Description: 6。《非数值并行算法——遗传算法》，刘勇等着 科学出版社 1995年第一版
本书系统地叙述了非数值并行算法之一的模拟退火算法的基本原理以及最新进展，同时为了便于读者解决实际问题，书中对具体算法的步骤作了详细介绍。本书共分七章，第一章介绍算法的思想、特点。发展过程和前景。第二章介绍算法的基本理论。第三章讨论算法解连续优化问题。第四章利用算法设计和优化神经网络。第五章介绍在组合优化中的应用。第六章介绍应用遗传程序设计解决程序设计自动化问题。第七章对遗传算法和其它适应性算法进行比较。
本书可供高校有关专业的师生、科研人员、工程技术人员阅读参考。-6. "Non - numerical parallel algorithms -- Genetic Algorithm" Liu Yong waiting 520-531 1995 version of the first book to systematically describe the non - numerical parallel algorithms one count of simulated annealing Law and the basic tenets of the latest progress, in order to help readers solve practical problems, book of the specific algorithm steps in detail. The book is divided into seven chapters, the first chapter describes the idea algorithm, characteristics. The development process and prospects. The second chapter describes the basic algorithm theory. The third chapter discusses Algorithm for continuous optimization problems. Chapter IV using the algorithm design and optimization of neural networks. The fifth chapter in combinatorial optimization applications. Chapter V Platform: |
Size: 4089255 |
Author:孙东 |
Hits:
Description: 一维边值问题线性有限元数值解法的matlab程序。有详尽的算法分析，程序注释清楚。适用于有限元方法的初学者，能帮助读者加深对基本概念的理解。-one-dimensional linear boundary value problems finite element numerical solution of Matlab procedures. A detailed analysis of algorithms, clear procedures for the Notes. Application of the finite element method in the beginners will help readers to better understand the basic concepts. Platform: |
Size: 107815 |
Author:姚飞 |
Hits:
Description: 本书是国内外业界广泛关注的7卷本《计算机程序设计艺术》第2卷的最新版。本卷对半数值算法领域做了全面介绍，分“随机数”和“算术”两章。本卷总结了主要算法范例及这些算法的基本理论，广泛剖析了计算机程序设计与数值分析间的相互联系，其中特别值得注意的是作者对随机数生成程序的重新处理和对形式幂级数计算的讨论。 本书附有大量习题和答案，标明了难易程度及数学概念的使用。 本书内容精辟，语言流畅，引人入胜，可供从事计算机科学、计算数学、计算技术诸方面的工作人员参考、研究和借鉴，也是相关专业高等院校的理想教材和教学参考书。 -the book industry at home and abroad is widespread concern in seven volumes of the "Art of Computer Programming" volume 2, the latest version. The present volume of semi - numerical algorithm to do a full field, at the "random" and "arithmetic" two chapters. This volume summary of the main examples of these algorithms on the basic theory and extensive analysis of the computer program design and numerical analysis of the linkages between, which is particularly noteworthy is the author of Random Number Generation procedures for processing and re-form calculation of power series. The book with a large number of exercises and answers, indicating the degree of difficulty and the use of mathematical concepts. The contents are incisive, language fluent and attractive Platform: |
Size: 38065202 |
Author:邵宇峰 |
Hits:
Description: ************************************************************************ * * * * * 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. Platform: |
Size: 11008084 |
Author:gtcewli3 |
Hits:
Description: 本书是国内外业界广泛关注的7卷本《计算机程序设计艺术》第2卷的最新版。本卷对半数值算法领域做了全面介绍，分“随机数”和“算术”两章。本卷总结了主要算法范例及这些算法的基本理论，广泛剖析了计算机程序设计与数值分析间的相互联系，其中特别值得注意的是作者对随机数生成程序的重新处理和对形式幂级数计算的讨论。 本书附有大量习题和答案，标明了难易程度及数学概念的使用。 本书内容精辟，语言流畅，引人入胜，可供从事计算机科学、计算数学、计算技术诸方面的工作人员参考、研究和借鉴，也是相关专业高等院校的理想教材和教学参考书。 -the book industry at home and abroad is widespread concern in seven volumes of the "Art of Computer Programming" volume 2, the latest version. The present volume of semi- numerical algorithm to do a full field, at the "random" and "arithmetic" two chapters. This volume summary of the main examples of these algorithms on the basic theory and extensive analysis of the computer program design and numerical analysis of the linkages between, which is particularly noteworthy is the author of Random Number Generation procedures for processing and re-form calculation of power series. The book with a large number of exercises and answers, indicating the degree of difficulty and the use of mathematical concepts. The contents are incisive, language fluent and attractive Platform: |
