Fast performance assessment of mechatronic designs integrating CAD and dynamical models with application on servo actuated designs

In this paper an approach is presented that allows to investigate, in a systematic way, how design parameters of a servo actuated system modeled in CAD influence the dynamic behaviour and performance of the system. This is achieved by extracting physical quantities such as mass and inertia in part designs as these are difficult to model analytically when having irregular part shapes. These values are then employed in a symbolic dynamic model to assess the behavior and associated performance through numerical integration. To illustrate the effectiveness of the integrated approach, different slider-crank designs are evaluated for a given motion path. A sensitivity analysis is performed on the basis of gradients extracted from the introduced symbolic dynamical model through algorithmic differentiation. Our approach is flexible and enables precise motion study of actuated mechanisms with calculation times improved by an order of magnitude compared to other methods. In addition, this novel approach depends only on open source software.