Multi-Objective Optimization of Thermo-Mechanical Modelling of Friction Stir Welding

The following research is primarily focused on the development of a reliable thermo-mechanical model of the Friction Stir Welding (FSW) for aerospace applications.

The main objective is to realistically simulate the FSW process characteristics using a synergic approach of advanced Finite Element (FE) modelling together with high-resolution thermal imaging.

Since FSW occurs at a solid state, the heat input mechanism, based on a complex frictional contact between the rotating tool and the workpiece, is strongly affected by several temperature-dependent material properties.

The new release of ANsys 14 FE software has innovative Parametric Design Language (APDL) command features specifically designed for frictional heat generation, plastic heat generation and temperature controlled bonding contacts, perfectly suited for a FSW process simulation that can efficiently be implemented into a multi-objective optimization platform, i.e. modeFRONTIER.

Experimentally, thermocouples are insufficient in obtaining complete thermal representations of the tool and workpiece as they can be used only as control points along the weld. Instead a high-resolution, high-sensitivity infrared (IR) camera and an IR real-time digital storage and analysis software can provide precise temperature information at all visible points, obtaining a complete external thermal assessment of the process.

Applying a Taguchi Design of Experiments method, FSW trials were performed on a standard aluminium alloy for airframe structures, using a Triflat design tool, and modifying the primary parameters for heat generation within the operating window. Welds were thermally monitored using a FLIR T640 IR camera to create a significant database to be used to optimize the developed ANsys/modeFRONTIER FSW parametric model.

An accurate FSW thermo-mechanical simulation will help to achieve a deeper understanding of the different phases of the FSW process and an enhanced control of the key parameters of this technology, substantially reducing the testing phase required to frame a robust sweet-spot.