Deformation Analysis and Path Planning for Thermal Forming of Complex Shapes
Open Access
- Author:
- Reutzel, Edward William
- Graduate Program:
- Mechanical Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- December 14, 2006
- Committee Members:
- Panagiotis Michaleris, Committee Chair/Co-Chair
Eric M Mockensturm, Committee Member
Stephen M Copley, Committee Member
Richard Martukanitz, Committee Member - Keywords:
- thermal forming
line heating
differential geometry
laser forming - Abstract:
- Many products, such as ship hulls, require that metal plates be formed into complex curvatures. Since the 1950’s, ship builders have investigated the use of thermal forming as a flexible and cost-effective replacement for rolling, blacksmithing, and creep forming. Thermal forming involves the application of heat in order to generate permanent plastic strains in sheet or plate to produce the desired shape. Historically, the heating pattern, or location of the heat lines, has been determined by skilled artisans employing heuristic rules and experience. Automated determination heating paths that will generate the deformations required to produce a desired shape is a complex task. A critical aspect of such a system is that it must be able to predict the deformation resulting from a prescribed heat path. To produce a desired shape, an automated thermal forming system must also incorporate an algorithm that can determine the location and parameters of heat lines that will generate the required curvature. To provide fast and robust solutions to each of these problems, the following approach has been taken: 1. Develop and demonstrate an automated iterative thermal forming algorithm, loosely based on conventional manual techniques, that relies on simple empirical relations and sensor feedback to determine heat paths necessary to form a desired shape, 2. Develop differential geometry-based techniques to efficiently predict the deformation caused by application of heat lines with improved accuracy and applicability to a broader set of heating conditions than the empirical relations employed in the earlier work, 3. Apply anisotropic hierarchical adaptive meshing to improve the efficiency of standard FEA simulations, then provide a detailed comparison of this technique to the differential geometry-based analysis technique and to the baseline standard FEA solution, and 4. Develop and evaluate a robust path planning approach that relies on the novel deformation analysis approach discussed above in order to achieve accurate results for arbitrary shapes in an efficient manner. This dissertation provides a summary of these efforts, and presents recommendations to further improve the techniques developed herein.