Friction Stir Welding

Problem Description
- Some metals such as aluminum and its alloys are known as nonweldable materials using traditional methods of welding and are unable to provide enough strength due to porosity in the fusion zone.
- Recent improvements in welding methodology have led to the development of a new welding technique known as friction stir welding (FSW).
- The basic concept is that a nondevourable solid-state heat-treated hard metal tool is introduced into the butting ends of sheets or plates to be joined and moved along the line of joint at specific rotational speed, traverse speed, axial force, and tilt angle.
- As the rotating tool is in contact with joining materials, it heats up the joint, which further leads to plastic deformation of the joint. This process comes under the solid-state welding category as there in no weld pool formation at that joing; rather a rotary motion of the tool transfers the material to produce the joint.
- In this study, two dissimilar aluminum alloys AA70775 and AA6061 having different thicknesses of 3, 4, and 5 mm were examined. Butt joint welding configuration using a vertical milling machine with special purpose tool was carried out.
Model Description
- A mathematical model is derived where mechanical properties of the joint formed by FSW are observed using different levels of process parameters.
- Mechancial properties of the joint are tensile strength, elongation, and impact strength. These values of these properties are used to assess the quality of the weld.
- Process parameters used are axial load, rotational speed of the tool, welding speed, tilt angle, and plate thickness.
- A nonlinear response function is used to model mechanical properties of the FSW joint as functions of the process parameters.
- The nonlinear response function variables are optimized to find the set that yields response curves to best fit the observed data.
- Predictions of tensile strength, elongation, and impact strength are made based on a set of process parameters which was not used to build the model (prediction set).
- A virtual screen is performed, whereby the process parameter space is searched to find the sets that yield the highest values of a given obective function (VSOF), where a high value of the objective function indicates a high quality weld.
- The objective function used in this example is a normalized sum of the resulting mechanical properties - tensile strength, elongation, and impact strength.
Results
- Tensile Strength vs. Process Parameters
  - Observed data points, nonlinear response function fit, and nonlinear response function prediction (for the prediction set) are shown here.
- Elongation vs. Process Parameters
  - Observed data points, nonlinear response function fit, and nonlinear response function prediction (for the prediction set) are shown here.
- Impact Strength vs. Process Parameters
  - Observed data points, nonlinear response function fit, and nonlinear response function prediction (for the prediction set) are shown here.
- Virtual Screen
  - Plots are generated for percent of maximum VSOF value vs process parameter, for the 5 different process parameters. These plots are shown here. The highest scoring process parameter set is also shown in the table at the bottom.
Conclusions
- The nonlinear response function shows fairly accurate fits to the observed data, over the entire range of input process parameters.
- The methods built in to the nonlinear response function approach avoid overfitting the model, even though the amount of observed data is relatively sparse.
- By avoiding overfit models, the nonlinear response function approach provides for better predictions of weld quality as in areas of the process parameter space that lacked observed data and were not used to build the model.
- Better predictions of weld quality over the entire process parameter space allows for virtual screening that can provide valuable insight into what process parameter sets will yield the highest quality welds using the FSW process.