TOWARDS EFFICIENT AND TIME-ACCURATE SIMULATIONS OF EARLY STAGES OF INDUSTRIAL SCALE EXPLOSIONS (original) (raw)
Related papers
Implicit-explicit Runge-Kutta method for combustion simulation
2006
New high order implicit-explicit Runge-Kutta methods have been developed and implemented into a finite volume code to solve the Navier-Stokes equations for reacting gas mixtures. The resulting nonlinear systems in each stage are solved by Newton's method. If only the chemistry is treated implicitly, the linear systems in each Newton iteration are simple and solved directly. If in addition certain convection or diffusion terms are treated implicitly as well, the sparse linear systems in each Newton iteration are solved by preconditioned GMRES. Numerical simulations of deflagration-to-detonation transition (DDT) show the potential of the new time integration for computaional combustion.
Numerical simulation of explosion phenomena in industrial environments
Journal of Hazardous Materials, 1996
The code REACFLOW developed at the JRC Ispra combines advanced numerical techniques for the simulation of transient, multi-dimensional, multi-component gas flows undergoing chemical reactions to a unique tool. It uses a true 2-D discretisation with an unstructured triangular grid to ensure a maximum of flexibility for the representation of complex geometries. The numerical discretisation uses a finite volume scheme based on an approximate Riemann solver. Explicit, implicit and semi-implicit methods cover the whole range of time scales. Compressible and incompressible flow is treated with an arbitrary number of components. Chemical reactions are calculated fully implicitly. Diffusion processes are also modelled using a finite volume equivalence to the finite element Galerkin method. A k-e turbulence model is currently being implemented. A system for dynamic grid adaptation automatically detects locations of refinement and coarsement based on local gradients of flow variables. The code capability will be demonstrated by various applications, including a hydrogen/air explosion in a containment and a 'tulip' flame calculation.
A computational method for combustion in high speed flows
Computers & Fluids, 2012
A two-dimensional time-accurate numerical model to simulate complex reacting flowfields in chemical non-equilibrium is presented. The aim of this study is to develop a computational tool which permits the analysis and the easy implementation of combustion phenomena for high speed flows. To construct an efficient numerical tool, while maintaining a reasonable accuracy, a semi-implicit numerical method was selected and verified for a hydrogen-air mixture. The numerical approach is based on a time-dependent, finite volume integration of the governing equations suitably modified for chemical non-equilibrium. The evaluation of the reacting constants based on Gibbs free energy and the Van't Hoff equation allows a very easy implementation of the chemical model used, regardless of its complexity. Calculations were performed with adeguate temporal and spatial resolution for modeling the physical process for practical calculation. Comparisons with numerical results are used for a verification of the numerical procedure.
Time–space adaptive numerical methods for the simulation of combustion fronts
Combustion and Flame, 2013
This paper presents a new computational strategy for the simulation of combustion fronts based on adaptive time operator splitting and spatial multiresolution. High-order and dedicated onestep solvers compose the splitting scheme for the reaction, diffusion, and convection subproblems, to independently cope with their inherent numerical difficulties and to properly solve the corresponding temporal scales. Adaptive and thus highly compressed spatial representations for localized fronts originating from multiresolution analysis result in important reductions of memory usage, and hence numerical simulations with sufficiently fine spatial resolution can be performed with standard computational resources. The computational efficiency is further enhanced by splitting time steps established beyond standard stability constraints associated to mesh size or stiff source time scales. The splitting time steps are chosen according to a dynamic splitting technique relying on solid mathematical foundations, which ensures error control of the time integration and successfully discriminates time-varying multi-scale physics. For a given semidiscretized problem, the solution scheme provides dynamic accuracy estimates that reflect the quality of numerical results in terms of numerical errors of integration and compressed spatial representations, for general multi-dimensional problems modeled by stiff PDEs. The strategy is efficiently applied to simulate the propagation of laminar premixed flames interacting with vortex structures, as well as various configurations of self-ignition processes of diffusion flames in similar vortical hydrodynamics fields. A detailed study of the error control is provided and show the potential of the approach. It yields large gains in CPU time, while consistently describing a broad spectrum of space and time scales as well as different physical scenarios.
