Downloads  | EASY-FIT demo version (3.34): Software system for parameter estimation in explicit model functions, Laplace transforms, steady-state systems, ordinary differential equations (ODE's), differential algebraic equations (DAE's), and one-dimensional partial differential equations (PDE's, PDAE's). A demo version is available containing all features of EASY-FIT with the restriction that model functions cannot be altered. The demo version is fully documented by interactive, context-sensitive help texts and contains 1,000 test examples, many of them based on real-life applications. The program runs under Windows 95, 98, NT4.0, and 2000.
To get the demo version, please contact the author and ask for a free CD-ROM, or download the software from this page. In the latter case, copy the self-extracting file EASYFIT.EXE or EASYFIT98.EXE, respectively, into a temporary directory, execute the code, and save all files in an empty installation directory of your choice. After successful extraction, start the setup program SETUP.EXE, and follow the instructions. To download the demo version of EASY-FIT for Win 95/98 (about 18 MB compressed!) click here: EASYFIT98.EXE To download the demo version of EASY-FIT for Win 2000 or higher (about 22 MB compressed!) click here: EASYFIT.EXE
IMPORTANT NOTE: To run the demo version, you need a password. Please fill out the download form completely and insert your e-mail address. You will get the actual password from the author.
</td msimagelist> | | EASY-OPT: Interactive user interface running under MS-Windows, to facilitate the formulation of nonlinear programming models, their implementation and numerical solution. It is possible to solve general nonlinear programming, least squares, L1, min-max and multicriteria problems interactively. In addition there may be any equality or inequality constraints. The software is fully documented by interactive, context-sensitive help texts and documentation in form of postscript and Acrobat Reader files. The program runs under Windows 95, 98, or 2000.
To get EASY-OPT, download the software from this page. Copy the self-extracting file EASYOPT.EXE into a temporary directory, execute the code, and save all files in an empty installation directory of your choice. After successful extraction, start the setup program SETUP.EXE, and follow the instructions (if not started automatically).
To download the full version of EASY-OPT (about 11 MB compressed!) click here: EASYOPT.EXE </td msimagelist> | | </table msimagelist> | Test problems for nonlinear programming:
306 test problems of the two collections
W. Hock, K. Schittkowski, Test Examples for Nonlinear Programming Codes, Lecture Notes in Economics and Mathematical Systems, Springer, No, 187, 1981
K. Schittkowski, More Test Examples for Nonlinear Programming Codes, Lecture Notes in Economics and Mathematical Systems, Springer, No, 282, 1987
are provided in form of Fortran source code together with a test frame. A decision is made which of the runs is successful, and performance results are evaluated. With the default tolerances given, all problems can be solved successfully by the code NLPQLP, a new version of the SQP implementation NLPQL of the author. Results of NLPQLP are included for comparative studies.
To download the test problems, click here: testprob.zip (zip-file, contains Fortran source codes for all test problems, see readme.txt for details)
</td msimagelist> | </table msimagelist> | NLPQLP: The Fortran subroutine NLPQLP solves smooth nonlinear programming problems and is an extension of the code NLPQL. The new version is specifically tuned to run under distributed systems. A new input parameter l is introduced for the number of parallel machines, that is the number of function calls to be executed simultaneously. In case of l=1, NLPQLP is identical to NLPQL. Otherwise the line search is modified to allow parallel function calls either for the line search or for approximating gradients by difference formulae. The mathematical background is outlined, in particular the modification of line search algorithms to retain convergence under parallel systems. Numerical results show the sensitivity of the new version with respect to the number of parallel machines, and the influence of different gradient approximations under uncertainty. The performance evaluation is obtained by more than 300 standard test problems. To download the report, click here: nlpqlp.zip (zip-file, contains postscript and Acrobat Reader version)
</td msimagelist> | | NLPQLG: Usually, global optimization codes with guaranteed convergence require a large number of function evaluations. On the other hand, there are efficient optimization methods which require gradient information, but only the approximation of a local minimizer can be expected. If, however, the underlying application model is expensive and if the existence of different local solutions is expected, then heuristic rules for successive switches from one local minimizer to a another is often the only applicable approach. For this specific situation, we present some simple rules for cutting off a local minimizer and to restart a new local optimization run. However, some safeguards are needed to stabilize the algorithm, since very little is usually known about the distribution of local minima. The paper introduces an approach where the nonlinear programs generated can be solved by any available black box software. For our implementation, a sequential quadratic programming code (NLPQLP) is chosen for the local optimization steps. The usage of the code is outlined and we present some numerical results based on a set of 30 test examples found in the literature.
