Publication

Abstract : Welding is considered as one of the most important process for the manufacturing industries. To be precise, the welding process has a supreme role to play in the automobile industry for fabrication and maintenance. Ultrasonic Welding (USW) is a very commonly used welding procedure in such industries, preferred because of its less processing time and versatility. In this study, three most influencing process parameters that includes welding pressure (P), weld time (Tw), and amplitude (A), are considered to determine the tensile strength of welded PC/ABS blend joints. The welding process is applied on a total of 26 samples and their tensile properties are obtained in terms of experimental results. It is clearly understood that all tensile strength of the joints welded by ultrasonic method is greatly influenced by all three process parameters. Further, with the help of data obtained from the experimentation, an Artificial Neural Network (ANN) based on Back Propagation model has been developed for analyzing the tensile strength of USW joints. By comparing the real time experimental results with the ANN predicted results, it is observed that the predicted and experimental models are in accordance with each other. This novel ANN model could be further employed for identifying the tensile strength of USW joints in various joining applications. Finally, the SEM images are analyzed to predict the nature of the weld condition.