System Identification Based Controller Design for Friction Stir Welding Process During Joining of Aluminium Metal Matrix Composites

Abstract

Friction Stir Welding (FSW) process was initially implemented by modifying the milling machines. With the advancements in robotics and automation, movement along three primary axes are made controllable using computer integration. FSW being a temperature-dependent process, the temperature at the weld zone affects the weld quality since the microstructure is totally altered by the temperature variations. If welding is done at constant process parameters by ignoring the process disturbances, it would also result in undesirable material properties. From research studies, it is understood that tool pin position and spindle speed are significant process parameters contributing to heat generation during the joining process. In this study, an attempt is made to control the spindle speed of the rotating tool by measuring tool-workpiece interface temperature and vibrational disturbances during joining of Aluminium Metal Matrix Composite (Al-MMC) plates. The system model is estimated from the experimental data using the concept of system identification. Followed by, a Smith Predictor control scheme is developed and validated in a closed loop FSW system to examine the tensile strength.