428964 Optimizing a Global Kinetic Mechanism for Use in CFD Modeling of Jet-a Combustion

Tuesday, November 10, 2015: 1:30 PM
355F (Salt Palace Convention Center)
Megan Karalus, CD-adapco, Bellevue, WA, Niveditha Krishnamoorthy, CD-adapco, Irvine, CA and Ravindra Aglave, CD-adapco, Houston, TX

Prediction of CO or NOx emissions is difficult due the influence of detailed chemistry. Yet predicting these emissions is very much desired due to the ever stricter environmental regulations. Use of detailed chemistry in a full 3D, CFD simulation is prohibitive due to the large number of stiff reactions involved. Therefore global mechanisms with limited reactions are preferred. Though valid in only a narrow range of operating conditions, they open an opportunity to get accurate CO predictions when tuned correctly to a given set of experimental data or results from a detailed reaction mechanism. This tuning can be a cumbersome process if done manually, especially if it needs to be repeated for a variety of operating conditions.

In this study, an automated optimization of a 4-step global mechanism describing Jet-A combustion is carried out. HEEDS, a design exploration tool, is paired with DARS-Basic, a CAE tool focused on the simulation of complex chemical kinetics in 0D and 1D reacting models. Both are CD-adapco software products.

In this presentation the basics of the optimization methodology are described. As an example, a 1D model of a freely propagating flame is chosen, with both laminar flame speed and CO profiles identified as the responses of interest. The target values for the study are generated using Dagaut’s [1] detailed mechanism containing 209 species and 1650 reactions. Four objectives are identified, all of which are created in HEEDS as curve fit responses. The first curve is laminar flame speed as a function of inlet fuel/air equivalence ratio (0.5 to 1.7). The other curves are CO as a function of temperature for three representative equivalence ratios: a lean condition (0.6), the stoichiometric condition (1.0) and a rich condition (1.4).

An overall optimized global mechanism is identified which significantly improves upon the baseline mechanism in predicting flame speed and CO profiles. In addition, a trade-off study is performed to look at the prediction of laminar flame speeds vs. CO profiles. This provides insight about the limitations of the chosen global mechanism.

1. Dagaut, P. & Cong, T.L., Experimental and Detailed Kinetic Modeling of the Oxidation of Methane and Methane/Syngas Mixtures and Effect of Carbon Dioxide Addition, Combustion Science and Technology Volume 180, Issue 10-11, 2008.

Extended Abstract: File Not Uploaded