System Identification (EE 672)

2021 Spring
Faculty of Engineering and Natural Sciences
Electronics Engineering(EE)
3
10.00
Mustafa Ünel munel@sabanciuniv.edu,
Click here to view.
English
Doctoral, Master
--
Formal lecture
Interactive,Simulation
Click here to view.

CONTENT

Aims to provide the fundamental theory of identification of dynamical systems, i.e. how to use measured input-output data to build mathematical models, typically in terms of differential or difference equations. It covers: The mathematical foundations of System Identification, Non-parametric techniques, Parametrizations and model structures, Parameter estimation, Asymptotic statistical theory, User choices, Experimental design, Choice of model structure.

OBJECTIVE

Objective of the course is to provide graduate students with a strong background in linear and nonlinear system identification to build mathematical models from experimental data.

LEARNING OUTCOME

- select inputs and outputs of a system, and characterize disturbances acting on the system.
- design suitable excitation signals,
- use measured input-output data to build mathematical models,
- solve linear regression problems by least squares methods,
- develop nonlinear NARX and Hammerstein-Wiener models
- preprocess data,
- validate obtained models

ASSESSMENT METHODS and CRITERIA

  Percentage (%)
Midterm 35
Assignment 30
Individual Project 35

RECOMENDED or REQUIRED READINGS

Textbook

System Identification, Theory for the User, 2nd Edition, Lennart Ljung, Prentice Hall, 1999.

Readings

System Identification, Karel J. Keesman, Springer-Verlag London Limited, 2011
Nonlinear System Identification, Oliver Nelles, Springer, 2001.