39-647: Special Topics in Design
Modeling and Simulation in Systems Engineering


(Spring 2002)


Lectures

Time: Tuesday 9:30-11:20
Room: Hamburg Hall 2224 (tentative)

Instructor

Chris Paredis
paredis@cmu.edu
Hamburg Hall 1205, 8-8299
Office Hours: TBA

Pre-requisites

Open to all seniors and graduate students in CIT, others with approval of the instructor.

 

COINSIDE.jpg (30822 bytes)

Course Overview

Modeling and Simulation (M&S) have become important tools for analyzing and designing complex systems in a broad array of disciplines ranging from business and engineering to biology and psychology. For example, in engineering design, M&S can be used to evaluate the effectiveness of a new product concept, verify whether all the functional design specifications are met, or suggest modifications for improving the manufacturability of a product. By using simulations in this fashion, designers can achieve significant reductions in design cycle time and overall lifetime cost of new products. In other disciplines, simulations allow us to perform experiments that cannot be realized in the real world due to physical, environmental, economic, or ethical restrictions. However, simulations are only meaningful if the underlying models are adequately accurate and if the models are evaluated using the proper simulation algorithms. To accomplish this, a systems engineer requires a variety of knowledge and skills in different disciplines. In this course, students will learn about different modeling and simulation techniques and how they can be used in real-world systems engineering problems. They will learn to develop system-level simulation models of both energy-based systems (continuous time) and discrete event systems (DEVS). Through course projects, the students will obtain hands-on experience in modeling and simulation of complex multi-disciplinary systems. The students will perform an entire simulation study of a multi-disciplinary system, from requirements definition and modeling, through simulation and analysis of the results.

 

Course Schedule

Lecture 1: Jan 15

Introduction:Course overview. What is modeling? What is simulation? Methodology for simulation studies.
Projects: Suggested topics. Group assignments.

 

Lecture 2: Jan 22

Types of Models and Simulations: Static v. discrete event v. continuous; algebraic equations v. differential equations; lumped parameter v. distributed parameter; physics-based v. behavioral; ODE v. PDE v. DAE;
Modeling of Energy-based Systems: across and through variables, energy conservation, and causality.

 

Lecture 3: Jan 29

Object Oriented Modeling: equation-based v. procedural modeling; Simulink v. Modelica; the Modelica language; the Dymola simulation software

 

Lecture 4: Feb 5

Ordinary Differential Equations: definition; initial value problems vs. boundary value problems; uniqueness of solutions;
Integration algorithms: Euler, Runge-Kutta, implicit v. explicit; illustrated with Dymola.

 

Lecture 5: Feb 12

Differential Algebraic Equations: as compared to ordinary differential equation; complexity and index; solvers; solving DAEs in Simulink v. Modelica.

 

Lecture 6: Feb 19

Intermediate project presentations: project description and approach.

 

Lecture 7: Feb 26

Modeling of physical systems: electrical, hydraulic, and thermal systems; electrical and hydraulic libraries in Modelica.

 

Lecture 8: Mar 5

Modeling of physical systems: mechanical and electro-mechanical systems; mechanical and electro-mechanical models in Modelica.

 

Lecture 9: Mar 12

Modeling of 3D mechanical systems: rigid body dynamics; Newton-Euler equations; kinematic loops; 3D mechanics in Modelica

 

Lecture 10: Mar 19

Discrete Event Systems (DEVS): definition, models, queuing example, comparison with energy-based systems.

 

Lecture 11: Mar 26

Discrete event simulation with Arena: the Arena simulation software; building models in Arena.

 

Lecture 12: Apr 9

Input distributions: random numbers; common probability distributions; which distribution to use?
Output data analysis: terminating simulations, estimating means, estimating steady-state parameters.

 

Lecture 13: Apr 16

Monte-Carlo Simulation: sources of uncertainty, variance reduction techniques.
Design of Experiments: basic principles, factorial design,

 

Lecture 14: Apr 23

Simulation-Based Design: the use of modeling and simulation in engineering design; Simulation and optimization.

 

Lecture 15: Apr 30

Final project presentations

 

Reference Books (not required)


page maintained by paredis@cmu.edu