Mehrkörpersimulation : Mit dem Computer Bewegungen von komplexen artikulierten mechanischen Systemen voraussagen
Die Mehrkörpersimulation stellt ein Teilgebiet der Mechanik dar, das sich mit dem Bewegungsverhalten von mechanischen Systemen beschäftigt, die aus mehreren, gelenkig miteinander verbundenen Körpern bestehen. Typische Beispiele sind schnell laufende Getriebe in der Textilindustrie, Fahrwerke von Flugzeugen, Roboter, aber auch komplette Fahrzeuge auf Schiene und Straße bis hin zum menschlichen Bewegungsapparat. Eine Herausforderung der Mehrkörpersimulationen besteht darin, Versuche am realen System teilweise oder wenn möglich ganz zu ersetzen. Dies ist zunehmend von Bedeutung, da Versuche oft zu teuer, zu zeitaufwendig, zu gefährlich oder nicht mit der erforderlichen Präzision durchführbar sind. Der folgende Beitrag erläutert einige Beispiele der Mehrkörpersimulation und deren Nutzen.
Multibody simulation is a branch of mechanics devoted to the prediction and analysis of the motion of systems consisting of many interconnected bodies undergoing large motions. Examples include robots, vehicles, textile mechanisms, and heavyweight machines; but also the extremities of the human body. Typical to these applications is that the basic laws underlying the physical effects have been known for some time; such as the “geometry of motion” in the area of kinematics, or the corresponding Newton and Euler Laws in the area of dynamics. However, their combination and numerical solution may pose huge problems in deciding whether or not a method is applicable in practise. For example, typical problems involve thousands of non-linear equations whose traditional solution is not possible within an acceptable timeframe. Instead, by careful investigation and exploitation of the underlying dependencies, one can devise compact solutions, which make their use in control systems or design environments feasible. This involves applying methods from mathematics, from modelling in mechanics in the application field and from computer science. This multifaceted approach has led to the development of this area into a fascinating and diverse interdisciplinary field. The examples shown in this article illustrate the broad spectrum of possible applications; ranging from rail vehicle dynamics, loading conditions of 200-ton mine excavators, design environments for roller coasters and robot motion simulators. There are also applications in biomechanics such as test devices for the human cervical spine, prediction systems for human gait and virtual microscopes for the reconstruction of internal bone movement out of non-invasive measurements of the human arm. The methodology itself was developed in the 1960s to predict how satellites and space vehicles would behave once they were irreversibly thrust into space. In the 1980s, it was employed widely and at enormous cost in the motor industry to replace some of the more expensive and riskier experimental tests. As can be seen from the article, this methodology has, in recent years, become an important engineering instrument in the development of new products and processes and the optimisation of existing ones, including medical operations and methods of diagnosis. As such, it has become a fixed component in the canon of engineering tools, paving the way for new solutions which never would have been possible without the aid of this method.