Size: 38065152 |
Author:邵宇峰 |
Hits:
Description: 6。《非数值并行算法——遗传算法》，刘勇等着 科学出版社 1995年第一版
本书系统地叙述了非数值并行算法之一的模拟退火算法的基本原理以及最新进展，同时为了便于读者解决实际问题，书中对具体算法的步骤作了详细介绍。本书共分七章，第一章介绍算法的思想、特点。发展过程和前景。第二章介绍算法的基本理论。第三章讨论算法解连续优化问题。第四章利用算法设计和优化神经网络。第五章介绍在组合优化中的应用。第六章介绍应用遗传程序设计解决程序设计自动化问题。第七章对遗传算法和其它适应性算法进行比较。
本书可供高校有关专业的师生、科研人员、工程技术人员阅读参考。-6. "Non- numerical parallel algorithms-- Genetic Algorithm" Liu Yong waiting 520-531 1995 version of the first book to systematically describe the non- numerical parallel algorithms one count of simulated annealing Law and the basic tenets of the latest progress, in order to help readers solve practical problems, book of the specific algorithm steps in detail. The book is divided into seven chapters, the first chapter describes the idea algorithm, characteristics. The development process and prospects. The second chapter describes the basic algorithm theory. The third chapter discusses Algorithm for continuous optimization problems. Chapter IV using the algorithm design and optimization of neural networks. The fifth chapter in combinatorial optimization applications. Chapter V Platform: |
Size: 4088832 |
Author:孙东 |
Hits:
Description: 一维边值问题线性有限元数值解法的matlab程序。有详尽的算法分析，程序注释清楚。适用于有限元方法的初学者，能帮助读者加深对基本概念的理解。-one-dimensional linear boundary value problems finite element numerical solution of Matlab procedures. A detailed analysis of algorithms, clear procedures for the Notes. Application of the finite element method in the beginners will help readers to better understand the basic concepts. Platform: |
Size: 107520 |
Author:姚飞 |
Hits:
Description: 本书首先介绍了MATLAB 语言的程序设计的基本内容，在此基础上系统介绍了各个应用数
学领域的问题求解，如基于MATLAB 的微积分问题、线性代数问题的计算机求解、积分变换和复变函数问题、非线性方程与最优化问题、常微分方程与偏微分方程问题、数据插值与函数逼近问题、概率论与数理统计问题的解析解和数值解法等。还介绍了较新的非传统方法如模糊逻辑与模糊推理、神经网络、遗传算法、小波分析、粗糙集及分数阶微积分学等领域。-The book begins with the introduction of the MATLAB programming language, the basic content, on the basis of the system introduced in all areas of applied mathematics problem solving, such as MATLAB-based calculus problems, linear algebra to solve computer problems, integral transforms and complex variable function problems, nonlinear equations and optimization problems, ordinary differential equations and partial differential equation problems, data interpolation and function approximation problem, Probability Theory and Mathematical Statistics Analytical Solution and Numerical Solution of such. Also introduce a relatively new non-traditional methods such as fuzzy logic and fuzzy reasoning, neural networks, genetic algorithms, wavelet analysis, rough sets and fractional calculus and other fields. Platform: |
Size: 797696 |
Author:gongyl |
Hits:
Description: Matlab是Matlab产品家族的基础，它提供了基本的数学算法，例如矩阵运算、数值分析算法，Matlab集成了 2D和3D图形功能，以完成相应数值可视化的工作，并且提供了一种交互式的高级编程语言——M语言，利用M语言可以通过编写脚本或者函数文件实现用户自己的算法。
这是关于matlab的教程-Matlab family of products is the basis of Matlab, which provides the basic mathematical algorithms, such as matrix computation, numerical analysis algorithms, Matlab integrates 2D and 3D graphics features, in order to complete the corresponding numerical visualization work, and provides an interactive high-level programming languages- M language, the use of M language can function through the preparation of the script or document to achieve the user Platform: |
Size: 189440 |
