Hot Search : Source embeded web remote control p2p game More...
Location : Home Search - h.3
Search - h.3 - List

 

#//u(i,j)        x方向的速度u;或者θ方向速度
#//u(i,j)        y方向的速度v;或者径向速度ur'h\K
#//pc(i,j)       压力修正 p'[OS
#//p(i,j)        压力p-=5-+
#//p(i,j)        密度ρBk
#//p(i,j)        扩散系数γQ"G48E
#//t(i,j)        温度Tn7/bKr
#//ake(i,j)        湍流脉动能量k971E
#//dis(i,j)      动能的耗散率ε4
 
//amut(i,j)     湍动扩散系数+
//gen(i,j)      湍流能量的生成率S
//f(i,jnf)     不同的φ变量M
//lsolve(nf)   1,求解变量f(i,j,nf)  8
//lprint(nf)   1,打印变量f(i,j,nf)   ~f
//lblk(nf)     1,对变量f(i,j,nf)应用块修正b5|{@U
//mode         选择坐标系的变量. 目前只支持mode=1.
mode=1 直角坐标系(x~y)BMI
mode=2 圆柱坐标系(r~z)^pg>)
mode=3 极坐标(r~θ)   Z^r[LE
solve子程序中求解变量f(i,j,nf)的重复扫描次数 b
变量fi,j,nf)的字符性标题 |L;
 
//xl    计算区域在x方向上的宽度)?
//yl    计算区域在y方向上的宽度@ob
//l1     x方向上主控制体的网格数。也是x方向上压力节点位置的最后一个i值。
//m1   y方向上主控制体的网格数。也是y方向上压力节点位置的最后一个j值。
//dt           时间步长 △t
//第三部分网格设定的变量
//x(i)        节点位置的x
//xu(i)       主控制容积在x方向的界面位置,即速度u(i,j)所在位置
//xdif(i)     差值 x(i)-x(i-1)
 //xcv(i)      主控制容积在x方向上的宽度
//xcvs(i)     速度u(i,j)的控制容积在x方向上的宽度
//y(j)        节点位置的y
//yv(j)       主控制容积在y方向的界面位置,即速度v(i,j)所在位置
//ydif(j)     差值 y(j)-x(j-1)
//ycv(j)      主控制容积在y方向上的宽度
//ycvs(j)     速度v(i,j)的控制容积在y方向上的宽度
////r(i,j)      主网络节点的半径r
////rmn(j)      在速度v(i,j)所在处的半径r之值
////sx (j)      主网格节点位置y(j)x方向上的标尺因子
//sxmn(j)     在界面位置yv(j)x方向的尺度因子
       //上面四个变量应用于非直角坐标系的情况
//xcvi(i,j)   xcv(i) 中与u(i,j)的控制容积相覆盖的部分
//xcvi(i,j)   xcv(i) 中与u(i+1,j)的控制容积相覆盖的部分
//ycvr(j)     主控制容积垂直于x方向的面的面积
//ycvrs(j)    速度v(i,j)的控制容积垂直于x方向的面的面积
//arx(j)      x方向相垂直的控制容积的面积
//arxj(j)     arx(j)中与速度v(i,j)的控制容积相覆盖的部分
//arxjp(j)    arx(j)中与速度v(i,j+1)的控制容积相覆盖的部分
//arxjarjxp实际上对应于x方向上的ycviycvip
 
//第四部分差分方程系数设定的变量
//con(i,j)    离散方程中的常数项b,在子程序gamsor中又作为存储sc的单元
//aip(i,j)    系数ae
//aim(i,j)    系数aw
//ajp(i,j)    系数an
//ajm(i,j)    系数as
//ap(i,j)     系数ap,在在子程序gamsor中又作为存储sp的单元
//flow         穿过控制容积界面的质量流率
//diff         扩散传导性 D
//acof         DIFLOW子程序计算的量,它给出了对流与扩散作用的联合影响
 
//第五部分求解差分方程过程中的变量求解
//du(i,j)     影响u(i,j)de
//dv(i,j)     影响v(i,j)dn
//pv(j)       用于计算主网络节点i,j上的质量流率的ρvr的插值因子:计算式如下: //fvp(j)      说明同上
//fx(i)       用于计算主控制容积界面(即速度u(i,j)所在处)的密度
 
//            rhom的插值因子,计算式如下:
 
//fxm(i)      说明同上
//pt(i)pt(j) tdma中的转换系数(消元过程中)l
//qt(i)qt(j) tdma中的转换系数(消元过程中)
//第六部分 index变量
int nf;               //nf           标明不同φ变量的下标值
int nfmax;         //nfmax        设有存储单元的nf的最大值
int np;              //(nfmax)      p(i,j)实际为f(i,j,nfmax)
int nrho;           //(nfmax+1)    rho(i,j)实际为f(i,j,nfmax)
int ngam;         //(nfmax+2)    gam(i,j)实际为f(i,j,nfmax+2)n
int l2;              //l2           (l1-1)
int l3;              //l3           (l1-2)
int m2;           //m2           (m1-1)
int m3;           //m3           (m1-2) 2
int ist;                    //ist          i的第一个内节点值`
int jst;                    //jst          j的第一个内节点值
int iter;                   //iter         不稳态问题的步进计数
int last;                   //last         用户所规定的最大迭代计数
int iter1;             //iter1        一个时间点求解setup2( )的迭代次数
double time;              //time         不稳态问题中的时间t
int ipref;              //ipref        压力参考结点的i
int jpref;                //jpref        压力参考结点的j=/
 
