Improved Algorithms Lead to Lab-Scale Combustion Simulations
In spite of the fundamental technological importance of basic combustion processes, our knowledge of them is surprisingly incomplete. Theoretical combustion science is unable to represent realistic flames in all of their complexity, and laboratory measurements are difficult to interpret and often limited in the type of flame and level of detail they can address. Lawrence Berkeley National Laboratory's Computational Research Division (CRD) Center for Computational Sciences and Engineering (CCSE) has teamed with the Environmental Energy Technologies Division (EETD) to build a high-performance computing solution to flame simulation and analysis. This solution allows detailed three- dimensional representation of flame behavior over time, which could not previously be studied using simulation technology. This new capability sets the stage for dramatic progress in combustion science research.
Current investigations are focusing on two primary areas: how turbulence in the fuel stream affects local combustion chemistry and how emissions are formed and released in the product stream. The first detailed simulations of laboratory-scale turbulent premixed flame experiments using CCSE's software have revealed flame chemistry and fluid transport in unprecedented level of detail. CCSE's Mark Day and John Bell, and EETD researchers are working together to validate the simulations with experimental data and to probe the computed results for information not easily obtainable from the experiment in any other way. The work has applications for devices ranging from power generators, to heating systems, water heaters, stove, ovens, and even clothes dryers.
Simulation at this level of detail was impossible just a few years ago. However, algorithmic improvements by DOE -funded applied mathematics groups such as CCSE have slashed computational costs for turbulent flow phenomena by a factor of 10,000. CCSE has implemented its advanced simulation algorithms in state-of-the-art parallel computing hardware to increase the number of variables available for describing a system from hundreds of thousands to more than a billion. These improvements mean that researchers can now simulate real experiments without having to create ad-hoc engineering models for under-resolved physical processes.
CCSE's modeling improvements take advantage of key mathematical characteristics of low-speed flows, which are common to most combustion applications, to eliminate components of the model that are relevant only to high-speed flow scenarios. These components have little effect on low-speed-flow system dynamics, but they drive down simulation efficiency by unnecessarily limiting the maximum numerical time-step size. The integration algorithms are implemented in a set of software tools based on adaptive mesh refinement (AMR), a dynamic grid-based system that automatically allocates computational resources to regions that contain the most interesting detail. The AMR methodology allows researchers to simultaneously incorporate large-scale effects that stabilize the flame and very fine-scale features of the combustion reaction zone itself.
The detailed solutions computed by CCSE are being validated with experimental data provided by the EETD Combustion Lab. Preliminary comparisons focus on global characteristics, such as mean flame locations and geometries, as well as statistics of instantaneous flame surface structures. In addition to validating the computed solutions, the research groups probe the vast arrays of data generated by the computations in order to learn more about flame details, such as the localized effects of large and small eddies on the structure of the combustion reaction zone.
For example, the distribution of hydrogen atoms in the thermal field is tightly coupled to key chain-branching reactions required to sustain the combustion process itself. The detailed models accurately represent the transport of hydrogen with respect to the other chemical species in the context of this turbulent flow. CCSE is currently using detailed chemistry and transport models containing 20 to 65 chemical species and hundreds of reactions.
Most combustion systems do not operate at low speeds but laboratory experiments where detailed laser diagnostics have been applied are in the low-speed regime. This makes the CCSE method the right one for comparing results with the laboratory experiments.
Example investigations by the group include the simulation and analysis of a three- dimensional V-flame experiment in the EETD Combustion Laboratory (see Figure 1). In this experiment, a thin rod is placed across the exit of a circular nozzle that issues turbulent premixed methane-air fuel vertically into an unconfined region open to the lab. The rod stabilizes a robust V-shaped flame that is highly corrugated and time dependent. Because of the camera's finite shutter speed, the surface of this turbulent flame appears smeared out over a finite flame brush thickness.
In the experiment, the instantaneous location of the flame may be visualized using particle image velocimetry (PIV), in which inert particles are distributed in the unburned gas with a uniform density. The upward-moving gas expands as it passes through the flame; the density of tracer particles shows a corresponding decrease. The location of the abrupt change in particle density, captured in the photograph (Figure 2, left), indicates the instantaneous flame position in a vertical plane through the center of the flame-stabilizing rod.
The photo may be compared to a representative planar slice of the simulated fuel concentration (right) because fuel is consumed at the flame front. The simulation was based on 20 chemical species and 84 fundamental reactions. The computed surface exhibits large-scale wrinkling of the instantaneous flame surface, and, when averaged over time, shows remarkable agreement with the laboratory photos in terms of flame brush thickness, spreading, and growth rates. The still images of the experiment and simulation demonstrate exceptional agreement, both in overall flame shape as well as brush growth characteristics, and fine-scale wrinkling of the flame surface.
The main purpose of using PIV in the experiment is to measure the velocity statistics within the turbulent flame for direct comparison with the simulated results. PIV is a diagnostic method developed for measuring instantaneous velocity distributions within a two- dimensional plane defined by a laser sheet. The system consists of a digital camera, synchronized with a double-pulsed laser beam that is shaped into a thin sheet that dissects the v-flame. The field of view for this PIV system (13 cm by 13 cm in this case), was adjusted to be comparable to the computational domain (12 by 12 by 12 cm). PIV measures the instantaneous velocity vectors by analyzing the spatial displacement of particles shown on a pair of images produced by the double pulsed laser. To obtain velocity statistics, EETD researchers analyzed 448 image pairs. The comparison with simulated results shows that the simulation captures the salient features of the v-flame flow field, including the flame-generated flow deflection, flame-generated acceleration, and buoyancy effects.
CCSE is now working with EETD researchers to develop additional statistical measures of both the simulation and experimental data, which will permit even more detailed quantitative comparisons. The solutions will also be used to track the fate of fuel particles as they wash through the flame and undergo a range of chemical reactions, ultimately contributing to pollutant or unburned hydrocarbon emissions, for this and other interesting experimental configurations. Because processing and understanding the wealth of data generated by these simulations continues to raise questions and because simulation of practical-scale combustion devices remains an immense undertaking, CCSE is constantly building and improving innovative diagnostic techniques that will help shed new light on the complex behavior of turbulent reacting flows.
For more information about combustion research in the Environmental Energy Technologies Division, contact:
- Robert Cheng
- (510) 486-5438; fax (510) 486-7303
- Ian Shepherd
- (510) 486-5645; fax (510)486-7303
For more information about CCSE's adaptive methodology for combustion modeling applications, contact:
- Marc Day
- John Bell
Turbulent Premixed Combustion Research at Center for Computational Sciences and Engineering (CCSE)
This work was funded by the U.S. Department of Energy's Office of Science.