Modeling and Simulation in Scilab/Scicos with ScicosLab 4.4

Transcription

1 Modeling and Simulation in Scilab/Scicos with ScicosLab 4.4

2

3 Stephen L. Campbell, Jean-Philippe Chancelier and Ramine Nikoukhah Modeling and Simulation in Scilab/Scicos with ScicosLab 4.4 Second Edition 123

4 Stephen L. Campbell Department of Mathematics North Carolina State University 2108 SAS Hall P.O. Box 8205 Raleigh NC USA Ramine Nikoukhah INRIA Rocquencourt Le Chesnay Cedex France Jean-Philippe Chancelier Ecole Nationale des Ponts et Chaussées Centre d Enseignement et de Recherche en Mathématiques et Calcul Scientifique 6-8 avenue Blaise Pascal Marne-la-Vallee Cité Descartes, Champs sur Marne France chancelier@cermics.enpc.fr ISBN e-isbn DOI / Springer New York Dordrecht Heidelberg London MSC 2010: 68U20, 68U07, 65-01, 37M05, 00A72, 65Y15, 34H05 Library of Congress Control Number: c Springer Science+Business Media, LLC 2010 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Springer is part of Springer Science+Business Media (

5 Preface ScicosLab ( is a free open-source software package for scientific computation. ScicosLab includes a fork of Scilab, based on Scilab 4, the modeling and simulation tool Scicos and a number of other toolboxes. Scilab is an interpreted language specifically developed for matrix based numerical computations. It includes hundreds of general purpose and specialized functions for numerical computation, organized in libraries called toolboxes that cover such areas as simulation, optimization, systems and control, and signal processing. These functions reduce considerably the burden of programming for scientific applications. One important ScicosLab toolbox is Scicos. Scicos ( provides a block-diagram graphical editor for the construction and simulation of dynamical systems. Scilab/Scicos is the only open-source alternative to commercial packages for dynamical system modeling and simulation packages such as MATLAB/Simulink and MA- TRIXx/SystemBuild. Widely used at universities and engineering schools, Scilab/Scicos has also gained ground in industrial environments. ScicosLab is developed and maintained by research groups, in particular, at INRIA 1 and ENPC. 2 ScicosLab includes full Scilab and Scicos user s manuals, which are available with search capabilities in a help window. All commands, their syntax, and simple illustrative examples are given. While very useful in finding out the details of a particular command, these manuals do not provide a tutorial on the philosophy of either Scilab or Scicos. Nor do they address how to use several of these commands together in the solution of a technical problem. The objective of this book is to provide a tutorial for the use of Scilab/Scicos with a special emphasis on modeling and simulation tools. While it will provide useful information to experienced users, it is designed to be accessible to beginning users from a variety of disciplines. Students [52] and academic and industrial scientists and engineers should find it useful. The discussion includes some information on modeling and simulation in order to assist the reader in deciding which simulation tools might be most useful to them. Every software environment has its special features, some would say quirks, that experienced users automatically take into account but often prove confusing to beginning users. We have tried to point these out where appropriate. The book is divided into two parts. The first part concerns Scilab and includes a tutorial covering the language features, the data structures and specialized functions for doing graphics, importing and exporting data, interfacing with external routines, etc. It 1 Institut National de Recherche en Informatique et en Automatique 2 Ecole Nationale des Ponts et des Chaussées V

6 VI Preface also covers in detail Scilab numerical solvers for ODEs (ordinary differential equations) and DAE s (differential-algebraic equations). Even though the emphasis is placed on modeling and simulation applications, this part provides a global view of the product. The second part is dedicated to modeling and simulation of dynamical systems in Scicos. Scicos provides a block-diagram editor for constructing models. This type of modeling tool is widely used in industry because it provides a means for constructing modular and reusable models. This part contains a detailed description of the editor and its usage, which is illustrated through numerous examples. It also covers advanced subjects such as constructing new blocks and batch simulation. Code generation and debugging are other topics covered. Finally, a new extension of Scicos is discussed. This extension allows the use of components described by the Modelica ( language. There have been several previous books written about Scilab. Most of them have been in French [20, 3, 2, 1, 29] and dealt with earlier versions of Scilab, as in [16]. The current book, like the earlier [18], is unique in a number of ways. They focus on simulation and modeling, put a major emphasis on Scicos, and discuss Scicos in depth. However, unlike [18] which was written before ScicosLab, Modeling and Simulation in Scilab/Scicos with ScicosLab 4.4 is the first to deal with the new ScicosLab 4.4 version and the latest version of Scicos, Scicos 4.4. The source of all the examples presented in this book can be downloaded from Finally, a large number of people have supported us in many ways. We would especially like to thank our wives, Gail Campbell and Homa Nikoukhah, and our parents, Aline and René Chancelier, for their support in this and everything else we do. Steve Campbell Jean-Philippe Chancelier Ramine Nikoukhah

