A Set-Based Design Method for Material-Geometry Structures by Design Space Mapping

Abstract

The objective of this work is a multiscale, set-based design method for mechanical components and their manufacturing processes and materials with interactive identification of feasible design regions. A unique aspect of the proposed method is the ability to adjust both material properties, through process planning, and part geometry, through exploration of various cellular structures (e.g., lattices, honeycombs), in order to achieve design goals. More specifically, the proposed design method can effectively explore the achievement of desired mechanical properties by controlling process conditions, material properties, and/or selecting appropriate feature configurations. Material process-property relationships and cellular structure structure-property relationships are modeled to create a comprehensive multiscale design space mapping database. Bayesian network classifiers enable real-time “inverse mapping” that identify feasible regions of the unit cell, feature, and process spaces, thus providing immediate feedback to the designer regarding the feasibility of the current component design. The design method is demonstrated on a class of mechanical components that are manufactured by electron-beam melting (EBM) in Ti-6Al-4V.