Mechanik und Mechatronik
> Zum Inhalt


We want cars to run smoothly and safely, we want buildings to maintain a comfortable living atmosphere, and we want to understand and optimise the flow of urban traffic. At first glance, these topics have nothing in common – but they are all about complex interconnected systems, which have to be controlled and optimised using advanced mathematical methods in order to make them behave the way we want. The availability of powerful computers enables the realisation of sophisticated model-based control concepts in many different application areas. This trend has evoked a continuous demand for new and effective methodologies in model design and process optimisation.

Many mechanical systems, such as a planet orbiting the sun, can be fully analysed from first principles of physics. But most complex real-world-systems have to be characterised by carefully studying the intricate connections between input parameters and system behaviour, casting them into mathematical equations.

The behaviour of many complex systems is usually described by a set of differential equations, most often nonlinear in their nature. Once the necessary parameters are available, such a description becomes a mathematical model, which can be used for the optimal operation of the system as well as for forecasting and diagnosis.

The list of possible application areas is endless – it ranges from controlling industrial machines in order to minimise vibrations over acoustic optimisation of loudspeakers to regulating eco-friendly combustion processes in chemical engineering.

Sometimes it is not even enough to have a precise mathematical description of a system available. The model also has to be sufficiently simple – especially if it is supposed to be used for real-time process control. This gives rise to the discipline of model reduction.

The Research Group of control and process automation specialises in model-based control and optimisation of complex industrial systems. They offer unique expertise in the following disciplines:

  • data-driven modelling of nonlinear multivariable systems
  • advanced model reduction methodologies
  • model predictive control of complex interconnected systems
  • active vibration control using robust design approaches
  • impedance control for hardware in the loop system
  • optimisation based control of energy systems (e.g. power plants, smart grids)

Selected Publications

J. Unger, C. Hametner, S. Jakubek, M. Quasthoff 
A novel methodology for non-linear system identification of battery cells used in non-road HEV, Journal of Power Sources, 2014

C.Hametner, C.Mayr, S.Jakubek 
Dynamic NOx emission modelling using local model networks, Int. Journal of Engine Research, 2014

N. Euler-Rolle, S. Jakubek, G. Offner
Model order reduction by projection applied to the universal Reynolds‘ equation, Math. and Computer Modelling of Dyn. Systems, 2014

O. König, G. Prochart, C. Hametner, S. Jakubek
Battery Emulation for Power-HIL Using Local Model Networks and Robust Impedance Control, Industrial Electronics, 2014

M. Killian, B. Mayer, A. Schirrer, M.Kozek
Cooperative Fuzzy Model Predictive Control, IEEE Transactions on Fuzzy Systems, 2015

E. Talic, A. Schirrer, M. Kozek, S. Jakubek 
Multi-objective parameter identification of Euler–Bernoulli beams under axial load, Journal of
Sound and Vibration, 2015