[Back]

B. Heinzl, C. Dorn, A. Dimitriou:
"Object-Oriented Modelling of Machine Tools for Energy Efficiency Analysis in Production";
Talk: MATHMOD 2012 - 7th Vienna Conference on Mathematical Modelling, Wien; 02-14-2012 - 02-17-2012; in: "Preprints Mathmod 2012 Vienna - Full Paper Volume", F. Breitenecker, I. Troch (ed.); Argesim / Asim, 38 (2012), 413 - 414.

Introduction. The research project INFO [1] (sponsored by the Austrian Research Promotion Agency (FFG))
pursues the primary goal to increase energy efficiency in production plants by considering various disciplines of
energy technology, production technology and building design in a holistic approach [2]. Qualified and customized
predictions and recommendations about the efficiency of different energy saving measures can be made using
comprehensive simulation models of the real production plant including all relevant micro- and macro-structures
and therefore identify potential savings in manufacturing plants.
One important part of this project investigates the micro-structures of production plants (individual processes and
machines) by making extensive energy analysis to point out optimization potential based on simulation models.
This also allows gaining knowledge for merging the individual levels in the modelling approach to an ultimately
complete simulation. Some of these aspects are studied in more detail by creating a multi-domain model of a
turning lathe as an example of a machine tool.
Modelling Method. A new high-level object-oriented modelling approach provides the necessary flexibility regarding
modularity and reusability. The structured nature of this modelling method allows for simple stepwise
development and easy expandability of a multi-domain model including electrical, mechanical and thermal aspects
of the machine tool. Simulation results are validated against real measurement data obtained from the actual turning
lathe, of which the model is created and which is provided by the Institute for Production Engineering and
Laser Technology from the Vienna University of Technology.
Although simulators for object-oriented component-based modelling of physical systems have evolved considerably
in the last years (e.g. Dymola, MATLAB/Simscape^TM, MapleSim), there are still numerical issues when it
comes to simulating complex multi-domain systems. Bottom-up modelling of this component-based approach is
therefore combined with stepwise top-down modelling in several stages with gradually increasing level of detail
for identifying numerical boundaries of the simulation arising from higher level of modelling detail as well as the
degree of modelling effort necessary for investigating certain aspects.
Modelling Stages. The top-down modelling process is divided into three stages. The first stage consists of the
basic mechanical and electrical model including an asynchronous motor for the main drive and servomotors for
automatic feed, both represented by linear equation models with parameters extracted from available data sheets.
The overall model also includes mechanical loads and a simple calculation model of the cutting forces using
common formulas and parameters [3]. However, a simple motor control implementation limits possible simulation
scenarios in this stage.
The second stage also takes into account accurate motor control parts, power electronics and waste heat calculations
in the drive motors and other components, therefore including thermal aspects in the model. In addition,
some adjustments have to be made due to numerical boundaries in order to assure sufficient performance of the
simulation and quality of the numerical solution.
The simulation model in the last stage will furthermore include remaining electrical loads, generated heat in the
cutting process and additional mechanical details.
Conclusion. Object-oriented modelling offers practical ways for structured multi-domain modelling of machine
tools leading to modular, easily refinable and modifiable simulation models. However, this approach leads to
comparatively complex models with a larger amount of equations, which can profoundly affect the performance
during simulation.
Future work will focus on refining the models as part of the next stages in the top-down modelling process as well
as parameterization of the remaining components. For model validation it is planned to compare the simulation
results of various scenarios against measurement data obtained from the turning lathe.