Shape Error Modelling and Analysis by Conditional Simulations of Gaussian Random Fields for Compliant Non-Ideal Sheet Metal Parts

Abstract

© 2018 The Authors. Published by Elsevier B.V. Accurate modelling of geometric and dimensional errors of sheet metal parts is crucial in designing correct GD&T and preventing unnecessary design changes during the development and launch of a new assembly process. A novel conditional simulation based methodology to probabilistically model and generate non-ideal sheet metal part geometric variations is developed. The methodology generates part geometric variations, which accurately emulate part fabrication process in terms of covariance of generated deviations. The methodology uses as inputs one or more of the following: Measurement data of current parts, historical measurements of similar parts or FEM-based simulations. The proposed methodology emulates real processes and products accurately by generating non-ideal part representatives based on the aforementioned input data. Results provide an easy engineering interpretation to the designer. The methodology is demonstrated using automotive door hinge reinforcement.

Publication
Procedia CIRP