Friction Stir Welding and Processing XI: Modeling: Process & Properties
Sponsored by: TMS Materials Processing and Manufacturing Division, TMS: Shaping and Forming Committee
Program Organizers: Yuri Hovanski, Brigham Young University; Piyush Upadhyay, Pacific Northwest National Laboratory; Yutaka Sato, Tohoku University; Nilesh Kumar, University of Alabama, Tuscaloosa; Anton Naumov, Peter The Great St. Petersburg Polytechnic University

Thursday 8:30 AM
March 18, 2021
Room: RM 39
Location: TMS2021 Virtual


8:30 AM  
Application of Machine Learning for Prediction of Microstructure and Mechanical Performances in Solid-state Joining Processes: Benjamin Klusemann1; Frederic Bock1; Uceu Suhuddin1; Lucian Blaga1; Jorge dos Santos1; 1Helmholtz-Zentrum Geesthacht
     Data-driven machine learning models exhibit strong capabilities to identify and quantify relationships along the process-microstructure-property chain. Via regression analysis, highly non-linear correlations along these chain domains can be established and used for predictions. The integration of microstructural feature classification and segmentation, enables strengthening and refinement of those correlations. Predictions of mechanical properties based on particular process parameters but also inverse determination of required process parameters for desired properties can be accomplished.In this contribution, results from the application of various machine-learning-models to correlate process parameters and microstructure characteristics with desired joint properties in solid-state joining techniques will be presented. Two exemplary processes of Refill Friction Stir Spot Welding and Friction Riveting will be analyzed: Experimental data is generated through Designs of Experiments with addition of many experiments that enrich the input data. The predicting capabilities and learning effects of such machine learning models for solid-state joining techniques will be discussed.

8:50 AM  
Friction Stir Welding Defect Prediction Using Computational Solid Mechanic's Modeling: Rafael Giorjao1; Julian Avila2; Eduardo Monlevade3; Antonio Ramirez1; Andre Tschiptschin3; 1The Ohio State University; 2UNESP; 3USP
    The tool role during the friction stir welding concerning the material flow and its thermomechanical behavior is still not entirely understood. In this matter, a computational solid mechanic's numerical model is proposed. This model uses an Arbitrary Lagrangian-Eulerian code in two different pin profile tools, with threaded and unthread pins. The model was able to reproduce the torque as a response of the tool, temperatures and the material flow in any position during the process, including the downward flow effect in the threaded pin. A point tracking analysis of the material flow revealed that threads increase the material velocity and strain rate, promoting a temperature increment during the process, which improved the material flow and avoided filling defects. The capability of this model to reproduce the defects observed in real welded joints and calculate the material thermomechanical characteristics showed its high sensitivity to welding parameters and tool geometries.

9:10 AM  
The Development of FSW Process Modelling for Use by Process Engineers: Mike Lewis1; Simon Smith2; 1FTS Engineering Answers Ltd.; 2Transforming Stress Ltd.
    Computational resources are such that process engineers might consider the use of computer models during the development of the manufacturing process of complex fabrications. However, it might not be possible for such an engineer to obtain reliable materials data, because manufacturing processes tends to require property data for a set of loading conditions that is so extensive that such data is only available for a small number of materials. The current paper discusses the development of a Friction Stir Weld (FSW) modelling method using only generally available materials data for a wide range of materials. The method assumes that accuracy depends upon the conditions close to the FSW tool plus some straightforward mechanics and thermodynamics remote from the tool. The material model was implemented in STAR-CCM+, a commercially available Computational Fluid Dynamics code, and the results from the model have been compared with measurements made on a test weld.

9:30 AM  
Effect of Tool Geometries on “Heat-input” during Friction Stir Welding of Aluminum Alloys: Yutaka Sato1; Yuichiro Tanai1; Tianbo Zhao1; Dalong Yi2; 1Tohoku University; 2Tsinghua University/Tohoku Univdersity
    Many previous studies proposed equations and models to estimate the heat input of FSW, but their verifications would be still insufficient. In this study, the heat input was calorimetrically measured during FSW of aluminum alloys with various welding tools having the different shapes, and effect of tool geometries on heat input was examined. The measured heat-input increased with increasing shoulder diameter, which was in good agreement with many previous works. Interestingly, the heat input strongly depended on the probe diameter and length. An attempt to correlate the heat input with the tool surface area was made, eventually showing that the heat input linearly increased with increasing “effective surface area” defined as (half of shoulder surface area + probe surface area). This result experimentally suggests that the probe dimension affected the heat input more strongly than the shoulder diameter.

9:50 AM  
Experimental and Numerical Investigations of High Strain Rate Torsion Tests of Al-based Alloys at Elevated Temperatures: Anton Naumov1; Anatolii Borisov1; Anastasiya Borisova1; 1Peter the Great St. Petersburg Polytechnic University
    The high strain rate torsion tests at elevated temperatures were realized on Gleeble-3800 System for different Al-based materials: AA 1050, AA 5082 O and AA 2024 T4. High strain rate torsion tests were provided at two elevated temperatures - 350 and 500 C. The stress-strain curves were obtained and analyzed after physical simulation. A finite element model was built and validated using temperature and torque data obtained after physical simulation. Experimental data was compared with a numerical simulation of the process. The possibility to use physical simulation for description of the processes during FSW is discussed.

10:10 AM  
Numerical Simulation and Analysis of Solid Phase Processing: A Validated Friction Extrusion Smoothed Particle Hydrodynamics Model: Lei Li1; Xiao Li1; Anthony Reynolds2; Glenn Grant1; Ayoub Soulami1; 1Pacific Northwest National Laboratory; 2University of South Carolina
    Shear Assisted Processing and Extrusion (ShAPE) is an emerging high-strain, solid-phase processing technique. It turns the metal powder, flake, or billet into high-performance parts without melting the feedstock materials. Currently, most of the studies on ShAPE are based on experiments, which can be costly and limited in data acquirement. To better understand the processing parameters on the performance of the processed parts, a computational model is required. In this study, we develop a smoothed particle hydrodynamics (SPH) model to simulate the ShAPE process. Due to its meshfree and Lagrangian nature, SPH can handle large plastic deformation without heavy re-meshing and track the history-dependent variables more easily. Model predictions evaluate the effects of the die geometry, rotation speed, billet advancing rate, and thermal boundaries on the material flow, stress-strain state, and temperature of the extruded parts. We show that predicted strains and temperature are in good agreement with the experimental data.

10:30 AM  
Effect of Temperature and Strain Parameters of High Strain Rate Torsion Tests on the Microstructure Evolution of Al-based Alloys: Anastasiya Borisova1; Elizaveta Anhimova1; Oleg Zotov1; Anton Naumov1; Anatolii Borisov1; 1Peter the Great St. Petersburg Polytechnic University
    To study the effect of temperature and strain parameters of high strain rate torsion on the microstructure evolution of Al-based alloys, such as AA 5056 and AA 2024 T4, torsion tests were realized under the same conditions on Gleeble-3800 System, as well as numerical simulation in Deform 3DTM software. The tests were provided at elevated temperatures, with varying strain and strain rates. A qualitative and quantitative analysis of the studied alloys using optical microscopy were performed. The strain distribution according to the quantitative analysis of grain size in various zones was studied depending on the temperature and strain parameters of the torsion tests. A qualitative assessment of the stress/strain effect on the grain size according to the results of physical and numerical simulation was realized. The pattern of grain size evolution at different temperature and strain parameters of high strain rate torsion is established.