INTROSPECTING THE THERMAL EFFECTS OF RADIOGRAPHIC WELD IMAGES

Abstract

ABSTRACT Welding is the experienced way of joining metals together. It is an efficient and economical process. It is an assembling process where materials are joined. It is achieved by adding some additional molten joining material on melting part of the materials to be joined. It forms strong bond when the molten material is cooled. It plays a major role in industries for the purpose of construction, joining and repairing of steel beams, reinforcing rods in buildings, bridges, spacecraft, pipe lines, nuclear containers etc. During the process of welding a number of different types of discontinuities can be produced, which may arise due to material inconsistencies of the material, error produced by the operator, or other factors that are beyond the operators control. Irrespective of the source of error, detection of discontinuities is critical. An unacceptable weld extremely reduces the bond between two materials and may cause failure. Various flaws occur in weld images. The flaws are as follows: gas cavity, Lack of Penetration, Porosity, Slag, cracks, lack of fusion,. Worm Hole, Under Cut. In the present work, in total 79 images have been considered. The image has been collected from Bharat Heavy Electricals Limited (BHEL), Haridwar and Indian Institute of Technology Roorkee. An attempt has been made to collect the feature of these images using wavelet transform. The extracted feature are thus subjected to Artificial Neural Network (ANN) and almost 89.88% of the images are classified. An algorithm has been developed and the results obtained are promising.