Title: Global Effects of Interactions Between Instances in N-Body-



Name: Dipl.-Ing. Felix Claus



Project description:


Starting situation

Nowadays, simulations are a very important component within Product-Development-

Processes. In many production procedures, a great potential for real-time controlling due

to the use of simulations exists. Simulations can replace physical processes to a great

extent. As such, industrial corporations can accelerate their complete supply-chain by

replacing time and money consuming processes in early planing stages. Furthermore,

the final simulation results provide advanced information of various physical quantities,

thus, the amount of correction loops, necessary in design processes, can be reduced. For

example, tensions and forces can be determined without costly physical experiments and

measuring machines. Nowadays, almost all types of problems are solved with the support

of simulations. The economy of time and money as well as the nomerous possibilities in

analysis processes is an enormous advantage for development and production processes.

But therefore, the simulation has to be precise and fast. Thus, the benefits of reducing

computational costs for high detailed simulations becomes crucially important.




However, to archive the performance of running real-time simulations, the simulation

model has to be optimized. The most common type of simulations is the Finite-Element-

Method (FEM) which, for instance, can be used to solve problems in the field of structural

mechanics. The availability of powerful computing systems, allows simulations with a

high degree of detail, more complex structures, and more precise models. By that the

industrial processes are getting progressive digitalized. In terms of "Industrie 4.0" simulations

are taking a major role.



Expected Results

In the course of this studies global effects of interactions between instances in N-Body-

Simulations will be examined. Therefore, with the use of FE-Simulation methods it will

be investigated how mechanical interactions between many bodies can be simplified while

keeping a high degree precision. The goal of this research is the development of methods,

that allow precise FE-Simulations with a user-chosen level of detail for each interaction

individually. These methods are intended to be generally applicable and focus on the

consideration of global effects. Therefore, a simplified model of interaction between instances

(bodies) is needed, whereas the impacts due to the simplification on the final

results need to be analyzed. Based on this studies, an efficient and precise model will be

created that can be solved in an acceptable amount of time.