1. Professorship of Production Systems and Processes, Mechanical Engineering, University of Technology Chemnitz, Chemnitz, Germany
2. Fraunhofer Institute for Machine Tools and Forming Technology, Chemnitz, Germany
Email: mei-yun.liu@mb.tu-chemnitz.de (M.Y.L.); holger.schlegel@mb.tu-chemnitz.de (H.S.);
martin.dix@mb.tu-chemnitz.de (M.D.)
*Corresponding author
Manuscript received December 8, 2023; revised January 10, 2024; accepted March 8, 2024; published May 14, 2024
Abstract—To achieve desired surface properties, various mechanical processes are used, including mechanical blasting, a technique involving the high-pressure projection of grains onto a surface. This study focuses on surfaces treated through mechanical blasting, specifically analyzing stainless steel components. The influence of key manufacturing parameters, such as grain shape and rotational speed, is systematically investigated across different stages. A comprehensive methodology for feature selection is presented, aiming to identify crucial roughness parameters and analyze their impact on the manufacturing process. The objective is to determine the most significant roughness parameters to establish a tailored quality control system aligned with the outcomes of mechanical blasting. This system provides targeted feedback on the manufacturing parameters, enabling precise adjustment and achieving the desired surface roughness. This approach contributes to sustainable process optimization by minimizing rework and reducing rejects 32 3D roughness parameters, defined by ISO standards, are calculated and analyzed. Based on a data set with 300 measured values, a statistical analysis was performed, which includes a correlation analysis and a regression analysis using Lasso regression for parameter selection. The results of the correlation analysis suggest that feature, functional and volume parameters seems to be important role for surface characterization. However, in further analysis by Lasso regression, the volume parameters were found to be irrelavant. In this context, the roughness parameters Spc, which represent the arithmetic mean peak curvature of surface features, and Spd, which signifies the number of peaks per unit area, stand out as notably significant and have been emphasized as the most crucial parameters.
Keywords—lasso regression, mechanical blasting, surface integrity, surface treatment
[PDF]
Cite: M. Y. Liu, H. Schlegel, and M. Dix, "Identification of the Most Important 3D Roughness Parameters for Surface Characterization for Enhanced Process Optimiztion in Mechanical Blasting," International Journal of Engineering and Technology vol. 16, no. 2, pp. 87-91, 2024.
