Functional Morphing for Manufacturing Process Design, Evaluation and Control.

Abstract

Shape changes are commonly identified in product development and manufacturing. These changes include part shape changes in a product family from one generation to another, surface geometric changes due to manufacturing operations, etc. Morphing is one method to mathematically model these shape changes. However, conventional morphing focuses only on geometric change without consideration of process mechanics/physics. It thus has limitations in representing a complex physical process involved in product development and manufacturing. This dissertation proposes a functional morphing methodology which integrates physical properties and feasibilities into geometric morphing to describe complex manufacturing processes and applies it to manufacturing process design, evaluation, and control. Three research topics are conducted in this dissertation in areas of manufacturing process design, evaluation and control. These are: • Development of evolutionary stamping die face morphing: Similarities which are identified among parts of the same product family allow the possibilities for the knowledge learned from the die design of one generation of sheet metal product to be morphed onto that of a new but similar product. A new concept for evolutionary die design is proposed using a functional morphing algorithm. Case studies show that the proposed method is able to capture the added features in the new part design as well as the springback compensation inherited from the existing die face. • Formability assessment in die face morphing: A strain increment method is proposed for early formability assessment by predicting strain distribution directly from the part-to-part mapping process based on the functional morphing algorithm. Since this method does not require the knowledge on the new die surface, such formability assessment can serve as an early manufacturing feasibility analysis on the new part design. • Functional morphing in monitoring and control of multi-stage manufacturing processes: A functional free form deformation (FFD) approach is developed to extract mapping functions between manufacturing stages. The obtained mapping functions enable multi-scale variation propagation analysis and intermediate-stage process monitoring. It also allows for accurate inter-stage adjustment that introduces shape deformation upstream to compensate for the errors downstream.