Skip to content

PH.D. AFAQUE SHAMS, ASES CONSULTANCY | TOMASZ KWIATKOWSKI, NCBJ
NATIONAL CENTRE FOR NUCLEAR RESEARCH (NCBJ)

PH.D. AFAQUE SHAMS, ASES CONSULTANCY | TOMASZ KWIATKOWSKI, NCBJ
NATIONAL CENTRE FOR NUCLEAR RESEARCH (NCBJ)

Preparation of reference databases of the Direct Numerical Simulations class for the design of new nuclear technologies and nuclear safety applications

PROJECY OBJECTIVE

Challenge

The project addresses two fundamental nuclear safety issues: thermal shock under high pressure and coolant mixing in fuel cassettes.

Project objective

Generation of two Direct Numerical Simulations (DNS) class reference databases for two physical phenomena occurring in nuclear reactors:

  • pressurized thermal shock (PTS)
  • mixing of coolant in fuel cassettes.

TASKS FOR THE SUPERCOMPUTER

01.

The generation of reference databases of the Direct Numerical Simulations (DNS) class is not a trivial issue from both a physical and technical point of view. The results from DNS calculations can be equated with the results of a physical experiment. The disadvantages of these calculations are the time required to carry them out (months or even years, depending on the case) and the need for large computational resources. Therefore, although DNS calculations are super-accurate, they are not attractive from the point of view of industry and various types of commercial work. On the other hand, in situations where all physical experiments are not feasible, simulations become an indispensable tool for understanding the physics of the phenomena under investigation.

02.

All work carried out as part of the research takes place using the computing resources of the Świerk Information Centre (CIŚ) cluster. In addition, most of the pre- and post-processing work is also carried out on the CIŚ cluster infrastructure, using the necessary scientific and engineering software.

03.

3. Software used: NEK5000, ANSYS Fluent, OpenFOAM, VisIT, ParaView, Matlab, Python.

BENEFITS OF COOPERATION WITH NCBJ

The research work carried out as part of the project required enormous computing power made available by the NCBJ. For the thermal shock simulation, 5 000 CPUs (central processing units) were used over a period of several months. In contrast, 10 000 CPUs were used to simulate coolant mixing over a period of almost three years.

The generated data takes up tens of TB.
Without the computing resources provided, as well as the storage space, it would not have been possible to carry out this research project.

As part of the project, two reference databases have been generated that will serve as a validation tool for lower order turbulent models. With the acquired data, it will be possible to improve existing and generally used turbulent models to more accurately represent the phenomena under study, i.e. thermal shock and coolant mixing in fuel rods. It is worth mentioning that a reference database for the issue of coolant mixing in fuel rods has been created considering the three different coolants most commonly used in various types of nuclear reactors (water, gas and liquid metal) and assuming two different temperature conditions on the rod walls – constant temperature and constant heat flux. Validation of lower-order turbulent models (RANS class) with the generated reference bases will enable their use in work related to nuclear reactor safety analyses, as well as in the design process of new nuclear technologies.

5 000 CPU

TO SIMULATE THERMAL SHOCK

10 000 CPU

TO SIMULATE COOLANT MIXING

3 years

DURATION OF SIMULATION

EFFECTS

The long-term effects of the project are:

  • the development of a set of so-called best practices, i.e. how such analyses should be carried out,
  • attempt to improve existing models/model by proposing, for example, a new correlation (based on results from the DNS reference database) so that the model in question models flow and heat transfer with greater accuracy in relation to the issues investigated in the project.

 

This will provide the scientific community and nuclear engineers with an ‘enhanced/improved’ tool to produce more reliable results (e.g. in safety analysis work or in the design of new nuclear technologies).

The results can be used for safety analyses of nuclear reactors and in conceptual work on new nuclear technologies.

Publications:

  • A. Shams, D. De Santis, D. Rosa, T. Kwiatkowski, E.J.M. Komen, Direct numerical simulation of flow and heat transfer in a simplified pressurized thermal shock scenario, International Journal of Heat and Mass Transfer, Volume 135, 2019, Pages 517–540, https://doi.org/10.1016/j.ijheatmasstransfer.2019.01.144.
  • Afaque Shams, Dante De Santis, Adam Padee, Piotr Wasiuk, Tobiasz Jarosiewicz, Tomasz Kwiatkowski & Sławomir Potempski, High-Performance Computing for Nuclear Reactor Design and Safety Applications, Nuclear Technology, 2020, 206:2, 283–295, DOI: 10.1080/00295450.2019.1642683.
  • Afaque Shams, Tomasz Kwiatkowski, Towards the Direct Numerical Simulation of a closely-spaced bare rod bundle, Annals of Nuclear Energy, Volume 121, 2018, pages 146–161, https://doi.org/10.1016/j.anucene.2018.07.022.
 
 

RECOMMENDATION

„I started my cooperation with NCBJ back in 2014, and from the beginning of this cooperation, I was using the computing infrastructure of the CIŚ cluster. During this period, we have carried out a number of one-of-a-kind calculations that would not be possible without these very computing resources. The CIŚ cluster has provided software. At the same time, I cannot fail to mention the excellent technical support that I have received from the entire team of administrators. All reported problems were resolved as a matter of urgency and accordingly communicated to me as per the need. The results of our work have been presented at international conferences and also published in different scientific journals. I am extremaly happy with this cooperation and the way of working of the whole team of administrators. I will be glad to continue futher computing projects on the CIŚ cluster. I highly recommend the services offered by the CIŚ team and cluster”.

PH.D. AFAQUE SHAMS, ASES CONSULTANCY | TOMASZ KWIATKOWSKI, NCBJ
NATIONAL CENTRE FOR NUCLEAR RESEARCH (NCBJ)

Information about cookies

The The National Competence Centre in HPC service uses cookies, which are pieces of information stored on users’ devices in the form of small text files. These data confirm that the user has visited the website and allow the recognition of their device to tailor subsequent page views to their preferences. The cookies used by the NCK service do not store any personal user data or information that could identify them; they only recognise the specific device’s browser.

Cookies are essential for the proper functioning of the service and help customise the content of the page to the user’s preferences. They provide statistical data on traffic on the site.

Websites, including the The National Competence Centre in HPC service, typically allow the storage of cookies by default, but this can be changed at any time to block the automatic addition of cookies or to be notified each time they are sent. It is important to note that disabled cookies may disrupt some functions of the website or prevent the use of certain services.

Skip to content