Abstract
Abstract. Verification, the process of checking a modelling output against a known reference model, is an important step in model development for the simulation of manufacturing processes. This manuscript provides details of a code-to-code verification between two thermal models used for simulating the melting and solidification processes in a 316 L stainless steel alloy: one model was developed using a non-commercial code and the Finite Volume Method (FVM) and the other used a commercial Finite Element Method (FEM) code available within COMSOL Multiphysics®. The application involved the transient case of heat-transfer from a point heat source into one end of a cylindrical sample geometry, thus melting and then re-solidifying the sample in a way similar to an autogenous welding process in metal fabrication. Temperature dependent material properties and progressive latent heat evolution through the freezing range of the alloy were included in the model. Both models were tested for mesh independency, permitting meaningful comparisons between thermal histories, temperature profiles and maximum temperature along the length of the cylindrical rod and melt pool depth. Acceptable agreement between the results obtained by the non-commercial and commercial models was achieved. This confidence building step will allow for further development of point-source heat models, which has a wide variety of applications in manufacturing processes.
Highlights
Verification and validation are two procedures in model development that are often discussed interchangeably
This paper describes the development of two modelling applications and their code-to-code verification
– Mesh independency tests revealed that the results obtained from the non-commercial and commercial models were mesh independent allowing for meaningful and appropriate model verification
Summary
Verification and validation are two procedures in model development that are often discussed interchangeably. In the context of the American Institute of Aeronautics and Astronautics (AIAA) definition (Roache, 2012), verification is the process of checking that a code’s simulation output is consistent with the underlying mathematical requirements, whereas validation is the process of checking a code’s simulated outputs against well-defined physical experimental results. Both steps are essential in model development, but verification precedes validation. In the case where an analytical solution is unavailable, mesh-independent numerical simulation results for identical modelling scenarios can be compared; this approach is known as code-to-code verification
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
