~~~ The edge list is generated in two steps. First, two temporary lists are generated in "edgesini.f". In a second step, the final edge list (edge()) is constructed, and edges to dummy points are included (subroutine "edgesfin.f"). The temporary lists are defined as follows: niedge(i) = pointer from node i to iedge() iedge() = linked list of edge endpoints: (_,1) = point j of edge (i,j) (_,2) = next point j which is also connected to i; if =0 - no further connection (_,3) = pointer to edge() - used in "initmet.f" to tie the face vector sij() to the correct edge Common blocks: ~~~~~~~~~~~~~~ "control.inc" /contr_c/ lrest,bname /contr_i/ maxiter,iter,nbounds,btype,ibound /contr_r/ convtol "files.inc" /files_c/ fn_grid,fn_gtop,fn_flow,fn_surf,fn_conv, fn_rsti,fn_rsto "geometry.inc" /geom_r/ xref,yref,cref "numerics.inc" /numer_c/ ktimst,lvort /numer_i/ nitirs,iorder,nrk,ldiss /numer_r/ cfl,epsirs,limfac,epsentr,ark,betrk,pi,rad, volref,limref "physics.inc" /phys_c/ kflow /phys_r/ gamma,cpgas, machinf,alpha,pinf,tinf,rhoinf,uinf,vinf,qinf, ptinl,ttinl,betainl,betaout,p12rat,pout "pltquant.inc" /pltq_c/ title,lquant,cquant /pltq_r/ drho,drho1,cl,cd,cm,mflow,mfrate Please note that all physical quantities are defined in SI units. -------------------------------------------------------------------------------- Program structure -------------------------------------------------------------------------------- Tree of subprogram calls: EUL2DU - main program (main.f) BCOND BC_FARF DEP_VARS_ONE FORCES BC_INFLOW DEP_VARS_ONE BC_OUTFLOW DEP_VARS_ONE BOUND_NODES CHECK_MET CONVER FORCES MASSFLOW DEP_VARS_ALL EDGES_FIN EDGES_INI FACEVEC_SYMM INIT_CONST INIT_FLOW INIT_MET INIT_METB PERIODIC LENGTH LIMITER_REF PLT_FLOW DEP_VARS_ALL LENGTH PLT_SURF DEP_VARS_ALL LENGTH READ_BFACES READ_DIMENS READ_GRID READ_PARAM READ_CHAR READ_SOLUT SOLVER BCOND (see above) DEP_VARS_ALL FLUXC1 FLUXC2 FLUX_ROE1 ENTROPY_CORR1 FLUX_ROE2 ENTROPY_CORR2 GRADIENT PERIODIC IRSMOO PERIODIC LIMITER VENKAT LIMITER_INIT PERIODIC TIME_STEP PERIODIC ZERO_RES WRITE_PARAM LENGTH WRITE_SOLUT Subroutines: BCOND - invokes application of boundary conditions (bcond.f) BC_FARF - sets farfield boundary conditions (bcfarf.f) BC_INFLOW - sets inflow boundary conditions (bcinflo.f) BC_OUTFLOW - sets outflow boundary conditions (bcoutflo.f) BOUND_NODES - stores indices of dummy nodes in bnode() (bnodes.f) CHECK_MET - checks metrics by computing min. and max. volume, and the sum of face vectors for each control volume (checkmet.f) CONVER - writes out convergence history (conver.f) DEP_VARS_ALL - values of dependent variables over whole domain (depvars.f) DEP_VARS_ONE - values of dependent variables at a point (depvars.f) EDGES_FIN - generates final edge list (edgesfin.f) EDGES_INI - generates temporary lists with nodes of an edge (edgesini.f) ENTROPY_CORR1- entropy correction for 1st-order scheme (fluxroe1.f) ENTROPY_CORR2- entropy correction for 2nd-order scheme (fluxroe2.f) FACEVEC_SYMM - corrects face vectors of edges which represent symmetry boundary (fvecsymm.f) FLUXC1 - calculates convective fluxes for 1st-order scheme (fluxc1.f) FLUXC2 - calculates convective fluxes for 2nd-order scheme (fluxc2.f) FLUX_ROE1 - upwind dissipation for 1st-order Roe scheme (fluxroe1.f) FLUX_ROE2 - upwind dissipation for 2nd-order Roe scheme (fluxroe2.f) FORCES - calculates pressure forces acting on the body (forces.f) GRADIENT - gradients of density, u, v, and pressure wrp. to x- and y-direction (gradient.f) INIT_CONST - initialises some important constants (initcon.f) INIT_FLOW - initialises flow variables and reference values (initflow.f) INIT_MET - calculates face vectors and control volumes (initmet.f) INIT_METB - duplicates control volumes at periodic boundaries and generates face vectors at boundaries (initmetb.f) IRSMOO - smoothes residuals (irsmoo.f) LENGTH - returns length of a string without trailing blanks (length.f) LIMITER - evaluates limiter functions (limiter.f) LIMITER_INIT - initialises limiter functions (limitini.f) LIMITER_REF - calculates reference values for the limiter (limitref.f) MASSFLOW - calculates mass flow at inlet and mass flow ratio (massflo.f) PERIODIC - adds variables at periodic nodes (periodic.f) PLT_FLOW - writes out selected quantities for the whole domain (pltflow.f) PLT_SURF - writes out selected quantities for all walls (pltsurf.f) READ_BFACES - reads in definition of boundary faces/periodic nodes (rbfaces.f) READ_CHAR - reads first ... ...

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：