MAT-31806 System Identification: learning for decision and control


Credits 6.00

Teaching methodContact hours
Independent study0
Course coordinator(s)Prof. dr. ir. KJ Keesman
dr. ir. LG van Willigenburg
Lecturer(s)Prof. dr. ir. KJ Keesman
dr. ir. LG van Willigenburg
Examiner(s)Prof. dr. ir. KJ Keesman
dr. ir. LG van Willigenburg

Language of instruction:


Assumed knowledge on:

BCT-20306 Modelling Dynamic Systems.

Continuation courses:

MAT-26306 Control Engineering; 
BCT-30806 Physical Modelling;
MAT-32306 Systems and Control Theory.


Mathematical models are crucial in many scientific fields. Modelling that starts from first principles can then be a good start. For operational use in real practice, however, the model should be in good accordance with experimental data.

The aim of this course is to introduce methods for the determination or identification of static/dynamic models starting from potentially limited to full prior system's knowledge, given experimental data and model objective, and to estimate the unknown parameters. In practice, the main model objectives are prediction, experiment design and management/control. In the first part of the course data-driven methods, in time and frequency domain, will also receive ample attention.
The main application fields are the agro/biotechnology, environmental sciences and food science. In particular, attention will be paid to parameter estimation and network identification within systems biology using the Systems Biology (SB) Toolbox for Matlab.
In the accompanying practical course the System Identification Toolbox will be used for the identification of a laboratory-scale process.

Learning outcomes:

After successful completion of this course students are expected to be able to:
- identify a simple dynamic process model from given input/output signals;
- model systems on the basis of prior system’s knowledge and experimental input/output data, and estimate the unknown model parameters;
- simulate and analyze “simple” models from systems biology and to estimate the unknown parameters;
- apply the 'System Identification Toolbox' of the MATLAB software package.


- self study to prepare for the lectures and practicals;
- lectures to provide some additional background material;
- computer practicals/hands-on training to apply the theory in a MATLAB environment;
- practicals using experimental data from a laboratory set-up.


- five take-home exams, containing one or two exercises (note: for submission of take home exams see time schedule);
- report of the practical exercises and observations during practicals;
- the final mark consists of the average of the take home exams (60%) and the practical report (40%);
- the marks for the individual parts need to be ≥5.5. The mark for the exam will remain valid for 6 academic years;
- the mark for the practical's and the case study will expire after one year.


Course information
K.J. Keesman (2011). System Identification: an Introduction. London. Springer. 323p. ISBN 978-0-85729-522-4. Series: Advanced Textbooks in Control and Signal Processing.
Additional Systems Biology material will be made available during the course.

Restricted Optional for: MBTBiotechnologyMScD: Spec. D - Process Technology4WD
MBEBiosystems EngineeringMSc4WD