Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/84699
Title: Fixture configuration design for sheet metal assembly with laser welding
Authors: Li, Bing
Degree: Ph.D.
Issue Date: 2001
Abstract: Sheet metal laser welding has gained acceptance for meeting the increasing demands for high weld quality in the assembly process due to its economic advantage and small thermal distortion caused. However, the application of laser welding in industry is still limited since many aspects of laser welding still require further investigation. Among such aspects metal fit-up is an important factor that affects the implementation of laser welding as well as the improvement of weld quality. Current stamping processes cannot meet the metal fit-up specification that laser welding requires. Assembly fixture will play a critical role in both control of assembly variation and improving Degree of Metal Fit-up(DMF). With the introduction of Optical Coordinate Measuring Machine (OCMM) and the application of flexible fixtures, a new assembly system using OCMM and flexible fixtures for laser welding is proposed. The variational information of the locating areas on the assembled parts can be obtained by OCMM. This information can be fed into the in-process fixture design module. Different cycles of the same assembly will result in different fixturing schemes that can be accommodated by adjusting the flexible fixtures. The in-process fixture design module is an important part that determines whether the acceptable metal fit-up can be obtained. This thesis primarily focuses on this module. In order to simplify the current complicated fixturing procedure a new locating scheme with both total locating and direct locating for welds is proposed. The total locating scheme is used to locate the entire sheet metal assembly. The direct locating scheme is used to locate the weld joints to meet the metal fit-up requirement. A finite-element(FE) model and a prediction and correction method for direct locator configuration are developed in the general fixture design. By setting the weld nodes to nominal values the nodal variation of the locating area can be obtained. Then the locating nodes can be grouped by setting different variational values that are obtained from FEA. Thus the feasible fixturing scheme can be obtained from applying the right group of direct locators. In the optimization for fixture configuration design, both number and location of the locators concerned are taken as design variables. Considering the DMF to be a fuzzy quantity, a feasible evaluation criterion about DMF is also developed using a fuzzy synthesis evaluation method. Genetic algorithm(GA) is employed as the optimal procedure. In order to reduce the sensitivity of the performance characteristic of DMF to the location fluctuation of designed locators, a two-stage response surface methodology is developed for the robust fixturing design. The first stage is to optimally find the Robust Design Space (RDS) where a relatively insensitive area can be reached. An optimal objective combining both robustness and performance is derived. Within the RDS a second-order response surface model is fitted by a 3k fractional factorial design in the second stage. Traditional experience-based determination of the weld pattern is easy to cause unexpected discrepancy from the quality requirements of the assembly weld. A new design approach of the assembly weld pattern is developed. A case study with the cross members assembly is presented, which demonstrates the feasibility of the proposed fixture design principles in industrial application. The results of the case study show that the proposed fixturing methodologies are reasonable and effective in meeting the requirement of sheet metal assembly with laser welding.
Subjects: Sheet-metal
Laser welding
Hong Kong Polytechnic University -- Dissertations
Pages: xii, 138 leaves : ill. ; 30 cm
Appears in Collections:Thesis

Show full item record

Page views

59
Last Week
0
Last month
Citations as of Mar 24, 2024

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.