Automatic Fixture Design Based on Formal Knowledge Representation, Design Synthesis and Verification
Author: Gmeiner, Thomas
Supervisor: Lindemann, Udo
Institution: Technische Universit
The fabrication of parts in manufacturing systems requires their fixation in a defined position for machining, assembly or inspection processes using fixture devices. Especially for low-volume production and the resulting frequent change-overs, as commonly found in modern manufacturing scenarios, the design and configuration of fixtures have a strong influence on the production time and costs. To address these tasks and increase the flexibility and degree of automation of manufacturing systems, flexible fixture devices and computer-aided or automated fixture design methods are developed. While flexible fixture devices are well established in industry nowadays, no general and commonly used methods, applications or systems for computational fixture design exist so far. This is due to the fact that the involved tasks are wide-spread, knowledge intensive and strongly intertwined with other steps of the design-to-fabrication process. Further, existing computer-aided fixture design systems are rarely integrated with the configuration or embodiment of flexible fixtures on the hardware level.
To address these issues, an automatic fixture design system integrated with a reconfigurable fixture device is presented in this thesis. The system is based on a formal ontology that serves as an integrated knowledge representation and basis for the generation of fixture design candidates through semantic reasoning. A spatial grammar is used for the synthesis of new problem-specific fixture components and design candidates. A tool-fixture interference analysis is used for fixture design verification and a workpiece deformation analysis for evaluation of the design candidates. Prototype software implementations for all methods are developed and their applicability to automate the design and drive the reconfiguration of a vise-type fixture device is evaluated by processing a set of case-study parts. The results show that the approach and system presented allows for (semi-)automatic fixturing of parts of different and non-predefined geometries and, thus, for an increase of the degree of flexibility and automation of a manufacturing system. The work further contributes to the areas of ontological fixture design knowledge representation, the use of spatial grammars for detailed engineering design tasks, the automation of cutting tool interference analysis based on NC codes and the automation of finite element analysis in the area of fixture design verification.