Please use this identifier to cite or link to this item:
https://www.um.edu.mt/library/oar/handle/123456789/74454| Title: | Improvement of sheet metal cutting using a laser cutting machine |
| Authors: | Cassar, Luke (2020) |
| Keywords: | Sheet-metal work Metal-cutting Laser beam cutting |
| Issue Date: | 2020 |
| Citation: | Cassar, L. (2020). Improvement of sheet metal cutting using a laser cutting machine (Bachelor's dissertation). |
| Abstract: | Laser cutting has become widely used for both prototyping and high-volume manufacturing in most sheet-metal processing applications. This is due to the relatively higher precision, consistency, cut quality and the capability for automation of this process. However, these beneficial qualities are dependent on many factors, including not just the laser cutting process parameters, but also the sheet material and thickness, such that when cutting different materials and thicknesses, the process parameters must be altered accordingly to achieve a successful or optimal laser cut. In practice, laser machine manufacturers provide preset parameter combinations for different materials in general such as stainless steel, mild steel etc. However, parameters for more specific applications such as cutting stainless steel 304 with compressed air using continuous cutting mode are not available. Consequently, using these general parameters provided by the laser machine manufacturer for more specific applications is suboptimal. Thus, the main objective of this study is to find the optimal process parameters that will maximise laser cut quality when cutting a 5mm stainless steel 304 sheet using the Bystronic Byspeed 3014 CO2 laser cutting machine. This was achieved through several steps. Firstly, the metrics that define cut quality were determined and prioritized. Secondly, the factors that affect these quality metrics the most were determined and prioritized. Thirdly, design of experiments, specifically the Box Behnken Design method, which is capable of creating a second-order model, was used to efficiently sample data on the laser cutting process when cutting a 5mm ss-304 sheet. Fourthly, this data was used to model the process using a hyper-parameter optimized shallow neural network for each quality metric, and the models where validated. Lastly, a multi-objective genetic algorithm was used in conjunction with the neural network to find the optimal solutions for the quality metrics and basic guidelines for the company that allowed us the use of their Byspeed 3014 laser cutting machine, where created. |
| Description: | B.ENG (HONS) |
| URI: | https://www.um.edu.mt/library/oar/handle/123456789/74454 |
| Appears in Collections: | Dissertations - FacEng - 2020 Dissertations - FacEngIME - 2020 |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 20BENGME011 - Luke Cassar.pdf Restricted Access | 3.19 MB | Adobe PDF | View/Open Request a copy |
Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.
