Nuclear Safety Research with Open Source CFD Software


Nuclear Safety Research with Open Source CFD Software

Schlegel, F.

During the last years, open source software (OSS) has become more and more popular in academia and industry. For academia in particular, transparency and reproducibility of results according to the FAIR principles is a fundamental requirement for good scientific work. Beside such considerations, OSS has several advantages over commercial software, e.g. the availability of the source code, transparency and reliability of the implemented algorithms, flexibility for own implementations and developments, long-term availability, independence from commercial interests of software manufactures, license models that allow easy collaborations and much more. However, OSS also has disadvantages, which limits the applicability, e.g., often brief documentation, requires high level programming knowledge, upstream contributions require high coding and software design standard, discussions in the community can be very time consuming, many open source licenses cause conflicts with other commercial software packages or national security aspects, and often a sufficient quality assurance is not available.

State-of-the-art for software developments follow nowadays an agile development strategy, which is based on the release of frequent and small changes instead of a long-term milestone driven development. Those frequent commits (modifications) require a high level of automation, unit and integration tests for quality assurance, version control, code style checking, automated packaging and deployment and automated documentation. Those tools can be utilized for a sustainable and efficient development of OSS.

Computational Fluid Dynamics (CFD) is a highly specialized field and is characterized by a vast complexity of the underlying physics and equation. In this field, OpenFOAM1 established itself as the leading open source software package for numerical simulations that owns today a significant share of the market. The software together with the source code is provided by the OpenFOAM Foundation, which ensures robustness, functionality, usability, extensibility and accessibility of OpenFOAM. However, OpenFOAM yet covers not all functionality required for nuclear safety research, and a lot of research and development work is done in this regard. Three years ago, the German CFD Network for Nuclear Safety Research selected OpenFOAM as reference software for containment and reactor coolant system (RCS) related topics. A key point associated with selection of OSS was the need to establish a coordinator, which ensures efficient and sustainable developments in the long term.

HZDR has a long experience in nuclear reactor thermal hydraulics and multiphase CFD, e.g., we have developed an automated scientific workflow based on Snakemake2. HZDR is also an active contributor to the OpenFOAM Foundation release and member of the German CFD Network for Nuclear Safety Research. As a member of the Helmholtz Society, HZDR has access to the Helmholtz Cloud Services3, which provide an excellent environment for agile software development (Gitlab, Mattermost, a.s.o.). Due to this expertise, HZDR was selected as coordinator and maintainer for the OpenFOAM_RCS4 project, funded by German Federal Ministry for the Environment, Nature Conservation, Nuclear Safety and Consumer Protection. Within the OpenFOAM_RCS project, HZDR will provide a common platform for nuclear safety research with respect to reactor coolant systems in Germany. The platform includes a repository for software code, a repository for restricted simulation setups, version control, automated testing with pipelines, validation tests and reports for new OpenFOAM version, and documentation. A key thing of the project is also the intensive collaboration with the OpenFOAM core developers (including some funding) for discussion of APIs, core maintenance, and future developments of OpenFOAM.

The lecture will present a comprehensive discussion of the pros and cons of OSS with a special focus on OpenFOAM and why this is a good choice as reference software. It will also show one way that is established and further developed within the OpenFOAM_RCS project to work as backend developer of a large OSS software project like OpenFOAM, and how to overcome the limitations and ensure sustainability of the additional developments. The lecture will discuss briefly the status of the OpenFOAM_RCS project and the planned applications for the next years.

  • Lecture (Conference)
    Advances in Thermal Hydraulics (ATH 2022), 12.-16.06.2022, Anaheim, USA

Permalink: https://www.hzdr.de/publications/Publ-34653