//第七部分其它变量
double rhocon;           //rhocon       密度为常数的问题中的ρ
int lstop;                 //lstop        =1时,停止计算
double  smax;            //smax         p'方程中的质源的最大值

double  ssum;            //ssum         p'方程中的质源的代数和


Date : 2008-12-03 Size : 11.17kb User : tanglincn

DL : 11
************************************************************************ * * * * * THIS IS THE H Y P L A S 2.0 README FILE * * ----------------- * * * * HYPLAS is a finite element program for implicit small and large * * strain analisys of hyperelastic and elasto-plastic two-dimensional * * and axisymmetric solids * * * * HYPLAS v2.0 is the companion software to the textbook: * * EA de Souza Neto, D Peric & DRJ Owen. Computational Methods for * * Plasticity: Theory and Applications. Wiley, Chichester, 2008. * * (www.wiley.com/go/desouzaneto) * * * * Copyright (c) 1998-2008 EA de Souza Neto, D Peric, D.R.J. Owen * *----------------------------------------------------------------------* * File last updated: 18 October 2008 * * * * This file belongs in the directory ../HYPLAS_v2.0 * ************************************************************************ * * * I M P O R T A N T * * * * READ SECTIONS 0 TO 3 OF THIS FILE CAREFULLY BEFORE ATTEMPTING * * TO COMPILE AND RUN THE PROGRAM HYPLAS ON YOUR COMPUTER !! * * * * THE AUTHORS DO NOT GUARANTEE THAT ANY SUGGESTIONS/INSTRUCTIONS * * GIVEN IN THIS README FILE WILL WORK ON ANY PARTICULAR OPERATING * * SYSTEM. IF YOU DECIDE TO FOLLOW ANY SUGGESTIONS/INSTRUCTIONS * * GIVEN HERE YOU MUST DO SO AT YOUR OWN RISK. * * * * * * BUG REPORTS: Please send bug reports to * * * * hyplas_v2.0@live.co.uk * * * * Messages sent to the authors' personal email addresses * * will NOT be answered. * ************************************************************************ This file contains the following sections: 0. Copyright statement and disclaimer 0.(a) Copyright statement 0.(b) Disclaimer 0.(c) Conditions of use 1. Introduction 1.(a) Note on portability 2. Compiling and running HYPLAS 2.(a) Memory requirements 2.(b) Testing a newly compiled executable 3. The HYPLAS directory tree 4. Cross-referencing between the source code and the textbook 5. HYPLAS error messaging 6. Further remarks on HYPLAS ************************************************************************ 0. COPYRIGHT STATEMENT AND DISCLAIMER ================================== 0.(a) Copyright statement ------------------- You may only use this program for your own private purposes. You are not allowed, in any circumstances, to distribute this program (including its source code, executable and any other files related to it, either in their original version or any modifications introduced by you, the authors or any other party) in whole or in part, either freely or otherwise, in any medium, without the prior written consent of the copyright holders. 0.(b) Disclaimer ---------- This program (including its source code, executable and any other files related to it) is provided "as is" without warranty of any kind, either expressed or implied, including, but not limited to, any implied warranties of fitness for purpose. In particular, THIS PROGRAM IS BY NO MEANS GUARANTEED TO BE FREE FROM ERRORS. This program (or any modification incorporated to it by you, the authors or any other party) will run entirely at your risk. The results produced by this program are in no way guaranteed to be fit for any purpose. Under no circumstances will the authors/copyright holders be liable to anyone for damages, including any general, special, incidental or consequential damages arising from the use or inability to use the program (including, but not limited to, loss or corruption of data, failure of the program to operate in any particular way as well as damages arising from the use of any results produced by the program for any purpose). 0.(c) Conditions of use ----------------- You may only use this program if you fully understand and agree with the terms of the above disclaimer. You must not use this program if you do not agree with or do not understand (fully or in part) these conditions of use. 1. INTRODUCTION ============ HYPLAS is a finite element code for small and large strain analysis of hyperelastic and elasto-plastic solids. Most procedures implemented in HYPLAS are described in detail in its companion textbook: EA de Souza Neto, D Peric & DRJ Owen. Computational Methods for Plasticity: Theory and Applications. Wiley, Chichester, 2008 (www.wiley.com/go/desouzaneto). 1.(a) Note on Portability ------------------- HYPLAS has been written in standard ANSI FORTRAN 77. Currently, the only known (and deliberate) exceptions to the FORTRAN 77 ANSI standard are the instructions: INCLUDE '' used in many routines to include the HYPLAS database files (common blocks and global variables), and; CALL GETENV('HYPLASHOME',HYPLASHOME) used in subroutine "ERRPRT" (file ../HYPLAS_v2.0/src/GENERAL/errprt.f). This instruction inquires the name of the system environment variable HYPLASHOME and writes it on the character string HYPLASHOME. This instruction is NOT part of the ANSI FORTRAN 77 standard, but seems to work in most currently available FORTRAN 77 compilers. 2. COMPILING AND RUNNING H Y P L A S ================================== The HYPLAS source code is stored in directory ../HYPLAS_v2.0/src/ (../HYPLAS_v2.0/ being the current directory) and all its subdirectories. To generate an executable file, you just need to compile the FORTRAN source files: ../HYPLAS_v2.0/src/hyplas.f and ../HYPLAS_v2.0/src/*/*.f together. We recommend that the executable HYPLAS be stored in the directory ../HYPLAS_v2.0/bin to which the environment variable HYPLASHOME should be set (see below how to set a system environmental variable). WINDOWS (R) systems ------------------- On Microsoft Windows(R) systems, HYPLAS has been successfully compiled using Intel Visual Fortran Compiler(R) integrated with Microsoft Visual Studio(R). Here you only need to create a project that contains all Fortran source files mentioned above as well as the include files ..