CS 367: Model-Based Reasoning Lecture 1 (01/10/2002) Gautam Biswas Logistics Class time: Tu.– Thurs. 1:10-2:25 pm Location: Room 313, Featheringill Hall Instructor: Gautam Biswas Office: Room 250, Featheringill Hall Tele: 343-6204 E-mail: biswas@vuse.vanderbilt.edu Office Hours: Tues., Friday: 9:00-10:00 Lab: EHS Lab, Room 331 Featheringill Topics Discrete-Event Systems Ref: Introduction to Discrete Event Systems by Cassandras and Lafortune (1999) Continuous Systems Ref: Systems Dynamics: A Unified Approach by Karnopp, Rosenberg and Margolis (1990) Hybrid Systems Ref: From lecture notes Grading Scheme Home Work Assignments (4): Mathlab, 20Sim, hcc, other packages: 25% Classroom Presentation(s) – Advanced topics from basic material+ classroom participation: 15% Class Project: 60% Project Report and Demonstration: 50% Presentation of Project: 10% Class projects will be individual efforts – it can be a theoretical and conceptual study, an experimental study, an implementation, or a combination of the three. What is Model-based Reasoning? Methodology for analyzing, understanding, and predicting behavior of a system, process, or phenomena using a model. System: engineering process (continuous) – chemical plants, aircraft, traffic system, information process (computer-based or digital) – computer network, telephone system, economic process – U.S. economy, world economy, software industry Social systems – suburban culture, welfare system Sometimes hard to categorize systems – notion of hybrid systems. What are we going to cover in this course? Modeling of systems – look at different modeling paradigms Continuous system modeling – quantitative and qualitative Discrete-event modeling Hybrid Modeling – modeling multiple paradigms in an integrated fashion (continuous + discrete) What are we going to cover in this course? Analysis of systems: understanding, explanation, prediction, and problem solving Analytic methods Simulation-based methods Special purpose methods for achieving particular tasks, e.g., verification, monitoring, and fault detection and isolation List of Topics Introductory Lecture (01/16): Modeling of Systems and its applications Topic 1(2-3 weeks): Discrete Event Modeling of Systems (ref: S. Lafortune, et al. – Automata based models, Petri Net based models(?)) Topic 2(2-3 weeks): Modeling and Analysis of Physical Systems: The Bond Graph Approach (Ref: D.C. Karnopp, D.L. Margolis, and R.C. Rosenberg Systems Dynamics: A Unified Approach, 2nd ed., John Wiley, NY, 1990). List of Topics (contd. …..) Topic 3 (2-3 weeks): Methods for analysis of DiscreteEvent and Continuous Dynamic Systems – applications to control and diagnosis. (Collection of papers). Topic 4 (2-3 weeks): Modeling and Analysis of Hybrid Systems. (Lecture Notes and collection of papers). Topic 5(2 weeks): Fault Detection and Isolation in Hybrid Systems - (Building Hybrid Observers and Fault Isolation units). Topic 6 (rest of semester): Advanced Topics, such as applications to computational architectures for embedded systems, embedded control, and fault-adaptive control. Lecture 1: (Home Work) Additional reading material: F.E. Cellier, H. Elmqvist, and M. Otter, “Modeling from Physical Principles,” (pdf file URL: http://www.vuse.vanderbilt.edu/~biswas/Courses/cs367/papers) Problems: 1. The height of water in a reservoir fluctuates with time. If you had to construct a dynamic system model to help water resource planners predict variations in the height, what input quantities would you consider? How many state variables would you need in your model? 2. Suppose you were a heating engineer and you wished to consider your house as a dynamic system. Without a heater the average temperature of the house would clearly vary over a 24 hour period. What might you consider as state variables for a simple dynamic model? How would you expand your model to predict the temperatures in several rooms in your house? How does the installation of a thermostat controlled heater change your model? Modeling of Dynamic Systems What exactly is a system? Ambiguous word, but some characterization: Entity that is separable from the rest of the universe (we call this the environment) by a physical or conceptual boundary e.g., human body is a system, coffee maker is a system, air traffic control is a system, school of fish is a system. Satisfies the concept of reticulation, i.e., a system can be looked upon as being made up of interacting parts e.g., human body made up of various organs and the nervous system; coffee maker, container for water, boiler (evaporator), passages, filter, etc.; air traffic control system – people, machines, schedules, etc. What is a Model? Simplified, abstracted constructs of a system used to predict the behavior of the system. Small (finite) description of a very complex reality Typically constructed to answer particular questions. If question changes, does the model change? Typically modeling involves two tasks: (i) model building, and (ii) model analysis Types of Models Ordinary Differential equations + Algebraic relations (Quantitative models) Qualitative Descriptions, especially if our knowledge of domain and situation is incomplete. Other issues – starting point for model building, causal versus acausal models, etc. Systems Viewpoint to Modeling Model to study operation of complete system as opposed to operation of the individual parts Method of model building: compositionality Method of analysis: isolate into parts Unified approach to modeling, rather than being domain-specific Qualitative reasoning schemes based on (incomplete) analytic models of system Energy-based modeling of physical systems Discrete-event models of systems – change in system directly linked to the occurrence of events Combine modeling paradigms – Hybrid Systems approach to modeling Modeling of Systems System In general, all system models will be mathematical functions Subsystems At times, we will introduce topological information (e.g., for diagnosis) Components --- primitives Black box Interface with other components Mathematical or Logical function Modeling of Dynamic Systems State-Determined Systems -- our goal is to start with physical component descriptions of systems understanding of component behavior to create mathematical models of the system. Mathematical model of state-determined system – defined by set of ordinary differential equations on the so-called state variables. Algebraic relations define values of other system variables to state variables. Dynamic behavior of state-determined system defined by (i) values of state variables at some initial time, and (ii) future time history of input quantities to system. In other words, our system models – satisfy the Markov property Uses of Dynamic Models Analysis for prediction, explanation, understanding, and control. Two types: (a) analytic methods, and (ii) simulation-based methods. Given S, X at present, and U for the future, predict future X and Y. Identification. Given U and Y find S and X consistent with U and Y. (under normal and faulty conditions) Synthesis. Given U and a desired Y, find S such that S acting on U produces Y. Input Variables U Dynamic System, S State Variables, X Output Variables Y