Entry requirements

120 credits and at least one of the courses Spectral processing of signals, Signals and systems or Stochastic modelling.

Learning outcomes

Students who pass the course should be able to

• describe the different phases that constitute the process of building models, from identification experiment to model validation
• account for and apply the stochastic concepts used in analysis of system identification methods, and of the models produced by system identification
• explain why different system identification methods and model structures are necessary in engineering practice
• reason about how to choose identification methods and model structures for real-life problems
• describe and motivate basic properties of identification methods like the least squares method, the prediction error method, the instrumental variable method, as well as to solve simple problems that illustrate these properties
• explain the advantages and challenges when identifying feedback systems in closed loop
• describe the principles behind recursive identification and its fields of application
• use available software to handle experimental data, for estimation and validation of process models
• interpret the outcome of model validation

Content

The system identification problem - from data to model. The least squares method. Model structures and input signals. The prediction error methods. The instrumental variable methods. Model validation and practical aspects. Identification of feedback systems.

Laboratory work:

• Five two-hour MATLAB laboratories.
• One half-day laboratory: Identification and control of a laboratory process.

Instruction

Lectures, Problem solving sessions, Laboratory work and homework assignments.

Assessment

Written examination at the end of the course. Passed laboratory course is also required.