A Modeling Approach for Mechatronic Systems Modeling and Simulation of an Elevator System Peter Schneider, Erich Huck, Peter Schwarz
Fraunhofer Institute for Integrated Circuits, Design Automation Division Dresden, Germany  this kind of formulation - eq. (2) - is needed.
Due to the complexity of mechatronic systems  there is a
strong demand for better assistance in formulating system
equations. To analyze real-world problems a powerful interdisciplinary modeling methodology covering: • a unified modeling approach,
• standardized modeling languages,
• algorithms and tools for model generation (order reduction, approximation etc.) and • properties for system optimization
Abstract -- Mechatronic systems as well as other technical systems (microsystems, distributed automation systems, ...) can be characterized as complex heterogeneous systems. Typically, they show some of the following features:
- mixed physical domains (mechanical, electrical, thermal,
fluidic, ... phenomena),
- partially close coupling between these domains,
- distributed and lumped effects or elements, respectively,
- continuous and discrete signals and systems
(in electronics: analog and digital).
Often, the modeling of continuous systems leads to very large systems of stiff differential equations. In contrast, system simulation requires simpler and faster models to investigate the interaction between all components. That’s why a powerful methodology for the modeling and simulation of mechatronic systems is demanded which considers their special characteristics. This methodology must cover a unified modeling approach, standardized modeling languages, algorithms and tools for model generation and capabilities for system optimization. In the following an approach for modeling is presented which meets these requirements. A realworld example, the modeling of an elevator system, shows the application of the modeling approach.
II. A MODELING APPROACH FOR MECHATRONIC
Elevator systems are typical examples for complex mechatronic systems. High safety-relevant demands are special properties of these systems. Practical experiments for the function validation of developed components (subsystems) and their
interactions are mandatory necessary. However, the opportunities for experimental system optimization are often limited by the system size and complexity. Experiments with system versions are too expensive or impossible because of safety requirements. Furthermore, the investigation of relevant operating conditions is very complicated. This difficult design situation can be improved by accompanying system simulations. A necessary precondition for simulation is the modeling of the system with all components.
Many physical effects in elevator systems can be formulated
mathematically by ordinary or partial differential equations. For information-processing subsystems, e.g. the control system, both the analog and the digital descriptions are needed. Therefore, simulators for analog and mixed analog-digital simulation (e.g. SABER , ELDO , Spectre) are very important. These simulators are suitable for solution of nonlinear ordinary differential equations, often with algebraic constraints. In addition, they provide opportunities for simulation of time-discrete system components, e.g. digital electronic circuits. Distributed systems can be described by ordinary differential equations after discretisation concerning the spatial coordinates. Furthermore, the above-mentioned tools provide
behavioral description languages (MAST, HDL-A, VHDLAMS ) which are suitable for modeling of electrical as well as nonelectrical systems and the mixture of time-continuous
and time-discrete systems.
A modeling methodology for complex systems can be
described as follows:
Index Terms -- mechatronics, modeling approach, differential algebraic...