Skip to main navigation Skip to search Skip to main content

Modélisation de la complexité géométrique des composants mécaniques

Translated title of the thesis: Modeling of geometrical complexity of mechanical parts
  • Badreddine Abouchalghouma

Student thesis: Master's thesisMaster in Engineering: Mechanical Engineering

Abstract

The geometrical complexity of the mechanical parts is a key parameter that intervenes in the process capability to ensure the manufacturability of products. Nevertheless, and to determine whether the process is able to manufacture a part having a certain geometry, it is necessary that the latter is actually manufactured. In this thesis and in the continuity of the work done previously, we present an approach for quantifying the geometrical complexity of parts produced with the automated fiber placement process AFP. Quantifying this complexity is based on the geometrical analysis of a surface laminated taking into account the fiber placement direction. Many candidate metrics characterizing the geometrical complexity will be implemented and their distribution on a truly manufactured part by our industrial Partner will be compared to the profile deviations recorded during the inspection of the plies constituting the part taken as a reference. This task will be done to identify the final index reflecting the degree of complexity of parts produced with the AFP process. Quantifying the geometrical complexity will give us the opportunity, by design, to estimate the profile tolerances to be applied to new parts having different geometries but from the same AFP process. This goal will be achieved by drawing on previously developed methods based on order statistics and which allow to have a model linking the geometrical complexity and the process capability while meeting the requirements of the standard (ASME Y14.5, 2009).
Date23 Jun 2016
Original languageFrench
Awarding Institution
  • École de technologie supérieure
SupervisorAntoine Tahan (Supervisor) & Roland Maranzana (Co-supervisor)

Cite this

'