Abdul Khalim, Abdul Zuhair (2020) Artificial Neural Network Modelling To Predict Laser Micro Grooving Quality Of Commercially Pure Titanium Grade 2. Doctoral thesis, Universiti Teknikal Malaysia Melaka.
![[img]](http://eprints.utem.edu.my/style/images/fileicons/text.png) |
Text (24 Pages)
Artificial Neural Network Modelling To Predict Laser Micro Grooving Quality Of Commercially Pure Titanium Grade 2.pdf - Submitted Version Download (879kB) |
|
Text (Full Text)
Artificial Neural Network Modelling To Predict Laser Micro Grooving Quality Of Commercially Pure Titanium Grade 2.pdf - Submitted Version Restricted to Registered users only Download (5MB) |
Abstract
Grooving is the process of making a narrow channel on a surface of flat or cylindrical workpiece. Often it is performed on workpiece shoulders to ensure the correct fit for mating parts. Groove is widely used in automotive industry, biomedical implants and electronic devices. Though, laser machining is known as advanced machining process is an alternative machining used by high precision industries as cutting tool. Due to its ultra- flexibility, high quality end product, tight tolerance, quick set up, high repeatability, and many other advantageous, it is being widely accepted and explored for it potential in machining industries. Unfortunately, machining of groove especially micro-groove by traditional machining is very much challenged in dimensional control due to mechanical contact. A micro groove is a basic geometric feature of a micro part. The disadvantages of traditional machining become much severe when it deals with micro-grooving where, not only the external dimensions, but also the metallurgy of the end product is being affected. Although laser lathing is available, it is very complex and expensive where most commonly used in the industries are flat stock laser machining. Thus, an existing 4 by 8 feet 3KW CO2 flatbed laser cutting machine has been transformed to perform laser lathing. The transformation of 2D flatbed laser cutting machine into 3D laser cutting ability is explored of its performance in producing micro-grooves on a Titanium Grade 2. This research work presents the modelling study of micro-grooving in laser machining of commercially pure titanium grade 2 material with CO2 laser by considering the power, gas pressure, cutting speed, depth of cut and focal distance as the designed process parameters. This research focuses on experimental of laser micro-grooving quality and development of artificial neural network (ANN) model. The experimental plans were conducted according to the design of experiment (DOE) to accommodate full range of experimental analysis. Therefore, three significant responses namely groove depth, groove width and groove corner radius were investigated to fall within desired values. Analysis found that an experimental error such as the discovery of an unknown effect, inherent variability in the system, inability to control complex variables, temperature or unexpected mechanical machining tolerance have influenced the predictive model. Two types of model are introduced which are namely singleton output model and multiple output model. The results indicate that, there is a difference of 2% between the two models. Therefore the significant error in predictive model occur due to the factors of pattern recognition. However, the developed ANN model is found valid through data testing where the mean absolute percentage error (MAPE) is less than 20 per cent.
| Item Type: | Thesis (Doctoral) |
|---|---|
| Uncontrolled Keywords: | Micromachining, Artificial Neural Network, Titanium Grade 2 |
| Subjects: | T Technology > T Technology (General) T Technology > TJ Mechanical engineering and machinery |
| Divisions: | Library > Tesis > FKP |
| Depositing User: | F Haslinda Harun |
| Date Deposited: | 07 Dec 2021 13:41 |
| Last Modified: | 07 Dec 2021 13:41 |
| URI: | http://eprints.utem.edu.my/id/eprint/25427 |
| Statistic Details: | View Download Statistic |
Actions (login required)
 |
View Item |