ABSTRACT
In this study, friction welded joints are monitored using machine vision system and the weld quality is predicted using 2D wavelet transformation method. Here, the two weld parameters mainly, welding speed and rotational speed are varied in three levels and nine experimental runs are obtained according to DoE techniques. Top surface weld images are captured and converted to binary images. Coefficients from the approximations of images are computed using 2D wavelet transformation. This is compared with the experimental results of UTS tests of the welded samples. It is observed that the with the increase in the proposed indicator, ultimate tensile strength of the joints is found to follow a decreasing. This paper is an attempt to implement the methodology used for friction stir welding in the quality prediction of rotary friction welded joints.
