Technische Universität Ilmenau

System Identification - Modultafeln der TU Ilmenau

Die Modultafeln sind ein Informationsangebot zu unseren Studiengängen. Rechtlich verbindliche Angaben zum Verlauf des Studiums entnehmen Sie bitte dem jeweiligen Studienplan (Anlage zur Studienordnung). Bitte beachten Sie diesen rechtlichen Hinweis. Angaben zum Raum und Zeitpunkt der einzelnen Lehrveranstaltungen entnehmen Sie bitte dem aktuellen Vorlesungsverzeichnis.

Fachinformationen zu System Identification im Studiengang Master Research in Computer & Systems Engineering 2016
Fachnummer101402
Prüfungsnummer2200507
FakultätFakultät für Informatik und Automatisierung
Fachgebietsnummer2211
Fachverantwortliche(r)Prof. Dr. Christoph Ament
TurnusSommersemester
SpracheEnglish
Leistungspunkte5
Präsenzstudium (h)34
Selbststudium (h)116
VerpflichtungPflicht
Abschlussmündliche Prüfungsleistung, 30 Minuten
Details zum Abschluss
max. Teilnehmerzahl
Vorkenntnisse

Lecture ‘Control Engineering’

Lernergebnisse

The students are able to model selected technical systems as well as signals and create a model suited for analyses, simulation and control unit design. They make experiences with both theoretical and experimental modelling principles and parameter identification, in offline and online mode. They understand the methods and algorithms for analyses and identification of dynamic systems and are able to review, select and adapt them to a precise problem.

Inhalt

If someone wants to evaluate the behavior of a technical system before it is build up, or to design a suitable controller, he is in need of a model that is a mathematical description of the system. The process of modelling is often very extensive in practice. In this lecture, the systematical procedure and methods for efficient modelling are developed. It is differentiated between theoretical and experimental modelling.

After the introduction (chapter 1) the principles of theoretical modelling (chapter 2) are introduced, based on different physical domains like e.g. electronics. These are linked by modelling analogies and displaying in block diagrams. In the area of experimental modelling, general model approaches are introduced (chapter 3) before the development of static identification principles for model parameters based on measurement data (chapter 4) and adequate proceedings for model validation (chapter 5) are presented. With the focus remaining on experimental modelling, the methods are expanded to include the modelling of signals (chapter 6) and of the dynamic behavior of technical systems, both for time discrete (chapter 8) and continuous ones (chapter 7). For the latter, three different classes of input signals are discussed (deterministic non-periodic, periodic, and stochastic).

In the final part, the focus is set on online methods for parameter identification which allow for the estimation of both parameters and system states during the runtime of the system (chapter 9). As a powerful tool, the Kalman-filter is introduced and adapted to concrete problems (chapter 10).

The lecture is structured as follows:

1 Introduction

2 Physical model approaches: White box models

3 General model approaches: Black box models

4 Parameter identification

5 Model verification

6 Identification of signal models

7 Identification of continuous-time systems

8 Identification of discrete-time systems

9 Recursive parameter estimation

10 Kalman filtering

Medienformen

During lectures, the concepts are developed at the blackboard. Also, a presenter is used to show relevant parts of the script, and to demonstrate numerical simulations.

The script is available at the copy shop and can also be downloaded. The website of the lecture features current information and tasks for the tutorials.

Literatur

D. Simon: Optimal State Estimation: Kalman, H Infinity, and Nonlinear Approaches, John Wiley & Sons, 2006

R. Isermann, M. Münchhof: Identification of Dynamic Systems - An Introduction with Applications, Springer, 2011

L. Ljung: System Identification, Theory for the user, Prentice Hall, 1999

 

 

Lehrevaluation

Pflichtevaluation:

Freiwillige Evaluation:

SS 2015 (Vorlesung)

SS 2015 (Übung)

SS 2016 (Vorlesung)

SS 2016 (Übung)

Hospitation:

Informationen und Handreichungen zur Pflege von Modul- und Fachbeschreibungen durch den Modul- oder Fachverantwortlichen finden Sie auf den Infoseiten zum Modulkatalog.