A Comparative Study of Predicting Burn off Length in Continuous Drive Solid State Friction Welding for ASTM A516 Steel by Regression Analysis, Fuzzy Logic Analysis and Finite Element Analysis

ABSTRACT The aim of this work is to recommend a method to model a friction welding process parameters for ASTM A516 Grade 70 alloy steel using techniques like regression analysis, fuzzy logic and finite element analysis. The techniques used in this study were used to determine the welding process variable by which the expected burn off length is obtained in friction welding. The 9 set of input parameters like friction pressure, upset pressure, forging time and rotational speed and corresponding output parameter burn off length gathered based on L9 Orthogonal array. While error estimated for regression analysis is 6.01 %, finite element analysis is 4.98 % marginally outperforms, fuzzy logic analysis which is yielded error only 3.26 %. The fuzzy logic analysis is the proposed methodology used to predict burn off length parameter for any set of welding input parameters. This model helps us to find out the actual length of the material in the joining process.