\HYPLAS_v2.0\src\*.INC On a Windows XP system, the system environment variable HYPLASHOME can be set as follows: 1. Open a File Manager 2. Right-click on the "My Computer" icon 3. Select "Properties" on the drop-down menu 4. A new window named "System Properties" will pop-up. Here select the "Advanced" tab. 5. On the "Advanced" tab, click the "Environment Variables" button. 6. A new window titled "Environment Variables" will pop-up. Here click the button "New" in the "System Variables" section of the window. 7. A new window will pop-up titled "New System Variable". Here you should fill the fields "Variable name" and "Variable Value", respectively, with HYPLASHOME and the path name (in full) of the directory ..\HYPLAS_v2.0\bin. 8. Press "OK" on the relevant pop-up windows. 9. The next time the computer is REBOOTED, this variable will be set to the correct path and HYPLAS should be able to find the error messages file ERROR.RUN if required. UNIX/LINUX systems ------------------ In a UNIX/LINUX operating system using a C-shell, for instance, the HYPLASHOME environment variable should be set with the command: setenv HYPLASHOME where here denotes the full path to the directory ../HYPLAS_v2.0/bin. To compile HYPLAS (from directory ../HYPLAS_v2.0/src) with a FORTRAN 77 compiler such as g77, you can use the command: g77 -o ../bin/hyplas hyplas.f */*.f Note that the executable file "hyplas" will be stored in the directory ../HYPLAS_2.0/bin (i.e. the directory set in the HYPLASHOME environment variable). Alternatively, you may use the Makefile provided (with suitable modifications, if needed) to create the HYPLAS executable. IMPORTANT: Before generating a HYPLAS executable, read Sections 2.(a) and 2.(b) below. 2.(a) Memory Requirements ------------------- HYPLAS memory requirements depend on the array dimensioning parameters set in files: ../HYPLAS_v2.0/src/ ELEMENTS.INC GLBDBASE.INC MATERIAL.INC MAXDIM.INC Files ELEMENTS.INC, GLBDBASE.INC and MATERIAL.INC contain parameters which are associated with the currently implemented finite elements and materials. DO NOT MODIFY THEM ! unless you are absolutely sure of what you are doing (only developers coding new elements or new material models/analysis types may need to modify them by changing the existing dimensioning parameters and/or including new parameters). The ONLY dimensioning file that can be safely modified by the average user is the file MAXDIM.INC This file contains the array dimensioning parameters related to the maximum permissible dimension of problems to be analysed by HYPLAS. These parameters include the maximum number of nodes, elements, element groups, etc. If necessary, CHANGE THESE PARAMETERS TO SUIT YOUR PROBLEM SIZE/MEMORY REQUIREMENTS before compiling HYPLAS. 2.(b) Testing a newly compiled executable ----------------------------------- After you have successfully compiled the HYPLAS source code and created an executable file, the next step is to run some tests to verify that HYPLAS is working well. To do this, proceed as follows: The directory ../HYPLAS_v2.0/book_examples/data_files contains a series of data files named .dat of benchmarked examples described in the companion textbook. The corresponding (benchmarked) result files are in the directory ../HYPLAS_v2.0/book_examples/result_files This directory contains a series of result files named .res generated with the current version of HYPLAS on a tested platform. All these files have been named such that their names start with the textbook section number where the corresponding example is described. For instance, files 14_9_2_tresca.dat and 14_9_2_tresca.res refer to a problem described in section 14.9.2 of the textbook, and so on. To check that HYPLAS is working well on your platform, after compiling HYPLAS, run the program HYPLAS for the examples of files .dat and compare the newly generated results .res with their benchmarked counterparts (of the same filename) in the result_files directory. To run an example, execute HYPLAS and use the keyboard to enter the name of the corresponding data file in full (including the extension .dat). To compare the benchmarked .res files against their newly generated you may proceed as follows: 1. On MICROSOFT WINDOWS systems - Here we have successfully used the software "ExamDiff" (the task was made particularly easy by selecting "View" and then the "Show Differences Only" option - this refers to version 1.8 of this software). 2. On UNIX/LINUX systems - Here we use the "diff" command from a shell window (and set the option to ignore blank spaces). A shell script may be used to perform this task automatically (including running HYPLAS and checking for result file differences) for all benchmarked examples provided. IMPORTANT: THE ONLY ACCEPTABLE DIFFERENCES BETWEEN A THE NEWLY GENERATED RESULT FILES AND THEIR BENCHMARKED COUNTERPARTS ARE THE DIMENSIONING PARAMETERS (FROM FILE MAXDIM.INC) USED TO COMPILE THE NEW EXECUTABLE (THESE PARAMETERS ARE PRINTED RIGHT AT THE BEGINNING OF THE RESULT FILES) AND NUMERICAL DIFFERENCES IN RESULTS DUE TO NUMERICAL "ROUNDING-OFF" (THESE ARE VERY SMALL DIFFERENCES THAT DEPEND ON THE PRECISION OF ARITHMETIC OPERATIONS IN THE PLATFORM USED). ALSO NOTE THAT THE EXAMPLES OF THE COMPANION TEXTBOOK DO NOT COVER ALL FEATURES OF HYPLAS. HENCE THIS TEST DOES NOT GUARANTEE THAT EVERYTHING IS WORKING PROPERLY. 