Abstract
The simulation of fire is a challenging task due to its occurrence on multiple space-time scales and the non-linear interaction of multiple physical processes. Current state-of-the-art software such as the Fire Dynamics Simulator (FDS) implements most of the required physics, yet a significant drawback of this implementation is its limited scalability on modern massively parallel hardware. The current paper presents a massively parallel implementation of a Gas Kinetic Scheme (GKS) on General Purpose Graphics Processing Units (GPGPUs) as a potential alternative modeling and simulation approach. The implementation is validated for turbulent natural convection against experimental data. Subsequently, it is validated for two simulations of fire plumes, including a small-scale table top setup and a fire on the scale of a few meters. We show that the present GKS achieves comparable accuracy to the results obtained by FDS. Yet, due to the parallel efficiency on dedicated hardware, our GKS implementation delivers a reduction of wall-clock times of more than an order of magnitude. This paper demonstrates the potential of explicit local schemes in massively parallel environments for the simulation of fire.
Highlights
Fire is typically associated with a form of combustion that emits sufficient amounts of light and energy to be perceptible and self-sustaining on a macroscopic scale [1]
We present the application of Gas Kinetic Scheme (GKS) for low speed flows for the simulation of three-dimensional turbulent natural convection and two fire plumes on two different spatial scales under the framework of Large Eddy Simulation (LES)
We demonstrated the efficient usage of gas kinetic schemes for the simulation of fire plumes
Summary
Fire is typically associated with a form of combustion that emits sufficient amounts of light and energy to be perceptible and self-sustaining on a macroscopic scale [1]. The current standard in fire simulation is the Fire Dynamics Simulator (FDS), which is freely accessible and well documented [12,13,14,15] It is based on a thermal compressible flow solver utilizing a low Mach number approximation. (LBM) [20,21,22,23], are utilized to compute approximate solutions of the compressible or weakly compressible Navier–Stokes equations Such explicit schemes tend to require a relatively small time step as compared to implicit methods, but usually require less operations per time step and allow for straight forward parallelization, resulting in near to optimal scaling beyond tens of thousands of compute cores [24,25,26]. Schemes (GKS) are another class of kinetic solvers for computational fluid dynamics They were originally developed as flow solvers for high Mach number flows with shocks on structured [27,28,29]. Massively parallel hardware in the sense of multi-GPGPU was utilized
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
