Eco-friendly drilling of AA 5052-H32 Alloy: influence of jasmine-based cutting fluid on surface quality and burr Formation
This research presents a study of biodegradable-cutting fluid made by 85 wt. (%) jasmine oil and 15 wt. (%) organic petroleum-based additive as a green substitute for conventional oils in drilling AA 5052-H32 aluminium alloy. Response Surface Methodology (RSM) was applied to investigate the factors...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | International Journal of Sustainable Engineering |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/19397038.2025.2538863 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This research presents a study of biodegradable-cutting fluid made by 85 wt. (%) jasmine oil and 15 wt. (%) organic petroleum-based additive as a green substitute for conventional oils in drilling AA 5052-H32 aluminium alloy. Response Surface Methodology (RSM) was applied to investigate the factors of spindle speed and feed rate on surface quality, burr formation and temperature across three lubrication situations: dry, 90–10% and 80–20% water-to-oil mix. Results show that a surface roughness of 7.3 µm at 6370 rpm and 2867 mm/min under 80–20% and retraction rate. This lubrication regime had the smallest amount of burr height of 0.07 mm as the most conducive cooling minimum temperature of 33.8°C. In addition, machine learning models are introduced to predict surface roughness that the Gaussian Process Regression (GPR) model yields the best prediction accuracy. Additionally, state-of-the-art machine learning models were applied to the experimental data to optimise the drilling process. The Optimized Gaussian Process Regression (OGPR) model showed the highest accuracy (R2 = 0.86, RMSE = 1.33, MAE = 1.12), followed by the WNN model (R2 = 0.89), while linear regression performed the worst. These results highlight the potential of ML models in enhancing machining efficiency and sustainability in drilling. |
|---|---|
| ISSN: | 1939-7038 1939-7046 |