3. THE H Y P L A S DIRECTORY TREE ================================ 3.(a) Summary ------- ../ HYPLAS_v2.0/ bin/ book_examples/ data_files/ result_files/ man/ html/ src/ CRYSTAL/ DAMAGE/ DAMAGED_ELASTIC/ DRUCKER_PRAGER/ ELASTIC/ ELEMENTS/ GENERAL/ MATERIALS/ MATHS/ MOHR_COULOMB/ OGDEN/ TRESCA/ VON_MISES/ VON_MISES_MIXED/ 3.(b) Description ----------- The HYPLAS program directory tree is organised as follows: ../HYPLAS_v2.0/ (this directory) This is the HYPLAS root directory, where the HYPLAS directory tree starts. ../HYPLAS_v2.0/bin/ This directory contains the file ERROR.RUN where most HYPLAS error/warning messages are. IMPORTANT: the environment variable HYPLASHOME should be set to this directory. Otherwise, HYPLAS will not find its error/warning messages when required. We also recommend that the EXECUTABLE of HYPLAS be stored in this directory. ../HYPLAS_v2.0/book_examples/ This directory has the following subdirectories: ../HYPLAS_v2.0/book_examples/data_files ../HYPLAS_v2.0/book_examples/result_files Refer to Section 2.(b) above for further details. ../HYPLAS_v2.0/man/ This is the HYPLAS documentation/manuals directory. It contains the following files: input_man.txt - A concise input data manual for HYPLAS in ASCII format; hyplas_calltree.txt - Contains a flowgraph (shows the call tree) of HYPLAS in ASCII-format. Note: calls to function subprograms are not included in this flowgraph; and the subdirectory: ../HYPLAS_v2.0/man/html This directory contains the hypertext (HTML) format Fortran source code and of manual pages of the entire HYPLAS program. Manual pages with descriptions of each function/subprogram including their argument list are linked to their corresponding HTML-format source code. This allows the user the navigate through the HYPLAS source code using a web browser. To start at the main program, use your web browser to open the file hyplas.html. This facility should be helpful to those trying to understand the flow of program HYPLAS. ../HYPLAS_v2.0/src/ This directory (and its subdirectories) contains the Fortran source code of HYPLAS. The files containing the sources are named following the standard practice: .f where is the name of the FORTRAN procedure (subroutine, function subprogram, etc.) whose source code is in file .f. The source code of the HYPLAS main program is in file hyplas.f and the HYPLAS database (COMMON blocks, array dimensioning parameters and other global parameters) is coded in the "include files" ELEMENTS.INC GLDBASE.INC MATERIAL.INC MAXDIM.INC in this directory. In addition, this directory contains a file named "Makefile" (UNIX-LINUX Release only) which may be used for compiling and linking HYPLAS in UNIX/LINUX systems. The subdirectories of ../HYPLAS_v2.0/src are as follows: ../HYPLAS_v2.0/src/CRYSTAL Contains the source code of all procedures related to the finite strain single crystal plasticity model implemented in HYPLAS. ../HYPLAS_v2.0/src/DAMAGE Source files of the procedures related to the Lemaitre ductile damage model implementation. ../HYPLAS_v2.0/src/DAMAGED_ELASTIC Source files of the procedures related to the damaged elasticity model with crack closure effect. ../HYPLAS_v2.0/src/DRUCKER_PRAGER Source files of the procedures related to the implemented Drucker-Prager plasticity model. ../HYPLAS_v2.0/src/ELASTIC Source files of the procedures related to the linear elasticity model (Hencky model under large strains) implemented. ../HYPLAS_v2.0/src/ELEMENTS Source files of the element interfaces and element-related procedures. ../HYPLAS_v2.0/src/GENERAL Source files of general procedures. ../HYPLAS_v2.0/src/MATERIALS Source files of the material interfaces. ../HYPLAS_v2.0/src/MATHS Source files of the mathematical procedures. ../HYPLAS_v2.0/src/MOHR_COULOMB Source files of the procedures related to the implemented Mohr-Coulomb plasticity model. ../HYPLAS_v2.0/src/OGDEN Source files of the procedures related to the implemented Ogden hyperelasticity model. ../HYPLAS_v2.0/src/TRESCA Source files of the procedures related to the implemented Tresca plasticity model. ../HYPLAS_v2.0/src/VON_MISES Source files of the procedures related to the implemented von Mises plasticity model with isotropic hardening. ../HYPLAS_v2.0/src/VON_MISES_MIXED Source files of the procedures related to the implemented von Mises plasticity model with mixed isotropic/kinematic hardening. 4. CROSS-REFERENCING BETWEEN THE SOURCE CODE AND THE TEXTBOOK ========================================================== Many references are made in the textbook to various subprograms of HYPLAS. These are usually made when a particular procedure described in the text is implemented in the program. The reader should refer to the textbook index. Also, a substantial number of comment lines have been added to the source code of HYPLAS with reference to sections, figures, boxes, etc of the textbook related to the part of the code in question. Such references are usually displayed after the word "REFERENCE:" (in capitals) on commented lines. Searching for this word will take you to the line of code where the particular routine has a reference to the textbook. NOTE: Occasional references to other textbooks/journal papers are also made following the word "REFERENCE:" on commented lines. 