Tantárgy adatlapja
Tárgy neve: Dinamikus modellek paramétereinek becslése
Tárgy kódja: P_DO_0250
Óraszám: N: 2/2/0, L: 0/0/0
Kreditérték: 6
Az oktatás nyelve: magyar
Követelmény típus: Kollokvium
Felelős kar: ITK
Felelős szervezeti egység: ITK Doktori és Habilitációs Iroda
Tárgyfelelős oktató:
Dr. Szederkényi Gábor
Tárgyleírás:
Planned weekly schedule:
- Revision of dynamical models, probability theory and statistics
- Linear regression and its statistical properties
- Least squares parameter estimation of predictive models
- Maximum likelihood estimation, theoretical limits: the Cramer-Rao inequality
- Recursive parameter estimation methods
- The instrumental variable method
- Optimization-based estimation of nonlinear models
- Structural identifiability and distinguishability
- Practical identifiability and sensitivity
- Parameter estimation using state estimators
- Practical implementation issues, application examples
- Project consultation
List of selected literature:
- Ljung, L. (1999). System identification, theory for the user. 2nd edition. Information and system science series. Prentice Hall.
- Walter, E., Pronzato, L., & Norton, J. (1997). Identification of parametric models from experimental data. Berlin: Springer.
- Boyd, S. P., & Vandenberghe, L. (2004). Convex optimization. Cambridge university press.
List of required ce
ompetences:
Understanding parameterized dynamical models and the corresponding parameter estimation problems. The students will learn different approaches for solving the parameter estimation problem, and will be able to evaluate the statistical quality of the results compared to the theoretical limits. Using the information on the model structure and the measured data, students will be able to choose and implement an appropriate method for parameter estimation.