- Title
- Adaptive Neuro-Fuzzy Inference System modelling of surface topology in ultra-high precision diamond turning of rapidly solidified aluminium grade (RSA 443)
- Creator
- Zvikomborero, Hweju
- Subject
- Mechatronics Surface roughness -- Measurement
- Date Issued
- 2020
- Date
- 2020
- Type
- Thesis
- Type
- Masters
- Type
- MEng
- Identifier
- http://hdl.handle.net/10948/49441
- Identifier
- vital:41721
- Description
- Surface roughness prediction is a crucial stage during product manufacturing since it acts as a quality indicator. This investigative research thesis presents an online surface roughness prediction, based on the Adaptive Neuro-Fuzzy Inference System (ANFIS) model during Ultra-High Precision Diamond Turning (UHPDT) of Rapidly Solidified Aluminium (RSA-443) using water and kerosene as coolants. Based on the Taguchi L9 orthogonal array, the cutting parameters (spindle speed, depth of cut and feed rate) are varied at three levels. Acoustic Emission (AE) signals are detected during the UHPDT process using a piezoelectric sensor. Spindle speed, depth of cut, feed rate, AE root mean square, prominent frequency and peak rate are considered as model inputs in this thesis. The experimental results reveal that a better surface finish is obtained using water coolant in comparison to kerosene coolant. Mean Absolute Percentage Error (MAPE) based comparison between ANFIS and Response Surface Method (RSM) is carried out. In this study, the ANFIS model has a prediction accuracy of 79.42% and 69.40% on water-based and kerosene-based results respectively. The RSM model yields higher prediction accuracies of 98.59% and 95.55% on water-based and kerosene-based results respectively.
- Format
- xxi, 141 leaves
- Format
- Publisher
- Nelson Mandela Metropolitan University
- Publisher
- Faculty of Engineering, the Built Environment and Technology
- Language
- English
- Rights
- Nelson Mandela Metropolitan University
- Hits: 1286
- Visitors: 1426
- Downloads: 192
Collections
NMU School of Engineering
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | SOURCE1 | Hweju, Z.pdf | 4 MB | Adobe Acrobat PDF | View Details Download |