To download the report, click here: nlpqlg.zip (zip-file, contains postscript and Acrobat Reader version) </td msimagelist> | | NLPJOB:
The paper describes the usage of the Fortran subroutine NLPJOB for solving smooth nonlinear multiobjective or multicriteria problems, respectively, by a transformation into a scalar nonlinear program. Provided are 15 different possibilities to perform the transformation, depending on the preferences of the user. The subproblem is solved by the sequential quadratic programming code NLPQL. The usage of the code is outlined and an illustrative example is presented.
To download the report, click here: nlpjob.zip (zip-file, contains postscript and Acrobat Reader version) </td msimagelist> | | </table msimagelist> | QL:
The algorithm is an implementation of the dual method of Goldfarb and Idnani and a modification of the original implementation of Powell. Initially, the algorithm computes a solution of the unconstrained problem by performing a Cholesky decomposition and by solving the triangular system. In an iterative way, violated constraints are added to a working set and a minimum with respect to the new subsystem with one additional constraint is calculated. Whenever necessary, a constraint is dropped from the working set. The internal matrix transformations are performed in numerically stable way. The usage of the code QL is outlined and a numerical example is presented.
To download the report, click here: ql.zip (zip-file, contains postscript and Acrobat Reader version)
</td msimagelist> | | DFNLP:
The Fortran subroutine DFNLP solves constrained nonlinear programming problems, where the objective function is of the form
- sum of squares of function values,
- sum of absolute function values,
- maximum of absolute function values,
- maximum of functions.
It is assumed that all individual problem functions are continuously differentiable. By introducing additional variables and constraints, the problem is transformed into a general smooth nonlinear program which is then solved by the SQP code NLPQL. For the least squares formulation, it can be shown that typical features of special purpose algorithms are retained, i.e., a combination of a Gauss-Newton and a quasi-Newton search direction. In this case, the additionally introduced variables are eliminated in the quadratic programming subproblem, so that calculation time is not increased significantly. Some comparative numerical results are included, the usage of the code is documented, and an illustrative example is presented.
To download the report, click here: dfnlp.zip (zip-file, contains postscript and Acrobat Reader version) </td msimagelist> | | </table msimagelist> | Nonlinear Programming Review: Nonlinear programming is a direct extension of linear programming, when we replace linear model functions by nonlinear ones. Numerical algorithms and computer programs are widely applicable and commercially available in form of black box software. However, to understand how optimization methods work, how corresponding programs are organized, how the results are to be interpreted, and, last not least, what are the limitations of the powerful mathematical technology, it is necessary to understand at least the basic terminology. Thus, we present a brief introduction into optimization theory, in particular we introduce optimality criteria for smooth problems. These conditions are extremely important to understand how mathematical algorithms work. The most popular classes of constrained nonlinear programming algorithms are introduced, i.e., penalty-barrier, interior point, augmented Lagrangian, sequential quadratic programming, sequential linear programming, generalized reduced gradient, and sequential convex programming methods. Common features and methodological differences are outlined. In particular we discuss extensions of these methods for solving large scale nonlinear programming problems. To download the report, click here: eolss.zip (zip-file, contains postscript and Acrobat Reader version) </td msimagelist> | | </table msimagelist> | Mathematische Grundlagen von Optimierungsverfahren: Vorlesungsmanuskript zur nichtlinearen Programmierung mit folgendem Inhalt: 1. Einführung
1.1 Das nichtlineare Optimierungsproblem
1.2 Definitionen und Beispiele
1.3 Sattelpunkt und Dualität
1.4 Notwendige und hinreichende Optimalitätsbedingungen 2. Unrestringierte Optimierung
2.1 Problemstellung
2.2 Verfahren mit konjugierten Richtungen
2.3 Newton-Verfahren
2.4 Quasi-Newton-Verfahren 3. Schrittlängenbestimmung
3.1 Problemstellung
3.2 Schrittweiten mit garantierter Konvergenz
3.3 Numerische Verfahren 4. Restringierte Optimierung
4.1 Voraussetzungen
4.2 Das quadratische Teilproblem
4.3 Die Meritfunktion
4.4 Der SQP-Algorithmus
4.5 Konvergenzresultate 5. Numerische Vergleichstests
5.1 Testbeispiele
5.2 Einige allgemeine Vergleichsresultate
5.3 Vergleichsresultate für Strukturoptimierungsprobleme Literatur Download: skript.zip (zip-Datei, enthält Postscript- und Acrobat Reader-Version) </td msimagelist> | | </table msimagelist> |
|
|
|
|
|
|
