Course Content

Fuzzy Part:

We are all familiar with vague statements from everyday language: “Today, it is partly cloudy and quite cold.” Similarly, expert knowledge for running a process can also be so described: “If the boiler temperature is a little too low, turn the heating on a little.” With the help of fuzzy logic, such statements can be formalized and made usable for engineering applications. As well, it is shown how fuzzy systems can also be determined on the basis of measurement data. The lecture first introduces the basics of fuzzy logic. A fuzzy system is defined which is used for the tasks of modeling, regulation, and classification.

This part is divided into the following chapters:

  • Motivation
  • Basics of fuzzy sets and fuzzy logic
  • Fuzzy modelling
  • Fuzzy control
  • Fuzzy classification
  • Industrial applications of fuzzy systems


Neuro Part:

  • Theoretical foundations of artificial neural networks
  • Learning strategies (Hebbian learning, delta rule learning, competitive learning)
  • Presentation of basic network types such as Perceptron, Adaline, Madaline, back-propagation networks, Kohonen networks
  • Modeling with the help of neural networks for static (polynomial model) and dynamic (difference equation model, Volterra series model) nonlinear systems including appropriate issues (possible errors, data preprocessing, design of the learning process)
  • Structures for control / regulation using neural networks (copying a conventional controller, inverse system model, internal model control, model predictive control, direct training of a neural controller, reinforcement learning)
  • Methods for neuroclassification (backpropagation, learning vector quantization). Neurofuzzy Methods. Application examples and presentation of development tools for artificial neural networks

Course Instructor

Dr. Fred Roß