Surface Roughness Prediction in Turning of Monel K 500 using DWT Technique

Abstract

Research on specimen surface roughness is crucial because of its impact on machined components' functionality. In the meanwhile, a vision system is a cutting-edge method that is becoming more popular for measuring pictures of the specimen to determine the roughness of the machined surface. Scanning electron microscope (SEM) images of the machined surface are captured by a vision system for this study. During the final turning process, two-dimensional pictures of the machined surface of the Monel K 500 alloy are used to estimate the profile of the surface of specimens. Surface roughness of simulated specimens was investigated by image analysis under different simulated machining settings. This study employs a method for identifying surface texture that combines the acquisition of 2D surface pictures with a wavelet transform-based strategy. The Two-Dimensional Wavelet Transform may be utilized in the process of assessing surfaces due to its ability to deconstruct an image of a machined surface into a multi-resolution representation of that surface's multiple attributes. Prediction errors of less than 1.674% were obtained by analysing the histogram frequency difference of a lit area of interest (ROI) in images of rotated surfaces.