These slides were presented at the NAFEMS World Congress 2025, held in Salzburg, Austria from May 19–22, 2025.

Abstract

While CFD methods for simpler flow conditions have already found their way into product and process development, complex problems such as multiphase or multiphysics applications often still require intensive model validation and, due to the usually long computing times, model optimisation. One way to provide the developer with a design tool for these more demanding CFD tasks is to create an automated CFD tool for a specific process that is validated and optimised for the application. This makes it particularly easy to use, especially in conjunction with cloud resources, as the frequently required high computing resources can be made available as needed. This means that multiple node computer architectures can be addressed in the background under Linux and the administration of hardware systems on site is no longer necessary. Using the example of modelling the cooling behaviour of titanium components during heat treatment, the creation and application of such a tool to reduce creep processes during cooling is demonstrated. Natural convection, heat conduction, radiation exchange and energy release due to microstructural transformation are taken into account in a transient conjugate heat transport analysis. The CFD model was created in OpenFOAM, validated by numerous experiments and optimised in terms of computing time. Both cooling by natural convection in quiescent air and cooling by forced convection in a rapid cooling chamber were considered. For both cases, good agreement was found with the experimentally determined temperature curves and cooling rates [1]. Different phases of the cooling mechanisms could be identified, e.g. periods in which microstructural transformation and radiation dominate the temperature curve could be distinguished from the phase of convection-dominated cooling. The CFD model setup and the simulation process are fully automated: The user provides the geometry as an STL assembly and defines his simulation parameters such as material data, initial temperatures and cooling times. The model is meshed fully automatically and the simulation is carried out. The simulation tool can be used both in-house and in the cloud. An interface to Abaqus for the use of transient temperature fields in creep analyses has been implemented. The tool is used by the end user to optimise the cooling conditions of titanium components in the preliminary design phase. In particular, it was shown that component distortion due to creep is minimised if the component is cooled as evenly as possible on the top and bottom surfaces. In this context, the grate on which the component lies is of particular importance, as a large amount of energy is stored here, which is exchanged with the component over a long period of time through radiation. As a result, the component cools down more slowly on the underside. Numerical simulations can be used to test various measures in advance in order to compensate for these effects and ensure uniform cooling of the component. References [1] Ulrich Heck, Martin Becker, Ralf Paßmann A coupled flow, heat and structural analysis tool to reduce creep during heat treatment of titanium components, Nafems Multiphysics Conferences, Munich 14 - 15 Nov. 2023