Abstract
The surface finish of industrial components has an important role in their performance and lifetime. Therefore, it is crucial to find the cutting parameters that provide the best surface finish. In this work, an experimental study of the effect of cutting parameters on ultra-high molecular weight polyethylene (UHMWPE) by a turning process was carried out. Today, the UHMWPE polymer continues to find applications mainly in the automotive industry and biomechanics because it is resistant to impact and corrosive materials to use. A face-centered Central Composite Design (CCD) and Response Surface Methodology (RSM) were applied to evaluate the influence of the cutting speed (Vc), feed rate (f) and depth of cut (ap) of the turning operation on the Average Surface Roughness (Ra) and Material Removal Rate (MRR). Results allowed obtaining an adjusted multivariable regression model that describes the behavior of the Ra that depends on the cutting parameters in the turning process. The predictive model of Ra showed that it fits well with a correlation coefficient (R2) around 0.9683 to the experimental data for Ra. The ANOVA results for Ra showed that the feed is the most significant factor with a contribution of 42.3 % for the term f 2, while the speed and depth of cut do not affect Ra with contributions of 0.19% and 0.18%, respectively. A reduction of feed from 0.30 to 0.18 mm·rev−1 produces a decrease in surface roughness from 6.68 to 3.81 μm. However, if the feed continued to reduce an increase in surface roughness, a feed of 0.05 mm·rev−1 induces a surface roughness of 14.93 μm. Feeds less than 0.18 mm·rev−1 cause a heat generation during turning that increases the temperature in the process zone, producing surface roughness damage of the UHMWPE polymer. Also, the results for MRR demonstrated that all of the cutting parameters are significant with contributions of 31.4%, 27.4% and 15.4% to feed, speed, and depth of cut, respectively. The desirability function allowed optimizing the cutting parameters (Vc = 250 m·min−1, ap = 1.5 mm y f = 0.27 mm·rev−1) to obtain a minimum surface roughness (Ra = 4.3 μm) with a maximum material removal rate (MMR = 97.1 cm3·min−1). Finally, the predictive model of Ra can be used in the industry to obtain predictions on the experimental range analyzed, reducing the surface roughness and the manufacturing time of UHMWPE cylindrical components.
A surrogate-assisted optimization approach for multi-response end milling of aluminum alloy AA3105