7 Contents Part I Scilab 1 General Information What Is Scilab? How to Start? Installation First Steps Line Editor Documentation Typical Usage ScicosLab on the Web Introduction to Scilab Scilab Objects Matrix Construction and Manipulation Strings Boolean Matrices Polynomial Matrices Sparse Matrices Lists Functions Scilab Programming Branching Iterations Scilab Functions Debugging Programs Input and Output Functions Display of Variables Formatted Input and Output Input Output in Binary Mode Accessing the Host System Graphical User Interface Scilab Graphics Basic Graphing Graphic Tour Graphics Objects VII

8 VIII Contents Scilab Graphics and L A TEX Old Graphics Style Interfacing Linking Code Writing an Interface Dynamic Loading Modeling and Simulation in Scilab Types of Models Ordinary Differential Equations Boundary Value Problems Difference Equations Differential Algebraic Equations Hybrid Systems Simulation Tools Ordinary Differential Equations Boundary Value Problems Difference Equations Differential Algebraic Equations Hybrid Systems Optimization Comments on Optimization and Solving Nonlinear Equations General Optimization Solving Nonlinear Equations Nonlinear Least Squares Parameter Fitting Linear and Quadratic Programming Linear Programs Quadratic Programs Semidefinite Programs Differentiation Utilities Higher Derivatives Examples Modeling and Simulation of an N-Link Pendulum Equations of Motion of the N-Link Pendulum Generated Code and Simulation Maple Code Modeling and Simulation of a Car Basic Model Equations of Motion Simulation Model Scilab Implementation Simulation Result Open-Loop Control to Swing Up a Pendulum Model Control Problem Formulation Optimization Problem Implementation in Scilab

9 Contents IX 5.4 Parameter Fitting and Implicit Models Mathematical Model Scilab Implementation Part II Scicos 6 Introduction Getting Started Construction of a Simple Diagram Running Scicos Editing a Model Diagram Simulation Changing Block Parameters Symbolic Parameters and Context Virtual blocks and Hierarchy Placing a Super Block in a Diagram Editing a Super Block Scope of Variables in Super Block Contexts and Masking Save and Load Operations Scicos File Formats Super Block and Palette Synchronism and Special Blocks Scicos Formalism Activation Signal Block Activation Activation Generation Inheritance Always Active Blocks Constant Blocks Conditional Blocks Scicos Blocks Block Behavior External Activation Always Activation Internal Zero-Crossing Blocks Inside Palettes Modifying Block Parameters Super Block and Scifunc Super Blocks Scifunc Constructing New Basic Blocks Interfacing Function Computational Function Saving New Blocks Constructing and Loading a New Palette

10 X Contents 10 Examples and Applications Academic examples Predator Prey Model Control Application Signal Processing Application Queuing Systems Neuroscience Application A Fluid Model of TCP-Like Behavior Interactive GUI Applications Batch Processing in Scilab Piloting Scicos via Scilab Commands Function scicosim Function scicos simulate Data Sharing Context Variables Input/Output Files Special blocks: From Workspace and To Workspace Steady-State Solution and Linearization Scilab Function steadycos Scilab Function lincos Code Generation Code Generation Procedure Limitations Continuous-Time Activation Synchronicism A Look Inside Some Pitfalls Applications Debugging Error Messages Block Errors Errors During Numerical Integration Other Errors Debugging Tools Examples Log File Animation Implicit Scicos and Modelica Introduction Internally Implicit Blocks Implicit Blocks Scicos Editor Scicos Compiler Block Construction Example

11 Contents XI 14.5 Scicos Block MBlock Initialization Example A Inside Scicos A.1 Scicos Editor A.1.1 Main Editor Function A.1.2 Structure of scs m A.2 Scicos Complier A.2.1 First Compilation Stage A.2.2 Second Compilation Stage A.2.3 Structure of %cpr A.2.4 Partial Compilation A.3 Scicos Simulator B Coding Examples B.1 Scicos Blocks of Type B.1.1 Type 5 Block for the Bouncing Ball Example B.1.2 Animation Block for the Cart Pendulum Example B.2 Animation Program for the Car Example B.3 Extraction Program for the L A TEX Graphic Example B.4 Maple Code Used for Modeling the N-Link Pendulum References Index