Multidisciplinary Optimization of Geometric Variation and Modal Behaviour for Squeak and Rattle Prevention

Abstract

Squeak and Rattle are two undesired phenomenon that occurs when adjacent parts come into contact with each other either by sliding or impacting. The Squeak and Rattle directly affects the end user’s perception towards the quality of the car which is not desirable for any automotive company. One of the main reason for the generation of Squeak and Rattle in a car is geometric variations which happen due to manufacturing tolerances, temperature etc. With modification in attachment points in an assembly, many other attributes like modal behaviour of the component are affected. This thesis aims to combine the design optimization process for geometric variation and modal behaviour for Squeak and Rattle prevention. The study is carried out on a passenger car door and deco panel-IP assembly. Geometric variation analysis is performed in CAT software package of RD&T whereas the modal behaviour of an assembly is determined by doing FEM analysis. To couple the optimization process for two different disciplines, modeFRONTIER is used which is an integration platform for multi-objective and Multi Disciplinary Optimization (MDO). Initially, single discipline is framed in the optimization workflow to understand the optimum or best attachment points in an assembly. Further, MDO workflow is created and the results were compared with deco upled optimization to analyze its effects on the selection of optimum attachment points. Multi Objective Genetic Algorithm (MOGA-II) is used as the optimization algorithm to find the best optimal positions within an assembly. The results show that the two objectives from MDO optimization workflow are conflicting with each other wherein the best solutions lie on a Pareto front. The MDO methodology can be applied for different types of geometries and attachment schemes. No experimental validation was done and could be considered as future scope of work.