Prediction of Process Parameters of Ultrasonically Welded PC/ABS Material Using Soft-Computing Techniques

Autor: Natarajan Prabaharan, Hassan Haes Alhelou, P. Pandiyan, S. Saravanan, T. Chinnadurai, M. Karthigai Pandean
Rok vydání: 2021
Předmět:
Zdroj: IEEE Access, Vol 9, Pp 33849-33859 (2021)
ISSN: 2169-3536
Popis: Welding process is found to be a predominant procedure in most of the processing industries, especially in the automobile sector for maintenance operation and fabrication. Ultrasonic Polymer Welding (USW) is used for the joining process because of its flexibility and short needed welding time. In this article, two different polymer materials PC and ABS are blended in the ratio of 60:40 and molded into a sheet. Furthermore, molded PC/ABS sheets are joined using USW with different processing parameter settings. Three major influencing process parameters like pressure (P), amplitude (A) and weld time (Tw) are considered and other processing parameters are kept at constant. The experiment is carried out for 26 welded samples and from the obtained results it is noticed that the above-mentioned process parameters directly influence the tensile strength of welded joints. Additionally, the ultrasonically welded samples tensile strength is analyzed with the help of Artificial Neural Network technique (ANN) and Adaptive Neuro-Fuzzy Inference System method (ANFIS). From the simulation results, an optimized ANFIS model provides the superior result as compared to ANN. Moreover, Scanning Electron Microscope analysis is carried out to visualize the weld interface between the joint. Also, Finite Element Modeling (ANSYS) is performed to understand the heat dissipation during the welding process.
Databáze: OpenAIRE