Course detail
Modelling and Identification
FEKT-MPC-MIDAcad. year: 2023/2024
The subject is oriented on:
- identification methods of dynamic systems
- approaches towards nonparametric and parametric identification
- on-line and off-line identification
- spectral estimation, assessment of noise and disturbance influence on identification results
Language of instruction
Number of ECTS credits
Mode of study
Guarantor
Entry knowledge
Rules for evaluation and completion of the course
Individual project - Max. 15 points.
Final Exam - Max. 70 points.
The content and forms of instruction in the evaluated course are specified by a regulation issued by the lecturer responsible for the course and updated for every academic year.
Aims
After passing the course, student should be able to
- use non-parametric identification methods
- select suitable excitation signal for the identification
- program and use the basic least squares method
- explain where biased estimate comes from and how to overcome this issue
- use the approaches which enable to enhance the quality of the estimates during practical applications
- utilize the universal programming equipment of MATLAB Simulink and its toolboxes for the identification of dynamic systems
Study aids
Prerequisites and corequisites
Basic literature
Soderstrom, T., Stoica, P.: System Identification. Prentice Hall International, 1989 (EN)
Recommended reading
Isemrann, R., Munchhof, M. : Identification of Dynamic Systems - An Introduction with Applications. Springer 978-540-78878-2, 2011. (EN)
Noskievič, P.: Modelování a identifikace systémů. Montanex Ostrava 1999 (CS)
Šimandl, M.: Identifikace systémů a filtrace. Západočeská univerzita v Plzni, 2001, ISBN 80-7082-170-1. (CS)
Elearning
Classification of course in study plans
Type of course unit
Lecture
Teacher / Lecturer
Syllabus
2. Nonparametric identification methods, correlation methods, frequency response measurement.
3. Input signal for identification, degree of persistent excitation, pseudorandom binary sequence.
4. Least squares method, derivation, geometric representation, properties.
5. Dynamic system models for system identification, ARX, ARMAX ARARX, general model, pseudolinear regression.
6. Recursive LSM. Numerically stable methods based on square root filtering.
7. Instrumental variable methods. Method with delayed observations, method with additional model.
8. Identification methods based on prediction error whitening. Noise model identification.
9. Practical notes on system identification.
10. Identification using neural nets and fuzzy modeling.
11. Another approaches to system identification.
12. Identification of nonlinear dynamic systems.
13. Course summary.
Exercise in computer lab
Teacher / Lecturer
Syllabus
2. Parameter identification from impulse and step response.
3. Identification using correlation methods, frequency.
4. Spectral analysis.
5. Input signals for identification, least squares method.
6. Recursive least squares method.
7. Instrumental variable methods.
8. System Identification Toolbox.
9. Identification methods base on whitening of prediction error.
10. Extended frequency analysis, PMS motor parameters identification.
11. Identification using recursive least squares method.
12. Test + work on project.
13. Real experiment - DC motor parameters identification.
Elearning