首页|Data on Machine Learning Described by Researchers at University of Technology Brunei (Optimization of Surface Roughness, Phase Transformation and Shear Bond Strength In Sandblasting Process of Ytzp Using Statistical Machine Learning)
Data on Machine Learning Described by Researchers at University of Technology Brunei (Optimization of Surface Roughness, Phase Transformation and Shear Bond Strength In Sandblasting Process of Ytzp Using Statistical Machine Learning)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
Elsevier
Data detailed on Machine Learning have been presented. According to news reporting out of Gadong, Brunei, by NewsRx editors, the research stated, “Sandblasting process is often applied to roughen the intaglio of yttria tetragonal zirconia polycrystals (YTZP) surfaces for easy and quality adhesion and micro-shear retention with dentine/resin cements. Sandblasting process parameters have shown to influence, at different scales, surface roughness, phase transformation and shear bond strength, all of which are referred, herein, as performance characteristics.” Our news journalists obtained a quote from the research from the University of Technology Brunei, “This study aimed to find the parametric settings of sandblasting parameters that could simultaneously optimize these performance characteristics, hypothetically testing the probability. YTZP surfaces were sandblasted at different levels of incidence angle (IA), abrasive particle size (AP), pressure(P) and sandblasting time (ST) following the Taguchi method based on the two-level parametric process settings (L8(27)). Surface morphologies, roughness (SR), monoclinic content (MC), and shear bond strength (SS) were characterized by the SEM, average surface roughness, XRD, and shear bond strength tests, respectively. Rough surfaces containing scratches, plastic deformation streaks, micro cracks and pitting were observed. According to the Taguchi method, the same optimum sandblasting parametric setting maximized SR and MC but failed to maximize SS. Subsequently, the principal component analysis embedded in statistical machine learning was applied to find the optimum sandblasting parametric setting that maximized all the performance characteristics. The optimum sandblasting setting of IA = 45 degrees, AP = 110 mu m, ST = 20 s and P = 400 kPa predicted the maximum values of SR = 0.773 mu m, MC = 36% and SS = 16.6 MPa. Analysis of variance confirmed AP and P as the most influencing parameters affecting all performance characteristics.”
GadongBruneiCyborgsEmerging TechnologiesMachine LearningUniversity of Technology Brunei
NETL
NSTL
Elsevier
