Investigation of Gaussian Surface Heat Source Loading Methods in TIG Welding Numerical Simulation

Literature Overview

This paper by Zhang Weizhi and Cheng Yanlin from the Institute of Electronics Engineering, China Academy of Engineering Physics, investigates three different Gaussian surface heat source loading methods for TIG welding numerical simulation. Published in Heat Processing Technology in 2015, the study uses ANSYS finite element software to simulate TIG welding on a flat plate and compares the effects of three loading approaches: surface element heat flux density, surface element body heat generation rate, and surface node heat generation rate. The research is supported by a science and technology innovation fund project (426010401).

Heat Source Modeling in Welding Simulation

Accurate heat source modeling is critical for the reliability of welding numerical simulations. The Gaussian surface heat source model is widely used in TIG welding simulation because it provides a reasonable representation of the arc heat distribution on the workpiece surface. However, the method of applying this heat source to the finite element model can significantly affect the simulation results.

The three loading methods investigated in this study are:

Comparative Analysis of Loading Methods

The study reveals significant differences in the behavior and accuracy of the three loading methods:

Surface element heat flux density loading: This method applies a heat flux to the surface elements of the finite element model. However, the element surface numbering can significantly affect the simulation results. Different element numbering sequences can lead to different heat flux distributions, resulting in inconsistent simulation outcomes. Additionally, this method can cause heat flux overlap issues, where adjacent elements receive overlapping heat flux contributions, leading to inaccurate heat input distribution.

Surface element body heat generation rate loading: This method converts the surface heat flux into a body heat generation rate by dividing by the element thickness. While this approach avoids the element numbering issue, the element thickness directly affects the peak temperature in the simulation. Thicker elements result in lower peak temperatures, and thinner elements result in higher peak temperatures. Furthermore, the convergence of the simulation calculation is poor, which can lead to unreliable results or failure to converge.

Surface node heat generation rate loading: This method applies heat generation rates to the surface nodes of the finite element model. While the element thickness still affects the peak temperature, the convergence of the calculation is good. More importantly, the simulated weld width and weld depth values do not change significantly with element thickness, which is a critical advantage for engineering applications. This method provides the most reliable and consistent simulation results.

Engineering Practice Implications

The choice of heat source loading method has direct implications for the accuracy and reliability of welding simulations used in engineering practice:

1. Weld geometry prediction: The surface node heat generation rate method provides the most consistent weld width and depth predictions, which is essential for designing welding procedures and predicting weld quality.
2. Residual stress analysis: Accurate temperature field simulation is the basis for residual stress prediction. The surface node method's good convergence and consistent results make it suitable for residual stress analysis.
3. Distortion prediction: Welding-induced distortion is driven by temperature gradients. The surface node method provides reliable temperature field predictions, enabling accurate distortion analysis.
4. Element mesh design: While the surface node method is less sensitive to element thickness, proper mesh design is still important for computational efficiency and accuracy. The mesh should be refined in the weld zone and coarsened in the far field.
5. Simulation validation: Regardless of the loading method chosen, the simulation results should be validated against experimental data, such as weld geometry, hardness profiles, and residual stress measurements.

Key Questions and Reflections

A key question arising from this study is the physical justification for the different loading methods. The surface element heat flux density method is the most physically intuitive, as it directly represents the heat flux from the arc to the workpiece surface. However, its sensitivity to element numbering and the heat flux overlap issue make it unreliable in practice. The surface node heat generation rate method, while less intuitive, provides the most practical and reliable results.

Another consideration is the applicability of these findings to other welding processes. The Gaussian surface heat source is specific to TIG welding, but similar loading method issues may arise in simulations of other processes such as MIG welding, laser welding, or electron beam welding. Engineers should be aware of the potential impact of heat source loading methods on simulation results regardless of the welding process being simulated.

Study Insights and Implications

This study provides valuable guidance for engineers using finite element simulation to analyze TIG welding processes. The recommendation to use the surface node heat generation rate loading method is supported by the superior convergence behavior and consistent weld geometry predictions. This method should be the default choice for TIG welding simulations in engineering practice. The study also highlights the importance of understanding the numerical aspects of heat source modeling, which is often overlooked in favor of more attention to material properties and boundary conditions. By selecting the appropriate loading method, engineers can improve the reliability of their simulations and make more informed decisions in welding procedure design and quality control.