Experimental investigation to predict the condition of cutting tool by surface texture analysis of images of machined surfaces based on amplitude parameters

Abstract

In this paper, an experimental investigation is presented for accomplishing surface texture analysis using machine vision-based system for predicting the condition of cutting tool. Texture of machined surface provides reliable information regarding the extent of the tool wear because tool wear affects the surface roughness dramatically. Analysis of machined surface images of different materials by turning process at different wear conditions cutting tool are grabbed using CCD camera are presented. In this paper, we propose an amplitude parameters based approach for analysis of machined surfaces. Machined surfaces with different wear conditions of the cutting tool,that is, sharp, semi-dull and dull are investigated by using surface metrology software TRUEMAP and also with conventional method using stylus instrument for comparative purpose. Since a machined surface is the negative replica of the shape of the cutting tool, nd reflects the volumetric changes in cutting edge shape, it is more suitable to analyse the machined surface than to look at a certain portion of the cutting tool. However, considerably less work has been performed on the development of surface texture of machined workpiece that provide information on the condition of the cutting tool, employed in machining the surface. In this paper, a non-contact method using machine vision for inspecting surface roughness of machined surfaces produced by varying conditions of turning process is studied to monitor and to predict the cutting tool condition has been presented. Through our experiments, we found a strong correlation between tool wear and surface roughness (surface texture) of the machined surfaces. Results proved that the approach is effective in predicting the condition of the cutting tool through amplitude parameters.