5. HYPLAS ERROR MESSAGING ====================== Most error/warning messages issued by HYPLAS are in the ASCII-format file ERROR.RUN (kept in the HYPLASHOME directory - ../HYPLAS_v2.0/bin). All such error/warning messages have an identification code (e.g. ED0015) which is printed both to the standard output (this is usually the computer screen) and to the relevant results file. If you wish to find where in the source code a particular message is being issued, then perform a search for the corresponding message identification code in the entire source code of HYPLAS. 6. FURTHER REMARKS ON HYPLAS ========================= 6.(a) Program efficiency THIS SECTION IS OF INTEREST ONLY TO THOSE WANTING TO MAKE HYPLAS RUN FASTER. It is particularly stressed in the textbook that this program has not been designed having efficiency in mind (refer to Section 5.1.2 of the textbook). Its structure has been designed mainly to illustrate in a relatively clear manner the computer implementation of the techniques and algorithms described in the text, with a particular view to the implementation of solid constitutive models and finite elements. For those who are especially interested in the speed of the code, there are a few tips that could help in this direction. Unfortunately, these involve modifications to the source code which is probably most appropriate to readers with a good level of experience in finite element programming. To those with this particular interest, we can suggest the following: (i) The use of faster linear solvers This is probably the change that would result in a greater gain in efficiency. The Frontal Method adopted in subroutine FRONT (file ../HYPLAS_v2.0/src/GENERAL/front.f) has been designed originally to save memory (back in the days when computer memory was severely limited). There are currently a vast number of methodologies which focus on speeding up the linear solution, in addition to reducing memory storage requirements (which is a particularly important issue in the solution of large scale problems). Some of these are extensions/refinements of the original Frontal solver. We remark that a number of such procedures (with their respective source codes) are available (conditions may apply) from the LAPACK (Linear Algebra PACKage - http://www.netlib.org/lapack) repository or from the HSL Library (http://www.cse.cse.scitech.ac.uk/nag/hsl). For the reader interested in gaining speed, we would recommend the replacement of the existing solver of FRONT by a faster one. We remark though that this is a substantial programming task. Another aspect here is the fact that computing times in FRONT are directly linked to the frontwidth of the system which, in the present version of HYPLAS is fixed and depends, for a given mesh, on how the degrees of freedom are numbered (node numbering). The incorporation of a frontwidth optimiser (which re-numbers the degrees of freedom in order to minimise the frontwidth) in FRONT could produce some good savings in computing times. Such savings become particularly noticeable in larger problems where the original node numbering produces an excessively large frontwidth. (ii) Material-specific computations The issues pointed out here affect only the computing times for specific material models and are expected to have a much lower impact in overall speed than the linear solver issue discussed above. Some of the material model-specific computations carried out in HYPLAS could be made a bit faster. For example, for isotropic models whose stress update is carried out in the principal stress space (such as the Tresca and Mohr-Coulomb models - see routines SUTR and SUMC, files ../HYPLAS_v2.0/src/TRESCA/sutr.f and ../HYPLAS_v2.0/MOHR_COULOMB/sumc.f, respectively) the spectral decomposition of the stress in carried out in the state update update routine and then repeated in the corresponding routine for computation of the consistent tangent operator (refer to files ../HYPLAS_v2.0/src/TRESCA/cttr.f and ../HYPLAS_v2.0/src/MOHR_COULOMB/ctmc.f, respectively, for the Tresca and Mohr-Coulomb plasticity models). Some savings in computing time can be achieved here by storing the stress eigenprojection tensors (these can be stored as state variables) during the execution of the state updating and then retrieving them later for use in the computation of the consistent tangent operator. This change can be incorporated to the code relatively easily. The computation of the exponential map and is derivative for the single crystal plasticity model (routines EXPMAP, file ../HYPLAS_v2.0/src/CRYSTAL/expmap.f and DEXPMP, file ../HYPLAS_v2.0/src/CRYSTAL/dexpmp.f) is carried out in three dimensions (these routines have been adapted from an earlier three-dimensional code). To improve efficiency, these can be adapted to work only in two-dimensional problems by removing the unnecessary operations related to the third dimension. 6.(b) Output of nodal averaged values The reader should be aware that the way in which nodal averaged values of stresses and other variables are calculated in HYPLAS is very basic (and rudimentary). This feature of the program is made available only to help those interested in producing contour plots, etc from results presented in HYPLAS result files and should be useful in many circumstances of interest. This facility has in fact been used in producing many of the figures presented in the textbook. But note, for example, that the values of incremental plastic multipliers for plasticity models may take (inadmissible) negative values when extrapolated from Gauss-point to nodes and averaged. We remark that more sophisticated and refined techniques of transferring Gauss point values of variables to nodal points and obtaining the corresponding smoothed field are available in the current literature. These fall outside the scope of the companion textbook of HYPLAS.
Date : 2011-07-29 Size : 10.5mb User : gtcewli3

