Raising Accuracy in Physically Based Simulations Through Scaling Equations

Abstract

In the recent years, the physically based simulation has been developed and applied to various engineering processes. So far the use of this simulation method was limited to calculate the behavior of objects with large dimensions, as the calculation of small objects leads to severe inaccuracies. Thus, simulation results for small objects cannot be used in the engineering process. However, technical systems often consist of a variety of small functional components and workpieces. This paper proposes a new method to significantly improve the accuracy of physically based simulations of small objects by scaling. First, a set of scaling equations is introduced, which allow physically correct scaling of dynamic rigid body systems. Second, the equations are validated by simulating a cube with an edge length of only 20 μm. In this simulation scenario, the new method is compared to the conventional, nonscaling physically based simulation and the improvements of the simulation results are examined. With the scaling equations, technical systems of small components and workpieces can virtually be tested and optimized. This affects a significant reduction of hardware based time and cost consuming experiments.