Author:lyd |
Hits:
Description: MATLAB作为一种智能化高级语言，集公式演算推异与数值计算于一体，兼有出色的图形处理和数据分析功能，作为多领域的应用软件，MATLAB的工具箱对相关学科的各处基本技术都采用了当今最先进的算法。本书共分7章，先简要介绍了MATLAB的安装及启动后，较为详细地介绍了MATLAB的基础知识，深入浅出地分析了MARLAB的符号运算功能以及绘图功能，并提供了大量的实例，便于读者自学。 -MATLAB as an intelligent high-level language, set up different calculus formulas and numerical calculation in one, a combination of excellent graphics processing and data analysis functions, as in many fields of application software, MATLAB Toolbox for related disciplines throughout the basic skills have adopted the most advanced algorithms. The book is divided into 7 chapters, briefly introduced the MATLAB installation and start-up, the more detailed description of the basic knowledge of MATLAB, in layman s language and analysis of the symbolic operation MARLAB functions and graphics functions, and provide a large number of examples, easy readers self-learning. Platform: |
Size: 6838272 |
Author:etcher |
Hits:
Description: 是《C++数值算法（第二版）》的配套源码。很好用。基本包含了所有数值计算的算法。-It is the source code of the book named Numerical Recipes in Cpp(second edition). It just works well and contains all the basic numerical algorithms. Platform: |
Size: 417792 |
Author:原非木 |
Hits:
Description: 计算机程序设计艺术
《计算机程序设计艺术》重译自Donald E. Knuth(汉名高德纳)的三卷著作："The Art of Computer Programming: 1. Fundamental Algorithms 2. Seminumerical Algorithms 3. Sorting and Searching"；三卷中文名为《基本算法》、《半数值算法》及《排序与查找》。本书内容博大精深，作者因为三卷书获得美国计算机协会1974年图灵奖(该奖被国际公认为计算机科学领域的最高奖项)。-
"Art of Computer Programming" re-translated from Donald E. Knuth (Chinese name Donald Knuth) three-volume work: "The Art of Computer Programming: 1. Fundamental Algorithms 2. Seminumerical Algorithms 3. Sorting and Searching" three volumes Chinese called "the basic algorithm", "semi-numerical algorithm" and "sort and find." Book content is profound, because the three-volume book on the U.S. Association for Computing Machinery Turing Award in 1974 (the prize is internationally recognized as the highest award in the field of computer science). Platform: |
Size: 1874944 |
Author:蟲子 |
Hits:
Description: 科学与工程数值算法〔Visual C++版〕
本书介绍了在科学与工程实际工作中常用的数值计算算法的原理和Visual C++编程方法。本书分为7章，前6章分别讨论了复数运算、矩阵运算、线性代数方程组的求解、非线性方程与方程组的求解、插值和数值积分等的面向对象编程方法，涉及使用频率非常高的近90个基本算法，按功能设计成了6类。第7章将这些算法类集成到一个静态库和一个动态库中，可以直接使用。每章节都用Visual C++程序示例了算法和算法库的调用方式。
本书适合涉及科学与工程数值计算工作的科研人员、工程技术人员、管理人员以及大专院校相关专业的师生参考阅读。-science and engineering numerical algorithm- Visual C++ version. This book describes the practical work in science and engineering numerical algorithm used in the principles and Visual C++ programming methods. This book is divided into seven chapters, the first six chapters discuss the complex arithmetic, matrix operations, linear Solving algebraic equations, nonlinear equations and solving equations, interpolation and numerical integration of surface To object programming, involving the use of very high frequency of nearly 90 basic algorithm, designed by function Six categories. Chapter 7 class of these algorithms will be integrated into a static library and a dynamic library can be directly Use. Each chapter of Visual C++ program with examples of algorithms and algorithm library called. Book is suitable for numerical calculation involved in the work of science and engineering researchers, engineers, management As well as relevant professional institutions and students refer to readin Platform: |
Size: 9784320 |
Author:周善人 |
Hits:
Description: 本书论述了解非线性方程组的基本理论和方法，着重介绍：New－ton法、单纯形算法、同伦延拓法、区间迭代法，以及计算机数学库中常用的新算法，还介绍了方法的收敛性定理和方程解的存在唯一性，并且给出了有实际应用价值的、效果好的算法步骤和数值例题本书可供高等学校数学系师生，有关研究人员和工程技术人员参考。