Combustion, Explosion, and Shock Waves, 1995
A new numerical algorithm for simulation of nonequilibrium chemically reacting flows of an inviscid muIticomponent gas is described. Application of this algorithm to the numerical solution of several problems of air-hydrogen mixture combustion in oblique detonation waves is demonstrated. Numerical simulation of hypersonic flows of a real multicomponent gas with finite-rate chemical reactions is of both theoretical and practical interest. Extensive investigations to create an aerospace plane have been pursued in recent years. Here the development of a hydrogen-fueled hypersonic ramjet engine (scramjet engine) is a major problem. Nonequilibrium chemical processes occur in all elements of this engine: in an air inlet (air dissociation behind intense shock waves), in a nozzle (combustion product recombination), and, of course, in a combustor, where supersonic mixing and burning of the air-hydrogen mixture take place. The calculation of nonequilibrium hypersonic flows presents a number of severe difficulties. First of all, the dimensionality of the problem increases abruptly because one must monitor the concentrations of all the components of the reacting mixture. Then, the system of equations of chemical kinetics is of a rigid type, i.e., the physical phenomenon involves a few characteristic times differing from one another by a hundred times. The application of explicit numerical algorithms to the simulation of such processes is hardly possible because this requires a time step equal to the minimum characteristic time, whereas the use of implicit algorithms involves problems of inverting large-dimensionality systems of nonlinear equations and adequacy of a numerical solution. In view of the problem complexity it is expedient to begin with the development of efficient numerical methods for calculation of reacting flows of an inviscid gas. To solve this problem, one can use a reliable base, the well-developed numerical simulation of ideal inviscid gas flows. The main problems arising within the framework of the model of an inviscid reacting gas are related to the consideration of shock wave structures with heat release and, first of all, detonation waves. The combustion behind shock waves and in detonation waves can be used in scramjet combustors, while, as noted in [1, 2], an oblique-detonation-wave engine is more promising for flight Mach numbers Moo > 15. That is why the numerical simulation of inviscid reacting flows is of scientific and practical interest. Let us mention some problems that call for solution: the interaction of detonation waves in an air-hydrogen mixture with combustor wails, flame stabilization behind a shock wave, i.e., combustion process localization and its maintenance in different regimes (for instance, under the conditions of a nonpremixed combustible mixture, with changing wedge angle, etc.). The present paper is devoted to the development of an efficient numerical method for calculation of reacting inviscid gas flows and numerical solution of the above problems.
Direct numerical simulation of premixed turbulent flames
Philosophical Transactions of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences, 1999
The enormous increase in computer power during the recent past has made it possible for the first time to carry out direct numerical simulation (DNS) of turbulent reacting flow fields, without the use of a turbulence model. This has opened up a significant new area of research into the fundamental processes of flame-turbulence interaction since it is now possible to solve the governing equations in full detail, resolving even the smallest of the important features of the flow and the flame. In many ways, the technology of DNS remains in its infancy, and the present work seeks to extend current methods in order to take advantage of massively parallel supercomputers. The governing equations are presented and necessary simplifications are justified in the light of current computational capabilities. A numerical discretization scheme suitable for parallel implementation is described, and results are presented for a series of simulations of simple test problems. The relevance of the results is discussed in the light of current turbulent-combustion modelling, and suggestions are made for future work.
Proceedings of the Combustion Institute, 2017
In the present work a sixth-order hybrid WENO-Centered Difference (CD) scheme with an adaptive mesh refinement method is employed to investigate gaseous detonation by injecting a hot jet into a hydrogen-oxygen combustible mixture flowing at supersonic speed. Two-dimensional reactive Navier-Stokes (NS) equations with one-step two-species chemistry model are solved numerically. The comparison between viscous and inviscid detonation structures shows that due to the absence of both the physical viscosity in Euler equations and minimization of numerical dissipation in the hybrid WENO-CD scheme, very small-scale vortices can be observed behind the detonation front. The diffusion effect in the NS equations suppresses the small-scale vortices, but it has negligible influence on the large-scale vortices generated by Richtmyer-Meshkov (RM) instability and those along the highly unstable shear layers induced by Kelvin-Helmholtz (KH) instability. When studying the same setup in an expanding channel and beyond the point of detonation initiation, it is found that because of the diffusion effect of detached shear layers, any unburned jet flow is consumed quickly and then additional energy is released periodically. Because of the formation of multiple secondary triple points and subsequent shear layers after the shutdown of the hot jet, a highly turbulent flow is produced behind the detonation front. Rather than the commonly known RM instability, the large-scale vortices involved in the highly unstable shear layers dominate the formation of the turbulent flow and the rapid turbulent mixing between the unburned jet flow and burned product. It is found that the size of unburned jets and vortices due to KH instability is growing for larger expansion angles. The further generated turbulent flow resulting from larger sized vortices, significantly enhances the mixing rate behind the Mach stem, leading to rapid consumption of the unburned reactants. Therefore, detonations propagate faster in channels with larger expansion angle and higher expansion ratio.
Influence of computational aspects on simulations of a turbulent jet diffusion flame
International Journal of Numerical Methods for Heat & Fluid Flow, 2003
The influence of computational aspects on simulation results is quantitatively investigated for the specific case of a turbulent piloted jet diffusion flame (Sandia Flame D). It is illustrated that, with a fixed turbulence and chemistry model, the results can heavily depend on the numerical aspects. The influence of inlet boundary conditions has already been discussed in an earlier paper. In this work, attention is focused onto the order of accuracy of the spatial discretization in the numerical scheme and onto the position of the outlet boundary. It is stressed that the purpose is not to judge the quality of the applied models, but to illustrate the possible impact of numerical influence factors. The conclusion is a warning message and a demonstration that all numerical aspects must be completely described when calculation results are presented.