Abstract—Surface roughness is a crucial parameter for mechanical products. To achieve small surface roughness, the grinding method is often chosen as the final machining process. The regression model of surface roughness forms the basis for controlling the grinding process and predicting surface roughness under specific conditions. The effectiveness of process control and the accuracy of predicted surface roughness depend on the precision of the surface roughness regression model. This study aims to enhance the accuracy of the surface roughness regression model by employing square root transformation. An experimental process was conducted with a total of eighteen experiments. In each experiment, three cutting parameters, including workpiece speed, tool feed rate, and cutting depth, were varied. Surface roughness was measured in each experiment. After conducting experiments, a surface roughness regression model was established, denoted as Model (1), without using any data transformation. The square root transformation was applied to convert the surface roughness dataset into another set of data. From this dataset, another surface roughness model, referred to as Model (2), was developed. Both models were used to predict surface roughness, and the predicted results were compared with the actual surface roughness in the experiments. Four parameters were used to compare Models (1) and (2), including the coefficient of determination (R-Sq), adjusted coefficient of determination (R-Sq(adj)), mean absolute error percentage (%MAE), and mean squared error (%MSE). All four parameters for Model (2) were superior to those for Model (1). The results confirmed that the square root transformation successfully improved the accuracy of the surface roughness regression model in grinding applications.

Keywords—surface roughness, grinding, square root transformation

Cite: Do Duc Trung and Nguyen Trong Mai, "Improving the Accuracy of the Surface Roughness Model in Grinding Through Square Root Transformation," International Journal of Mechanical Engineering and Robotics Research, Vol. 13, No. 2, pp. 249-253, 2024.

Copyright © 2024 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.