-This book discusses the nonlinear equations about the basic theory and methods, focuses on: New-ton method, simplex algorithm, homotopy continuation method, interval iteration method, and computer math libraries commonly used in the new algorithm, also described the method of convergence theorem and the equation of the existence, uniqueness, and gives practical value to the good effects algorithms and numerical examples the steps of Mathematics teachers and students of the book for colleges and universities, the researchers and engineers officers. Platform: |
Size: 2801664 |
Author: |
Hits:
Description: 《计算机程序设计艺术》重译自Donald E. Knuth(汉名高德纳)的三卷著作："The Art of Computer Programming: 1. Fundamental Algorithms 2. Seminumerical Algorithms 3. Sorting and Searching"；三卷中文名为《基本算法》、《半数值算法》及《排序与查找》。本书内容博大精深，作者因为三卷书获得美国计算机协会1974年图灵奖(该奖被国际公认为计算机科学领域的最高奖项)。-Re-translated from Donald E. Knuth, The Art of Computer Programming " (Chinese name Donald Knuth) of the three-volume book:" The Art of Computer Programming: 1. Fundamental Algorithms 2. Seminumerical Algorithms 3. Sorting and Searching " three volume Chinese called " basic algorithm" , " semi-numerical algorithm" and " Sort and Find. This book is profound, of the Association for Computing Machinery' s Turing Award in 1974 because of three books (The award is internationally recognized as the highest award in the field of computer science). Platform: |
Size: 48766976 |
Author:Andrew |
Hits:
Description: 非数值并行算法：遗传算法
本书系统的叙述了非数值并行算法之一的遗传算法的基本原理及最新进展，同时为了便于读者解决实际问题，书中对具体算法步骤做了详细介绍-Non-numerical parallel algorithms: genetic algorithms book describes the system of non-numerical parallel algorithms basic principles of genetic algorithms and the latest progress, and in order to facilitate the reader to solve practical problems, the book made for the specific algorithm steps described in detail Platform: |
Size: 4104192 |
Author:123 |
Hits:
Description: 《计算机程序设计艺术》重译自Donald E. Knuth(汉名高德纳)的三卷著作： The Art of Computer Programming: 1. Fundamental Algorithms 2. Seminumerical Algorithms 3. Sorting and Searching ；三卷中文名为《基本算法》、《半数值算法》及《排序与查找》。本书内容博大精深，作者因为三卷书获得美国计算机协会1974年图灵奖（该奖被国际公认为计算机科学领域的最高奖项）。-The Art of Computer Programming retranslation Donald e. Knuth (han gartner) three volumes of The book: The Art of Computer Programming: 1. The Fundamental Algorithms 2. The Seminumerical Algorithms 3. The Sorting and Searching Three volumes in Chinese, the basic algorithm, a numerical algorithm and the sorting and searching.And profound content of the book, the author for three books for the computer association in 1974 Turing award (the prize is internationally recognized as the highest award in the field of computer science). Platform: |
Size: 26260480 |
Author:刘烨 |
Hits:
Description: 《计算机程序设计艺术》重译自Donald E. Knuth(汉名高德纳)的三卷著作： The Art of Computer Programming: 1. Fundamental Algorithms 2. Seminumerical Algorithms 3. Sorting and Searching ；三卷中文名为《基本算法》、《半数值算法》及《排序与查找》。本书内容博大精深，作者因为三卷书获得美国计算机协会1974年图灵奖（该奖被国际公认为计算机科学领域的最高奖项）。-The Art of Computer Programming retranslation Donald e. Knuth (han gartner) three volumes of The book: The Art of Computer Programming: 1. The Fundamental Algorithms 2. The Seminumerical Algorithms 3. The Sorting and Searching Three volumes in Chinese, the basic algorithm, a numerical algorithm and the sorting and searching.And profound content of the book, the author for three books for the computer association in 1974 Turing award (the prize is internationally recognized as the highest award in the field of computer science). Platform: |
Size: 16496640 |
Author:刘烨 |
Hits:
Search - numerical algorithms in Basic