Drill-informatics assessment of silanized Corchorus olitorius based composites applying supervised machine learning strategies
摘要
A growing apprehension about global warming has sparked a pursuit among the scientific community to produce ecological friendly substances that may mitigate carbon effects. The atmospheric and ecological features of natural fiber-based composites have had substantial consequences in the field of sustainability. This research investigates how different tool characteristics, such as feed rate and cutting speed, affect the drilling performance of composites based on silanized Corchorus olitorius particles. A number of trails are undertaken to quantify the influence of feed and cutting speed on multiple variables which involves drill force, material removal rate, roundness inaccuracy and surface roughness. The findings demonstrated that the arrangement of drilling has had an immense impact on the emergence of flaws in the vicinity of periphery hole. With an 8 mm drill bit, the lowest feed rate (20 mm/rev) and maximum cutting speed (2000 rpm) resulted in the lowest thrust force (44.25 N) and surface roughness (1.256 μm). On the other hand, maximum values occurred with greater feed rates and lower cutting speeds. In order to identify the potential drilling attributes of established composite substrate, some machine learning models have been devised. Among these models, the support vector machines model has achieved the greatest accuracy (R2 up to 0.951). These findings illustrate the composite’s viability for lightweight residential applications because of its machinability, cost-effectiveness, and satisfactory mechanical performance.
Graphical abstract