Design and Simulation Technologies (DSTECH) Inc. is an SME company supplying and services with customers in the solution of complex engineering problems encountered in different disciplines such as Environment, Energy, and Aerospace sciences. DSTECH develops and employs open-source platforms for the realistic simulation of complex flows on HPC clusters. Novel designs are developed to enhance the efficiency of engineering systems based on high-resolution numerical simulations.

TRUBA HPC Center, operating under TUBITAK ULAKBIM, coordinates NCC Turkey. Middle East Technical University (METU), Sabancı University (SU), and Istanbul Technical University National Center for High-Performance Computing (UHeM) are the third parties of the NCC. Our objective is to level up the industry, public sector, and academia in High-Performance Computing (HPC), High-Performance Data Analytics (HPDA), Artificial Intelligence (AI), and prepare for the developing EuroHPC ecosystem. The particular third party presenting this success story is TRUBA.

         

Technical/scientific Challenge

Dynamic systems should be optimised using a simulation and optimisation framework to find out design parameters without using a blind-search approach. DSTECH intended to optimise geometrical specifications of a previously patented slot-baffle design for the enhancement of the efficiency of a disinfection contact system. Specifications of the slot-baffle design need to be optimised considering single and dual objectives of reducing short-circuiting effects and increasing mixing. Optimisation of the design requires a series of numerical simulations that may take a long simulation duration.

Solution

DSTECH has developed an open-source simulation and optimisation framework augmented with OpenFoam, Dakota, and Python, which can be freely used on HPC clusters without any restrictions such as license, commercialization, and parallelization. The present simulation and optimisation framework is employed to optimise the slot-baffle design in three stages without any convergence issues. Numerical simulations are performed with parallel computing strategies using an intense computational resource allocated on the TRUBA.  Numerical simulation results in the present study reveal that the efficiency of the conventional design can be improved by 12.47% when the optimised design is implemented to the present tank.

Figure 1. Optimisation of the slot dimensions for constant locations. (a) Three-dimensional view of the optimised baffle and (b) variation of the object function with the iteration

Figure 2. Variations of the parameters during optimisation. (a) Variations of width and height, variation of (b) height and (c) width.

Business impact

Intense computational power is required for the accurate simulation of turbulent flows in large-scale engineering systems. Use of HPC enables us to perform a series of high-resolution numerical simulations for the optimisation of engineering designs in the project schedule. Moreover, there is no need to install expensive infrastructures and maintenance costs are not required with the applications of parallel computing on the HPC, which are important issues for SMEs.

The case study also started a new partnership between TRUBA and DSTECH. DSTECH has developed an open-source CFD code for the simulation of high-speed turbulent flows over aircrafts with the partnership of TAI (Turkish Aerospace Industries Inc.). In the Horizon 2020 project ECO-Qube, an open-source computational framework has been developed for the simulations of flow and thermal structures in data centres. A series of numerical simulations will be performed for different thermal scenarios to generate extensive data required in the development of an AI-augmented cooling system in ECO-Qube project. DSTECH is developing an AI-augmented disinfection system for the smart management of potable water treatment plants in another research and development project funded by Small and Medium Enterprises Development Organization (KOSGEB) in Turkey. Both TRUBA and DSTECH will evaluate collaboration around these topics in the scope of the EuroCC project.

Benefits

The following benefits are achieved;

  • An automatic-optimisation framework has been developed.
  • A patented design is optimised with parallel computing on the HPC cluster.
  • The optimised design has improved the efficiency of the system by 12.47%.
  • A new partnership is formed. DSTECH and TRUBA HPC Center will collaborate further around new topics.

Keywords

  • CFD, HPC, OpenFoam, Optimisation, Dakota
  • Industry sector: Environment
  • Technology: CFD Simulation

DSTECH developed a simulation and optimisation framework integrating open-source CFD and optimisation software, which can be freely accessed in TUBITAK open archive Aperta https://aperta.ulakbim.gov.tr/record/227987#.YfbwYepByUk