Scalable computational steering system for vizualization of large scale CFD simulations

2002-06-27
Tonkal, Ozan Çağrı
Pehlivan, Sercan
Sezer Uzol, Nilay
İşler, Veysi
A general-purpose computational steering system (POSSE) which can be coupled to any C/C++ simulation code, has been developed and tested with a 3-D Navier-Stokes flow solver (PUMA2). This paper illustrates how to use “computational steering” with PUMA2 to visualize CFD solutions while they are being computed, and even change the input data while it is running. In addition, the visualizations can be displayed using virtual reality facilities (such as CAVEs and RAVEs) to better understand the 3-D nature of the flowfields. The simulations can be run on parallel computers or Beowulf clusters, while the visualization is performed on other computers, through a clientserver approach. A key advantage of our system is its scalability. The visualization is performed using a parallel approach. This is essential for large-scale simulations, since it is often not possible to postprocess the entire flowfield on a single computer due to memory and speed constraints. Example solutions from this solver are presented to show the usefulness of POSSE. The examples include unsteady ship airwake simulations, unsteady flow over a helicopter fuselage, and unsteady simulations of a helicopter rotor. The results of the rotor simulations in hover are compared with the experimental measurement and discussed in some detail. The advantages of using object-oriented programming are also discussed.
AIAA Fluid Dynamics Conference and Exhibit, (24 - 27 Haziran 2002)

Suggestions

Scalable computational steering system for visualization of large-scale CFD simulations
Modi, Anirudh; Sezer Uzol, Nilay; Long, Lyle N.; Plassmann, Paul E. (2002-12-01)
A general-purpose computational steering system (POSSE) which can be coupled to any C/C++ simulation code, has been developed and tested with a 3-D Navier-Stokes flow solver (PUMA2). This paper illustrates how to use "computational steering" with PUMA2 to visualize CFD solutions while they are being computed, and even change the input data while it is running. In addition, the visualizations can be displayed using virtual reality facilities (such as CAVEs and RAVEs) to better understand the 3-D nature of th...
Scalable computational steering for visualization/control of large-scale fluid dynamics simulations
Modi, Anirudh; Sezer Uzol, Nilay; Long, Lyle N.; Plassmann, Paul E. (2005-01-01)
The development, integration, and testing of a general-purpose "computational steering" software library with a three-dimensional Navier-Stokes flow solver is described. For this purpose, the portable object-oriented scientific steering environment (called POSSE) library was used. This library can be coupled to any C/C++ simulation code. The paper illustrates how to integrate computational steering into a code, how to monitor the solution while it is being computed, and how to adjust the parameters of the a...
Implementation of Robot Formation Control and Navigation Using Real Time Panel Method
Merheb, Abdel Razzak; Atas, Yunus; Gazi, Veysel; Sezer Uzol, Nilay (null; 2010-10-22)
Parallel implementation of a gas-kinetic BGK method on unstructured grids for 3-D inviscid missile flows
Ilgaz, Murat; Tuncer, İsmail Hakkı (2009-10-12)
A 3-D gas-kinetic BGK method and its parallel solution algorithm are developed for the computation of inviscid missile flows on unstructured grids. Flow solutions over a supersonic missile are presented to validate the accuracy and robustness of the method. It is shown that the computation time, which is an important deficiency of gas-kinetic BGK methods, may significantly be reduced by performing computations in parallel. © 2009 Springer-Verlag Berlin Heidelberg.
FPGA implementation of neuro-fuzzy system with improved PSO learning
KARAKUZU, CİHAN; KARAKAYA, FUAT; Cavuslu, Mehmet Ali (2016-07-01)
This paper presents the first hardware implementation of neuro-fuzzy system (NFS) with its metaheuristic learning ability on field programmable gate array (FPGA). Metaheuristic learning of NFS for all of its parameters is accomplished by using the improved particle swarm optimization (iPSO). As a second novelty, a new functional approach, which does not require any memory and multiplier usage, is proposed for the Gaussian membership functions of NFS. NFS and its learning using iPSO are implemented on Xilinx...
Citation Formats
O. Ç. Tonkal, S. Pehlivan, N. Sezer Uzol, and V. İşler, “Scalable computational steering system for vizualization of large scale CFD simulations,” presented at the AIAA Fluid Dynamics Conference and Exhibit, (24 - 27 Haziran 2002), St. Louis, Missouri, Amerika Birleşik Devletleri, 2002, Accessed: 00, 2021. [Online]. Available: http://arc.aiaa.org/doi/abs/10.2514/6.2002-2750.