数字带通滤波器的设计过程。共3个源程序。其中,dos.cpp为带通FIR的源程序,其中的滤波器阶数n,滤波器的低频fln,滤波器的高频fhn,窗函数w,可以按照自己的要求改变。这里采用的是汉宁窗。主函数main.cpp实现输入与fir的h(n)的卷积,输出结果Y(n),并将其存储在y.txt中。最后,graphics.cpp从y.txt读取数据,再由语句画出波形,可以清楚地看出计算结果是否正确,并进行比较。-digital band-pass filter design process. A total of three sources. Among them, dos.cpp FIR bandpass to the source, the filter order n, the low-frequency filter FLN, the high-frequency filter fhn, window function w, in accordance with the requirements change. Here is the Hanning window. The main function main.cpp achieve the input and fir h (n) convolution, the output Y (n), and store them in y.txt China. Finally, graphics.cpp y.txt read data from and then draw waveform statement, we can clearly see whether calculations are correct, and comparisons.
Date : 2025-12-16 Size : 2kb User : feide

该文件为c++的数学函数库!是一个非常有用的编程工具.它含有各种数学函数,为科学计算、工程应用等程序编写提供方便!-the document to the c Math Library! It is a very useful programming tools. It contains a variety of mathematical functions, for science, engineering application procedures for the preparation of convenience!
Date : 2025-12-16 Size : 7.62mb User : 唐山

理查森外推程序输入编号(1,2,3,分别表示lnx,tanx,sin(x^2+1/3x)计算函数)以及M和h的值,输出给出例6中的表格-Richardson extrapolation process input number (1,2,3, respectively, express lnx, tanx, sin (x ^ 2+ 1/3x) calculation function), as well as the value of M and h, the output is given in the table in Example 6
Date : 2025-12-16 Size : 907kb User : sara

DL : 0
(1)Msls分三步对系统和噪声模型进行辨识,采用脉冲序列作为辅助系统模型,用 计算输出数据 ;用原输出数据 计算 ,用递推最小二乘方法分别对系统参数和模型参数进行估计。 (2)M.dat,wnoise1.dat分别为M和白噪声序列。Wnoise1.dat的长度为700,wnoise2.dat的长度为1000。Msls6.c为N=600的程序,Msls8.c为N=800的程序。 (3)程序运行后,生成的两个h文件为产生的脉冲响应函数。Msls6.dat为msls6.c的参数估计结果,msls8.dat为msls8.c的参数辨识结果。分别如下所示: a1=0.906331 a2=0.160170 a3=0.025525 b1=0.704475 b2=-1.497551 c1=1.009114 c2=0.446890 a1=0.906347 a2=0.159066 a3=0.024650 b1=0.700720 b2=-1.493327 c1=1.008787 c2=0.425714 (4)由数据结果可以看出,采用msls辨识方法估计精度要比els法的估计精度差一些。尤其是噪声参数c2的估计误差不在1%以内。这是由于msls法计算上较为简便,计算上的简化就带来了估计精度上的误差。由N=600和N=800相比较,可以看出当N增大时,误差有所减小。理论上当N趋于无穷时, 。 -err
Date : 2025-12-16 Size : 270kb User : 陈栋

