MSc Advanced Control and Systems Engineering / Course details

Year of entry: 2025

Course unit details:
System Identification and Artificial Intelligence

Course unit fact file
Unit code EEEN60231
Credit rating 15
Unit level FHEQ level 7 – master's degree or fourth year of an integrated master's degree
Teaching period(s) Semester 1
Available as a free choice unit? No

Overview

The unit has two different parts.

Part A - System Identification

  • Examplar system identification problems.
  • Measurements and statistics.
  • Non-parametric methods: Time and frequency domain
  • Least square problem. Statistic foundation.
  • Parametric methods (ARX, OE).
  • Input design.
  • Optimisation: gradient method for OE.
  • Recursive estimation.
  • Validation.

Part B - Artificial Intelligence (for dynamic systems modelling)

  • Introduction to AI.
  • Neural network models (single-layer and multiple-layer neural networks).
  • Deep learning (neural networks).
  • Learning and optimisation methods.
  • Performance evaluation.

Aims

The unit aims to give students an understanding of how system identification and artificial intelligent algorithms can be used to find models of dynamic systems; how least squares and optimisation approaches can be used for parameter estimation; the influence of noise on the parameter estimation and the relevance of measurement theory for the identification process.

Learning outcomes

On successful completion of the course, a student will be able to:

ILO 1: Test and validate the neural networks models with different selecting criteria.

ILO 2: Optimise the neural networks model parameters with input-output data.

ILO 3: Demonstrate understanding of the potential and limitations of AI and their essential steps for dynamic system modelling, including choices of AI models and performance evaluation.

ILO 4: Describe the implementation of the data-driven techniques that can be used to identify linear and nonlinear dynamic systems.

ILO 5: Use statistical analysis techniques to describe random variation in measured data.

Syllabus

The unit has two different parts.

Part A - System Identification

  • Examplar system identification problems.
  • Measurements and statistics.
  • Non-parametric methods: Time and frequency domain
  • Least square problem. Statistic foundation.
  • Parametric methods (ARX, OE).
  • Input design.
  • Optimisation: gradient method for OE.
  • Recursive estimation.
  • Validation.

Part B - Artificial Intelligence (for dynamic systems modelling)

  • Introduction to AI.
  • Neural network models (single-layer and multiple-layer neural networks).
  • Deep learning (neural networks).
  • Learning and optimisation methods.
  • Performance evaluation.

Teaching and learning methods

Lectures, tutorial and laboratory.

Assessment methods

Method Weight
Other 10%
Written exam 80%
Written assignment (inc essay) 10%

Feedback methods

Examination - feedback will be given after the exam board.

System Identifaction coursework - individual feedback is provided 3 weeks after submission.

AI coursework - individual feedback is provided 3 weeks after submission.

Recommended reading

Söderström, T. (1989). System Identification. Prentice Hall.
Ljung, L. (1987). System Identification: Theory for the User. Prentice-Hall.
Goodwin, G.C., & Payne, R.L. (Ed.). (1977). Dynamic System Identification: Experiment Design and Data Analysis. Academic Press.
Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press.
Alpaydin, E. (2020). Introduction to Machine Learning (4th ed.). MIT Press

Study hours

Scheduled activity hours
Lectures 42
Practical classes & workshops 6
Tutorials 6
Independent study hours
Independent study 96

Teaching staff

Staff member Role
Ognjen Marjanovic Unit coordinator
Long Zhang Unit coordinator

Return to course details