DL : 0
Algo2-3.c C program for implementing Algorithm 2.3 Algorithm translated to C by: Dr. Norman Fahrer IBM and Macintosh verification by: Daniel Mathews NUMERICAL METHODS: C Programs, (c) John H. Mathews 1995 To accompany the text: NUMERICAL METHODS for Mathematics, Science and Engineering, 2nd Ed, 1992 Prentice Hall, Englewood Cliffs, New Jersey, 07632, U.S.A. Prentice Hall, Inc. USA, Canada, Mexico ISBN 0-13-624990-6 Prentice Hall, International Editions: ISBN 0-13-625047-5 This free software is compliments of the author. E-mail address: in "mathews@fullerton.edu" Algorithm 2.3 (False position or Regula Falsi Method). Section 2.2, Bracketing Methods for Locating a Root, Page 62-Algo2-3.c C program for implementing Algorithm 2.3 Algorithm translated to C by: Dr. Norman Fahrer IBM and Macintosh verification by: Daniel Mathews NUMERICAL METHODS: C Programs, (c) John H. Mathews 1995 To accompany the text: NUMERICAL METHODS for Mathematics, Science and Engineering, 2nd Ed, 1992 Prentice Hall, Englewood Cliffs, New Jersey, 07632, U.S.A. Prentice Hall, Inc. USA, Canada, Mexico ISBN 0-13-624990-6 Prentice Hall, International Editions: ISBN 0-13-625047-5 This free software is compliments of the author. E-mail address: in "mathews@fullerton.edu" Algorithm 2.3 (False position or Regula Falsi Method). Section 2.2, Bracketing Methods for Locating a Root, Page 62
Date : 2025-12-16 Size : 2kb User : virumpa

DL : 0
(1)已知甲数的35 是36,甲数是( )。已知乙数的47 是12,乙数是( )。   (2)40分钟是1小时的( ),1小时的35 是( )分钟。 (3)铅笔的单价是钢笔的920 ,把( )看作单位“1”。 二、根据算式补充条件   (1)停车场有24辆大汽车,( ),有多少辆小汽车?     24÷23   (2)停车场有24辆大汽车,( ),有多少辆小汽车?     24×23   (3)停车场有24辆小汽车,( ),有多少辆大汽车?     24×13 三、应用题   (1)一个长方体的宽是长的23 ,长是高的56 ,它的宽是10厘米,它的高是多少厘米?   (2)六年级(2)班学生人数是本年的学生人数的13 ,六年级学生人数是全校的16 ,如果六(2)班有学生48人,全校有学生多少人? 参考答案 一、 (1)60 ,21 (2)23 ,36 (3)钢笔的单价 二、 (1)正好是小汽车的23 -Th e i nf o r ma t i o n o f e n v i r on me n t c o n s t r a i n s a nd pa t h l e n g t h Wa s i n t e g r a t e d i n t he f it n e s s f un c t i o n whi c h Wa s c o mp ut e d by n e u r a l n e t wo r k,t h e pa t h no de s wa s v i e we d a s a p a r t i c l e,8 0 wi t h t h e qu a l i t y o f o p t i mi z a t i o n o f h y br id p a r t i c l e s wa r m a lg o . r i t h m,a b e s t p a t h W a s f o u n d .F i n a ll y b y c o mp u t e r s i mu l a t i o n,i t i s p r ov e d t h a t t h e a l g o r i t h m i s r a t i o n a l a n d c a n b e u s e d i n mo — bi l e r ob o t r e a l— t i me n a v i g a t i o n
Date : 2025-12-16 Size : 4kb User : wanchao

DL : 0
The Levenshtein Distance between two strings a, b is the number of changes to transform one string into another. For example to transform mouse into house would take 1 (change the m to an h). If the strings are different length then either characters need to be added or removed. The total number of operations needed (inserts, deletes or changes) is the Levenshtein distance.
Date : 2025-12-16 Size : 4kb User : ed

可以处理三维矩阵数据,三维数组的线性化处理,-#include<stdio.h> #define N 56 #define M 56 #define L 336 #define Leng 23.693 double V[N][M][L] void main() { int i, j, k double Z[L], ZL[L], S, A FILE*fp1,*fp2 if ((fp1 = fopen("LOCPOT","r+"))==NULL) printf("ERRO,can not open the file1") else for(k=0 k<=L-1 k++) for(j=0 j<=M-1 j++) for (i=0 i<=N-1 i++) fscanf(fp1," lf",&V[i][j][k]) fclose(fp1)
Date : 2025-12-16 Size : 1kb User : 王思栋

设计简易交互式计算器,功能如下: (1) 计算器功能与Windows操作系统带的标准型的计算器系统相同。至少有+、-、*、/、开方、取倒数、清零、backspace的功能,可以对小数进行运算。 (2) 程序可以实现简单的没有括号的表达式计算, (3) 不使用math.h头文件,求开方运算时使用多项式拟合方式。 -Design simple interactive calculator, function as follows: (1) calculator function and Windows operating system with the standard system of the same type of calculator. At least+,-,*,/, prescription, take the countdown, clear, backspace function, can operate on decimal. (2) The program can not simply calculate the expression in brackets, (3) does not use the math.h header file, find the root operation means the use of polynomial fitting.
Date : 2025-12-16 Size : 2.59mb User : onway

其基本思想是取B样条函数 为及基函数,在[a,b]中均匀划分区间,xi=a+ih,h=(b-a)/n,由于三次样条函数空间是n+3维的,所以把分点扩充到x-1,xn+1,则任意三次样条函数可用 线性组合来表示:-The basic idea is to take B-spline function and basis functions in [a, b] in the evenly divided intervals, xi = a+ ih, h = (ba)/n, because the cubic spline function is n+3 dimensional space , so the expansion points to the x-1, xn+1, then any cubic spline function can be expressed as a linear combination:
Date : 2025-12-16 Size : 6kb User : 咪噜

DL : 0
凸轮轮廓计算: 3#号凸轮:正弦加速度运动规律 推程h=20mm; 回程h=20mm; 凸轮基圆半径r=50mm; 推程运动角φ=180°; 回程运动角φ=180°; -Cam profile calculation: the# 3 cam: sine acceleration movement of push process h = 20mm return h = 20mm cam base circle radius r = 50mm push the process movement angle φ = 180 ° return movement angle φ = 180 °
Date : 2025-12-16 Size : 216kb User : feng

DL : 0
纯C语言编写的常见函数(冲激、阶跃、三角、指数、自定义函数,均可设置自定义系数)的FFT运算并画图显示。运行需要graphics.h。TURBO C++ 3.0运行通过。压缩包中包括一些C语言图形界面的简单教程。-A C program that can calculate common functions FFT and show the graphics. It is run successfully in Turbo C++ 3.0 with graphics.h.
Date : 2025-12-16 Size : 554kb User : 木偶

DL : 0
sgu250:Constructive Plan 题目大意: 给出一个n∗ m的01矩阵,0表示不能放,1表示能放,在其中放入三个矩形,要求满足如下条件: 1.每个矩形面积大于0。 2.这些矩形必须是一个联通块,矩形之间不能重叠。 3.矩形的左边界在同一条线上。 4.中间矩形的横向长度小于两边矩形的横向长度。 求出最大的三个矩形的总面积,无解输出− 1。-250. Constructive Plan time limit per test: 0.25 sec. memory limit per test: 65536 KB input: standard output: standard Oh, no! - Petya said, walking around his recently bought ground plot. Petya wants to build a new house on it. According to Petya s building project the house should look above like C character. There are many trees growing on Petya s plot. But everyone who cuts down a tree in Petya s country is sent to cut down trees for the rest of his life. So first of all Petya has to choose a place for building the house without cutting any tree. He is feeling that he is not able to find the solution on his own, so he decided to ask you to help him. The task is simplified a little by the fact that Petya s plot has a rectangular shape of size N*M, divided into 1*1 square cells. For each cell it is known whether there are any trees growing there. House can t occupy cells where trees grow. Fortunately Petya could explain how his house must look above. 1) H
Date : 2025-12-16 Size : 1kb User : owaski

该函数是将三维核函数拆解为三个一维核函数,尽管只有部分三维核函数可以被准确拆解(如高斯核函数),但该函数对于不能准确拆解的核函数仍可以采用最小二乘的办法拆解该函数,对于图像滤波,将三维核函数拆解为三个一维核函数的做法将在一定程度上提高滤波效率。 输入: H [h(1,1,1), h(1,2,1), h(1,3,1)] (第1层) [h(2,1,1), h(2,2,1), h(2,3,1)] [h(3,1,1), h(3,2,1), h(3,3,1)] [h(1,1,2), h(1,2,2), h(1,3,2)] (第2层) [h(2,1,2), h(2,2,2), h(2,3,2)] [h(3,1,2), h(3,2,2), h(3,3,2)] [h(1,1,3), h(1,2,3), h(1,3,3)] (第3层) [h(2,1,3), h(2,2,3), h(2,3,3)] [h(3,1,3), h(3,2,3), h(3,3,3)] 输出: a [a(1), a(2), a(3)] b [b(1), b(2), b(3)] c [c(1), c(2), c(3)] 这里,h(i, j, k) a(i) * b(j) * c(k) Matlab代码见:SeparateKernel.m。 -it is to make a 3 dimentional function be expressed as three 1 dimntional funtions
Date : 2025-12-16 Size : 23kb User : lynn

这个c++程序用于求解形如 -(pu')'+qu=f;u=0于左边界;u'=alpha于右边界。 的椭圆型偏微分方程。采用有限元解法,二次元差分格式。 使用说明: 在corefunc.h中自己定义修改方程中的函数p,q,f,它们分别对应文件中的3个函数funcp,funcq,funcf。 核心函数是double *Calcu(double *x, int n,double alpha=0.0) 其中x是节点数组,n+1应等于节点数量。alpha是右边值条件。核心函数返回计算的 近似解u在节点及对偶节点上的函数值。(A function for finite element experiment.)
Date : 2025-12-16 Size : 4kb User : random海
CodeBus is one of the largest source code repositories on the Internet!
Contact us :
1999-2046 CodeBus All Rights Reserved.