Kinetic mechanism reduction

In CFD simulations, solving the system of equations describing the oxidation of fuels dominates the computational cost and becomes prohibitive when several tens of species are included. Therefore, when complex fuel kinetic mechanisms are required, the application of CFD is strongly limited, since hundreds of species and thousands of reactions are involved. In that context, we are developing different techniques to accommodate detailed kinetic mechanisms in CFD simulations.

Tabulation of Dynamic Adaptive Chemistry


 The Tabulation of Dynamic Adaptive Chemistry (TDAC) method accommodates large kinetic mechanisms with CFD. It consists of two intrinsically coupled layers: tabulation and mechanism reduction. The tabulation layer tries to retrieve the solution of the ODE system at the query composition by linear approximation. For the cells where this fails, the mechanism reduction layer reduces the mechanism at runtime. It then only provides the active species to the ODE solver, which computes the mapping. ISAT grows or adds a point and extend this mapping to the full composition space.

The TDAC method has been implemented in the official OpenFOAM release. More details are provided in the following publications: [1] F. Contino, H. Jeanmart, T. Lucchini, and G. D’Errico. Coupling of in situ adaptive tabulation and dynamic adaptive chemistry: An effective method for solving combustion in engine simulations. Proc. Combust. Inst., 33(2):3057–3064, 2011. link [2]  F. Contino, F. Foucher, P. Dagaut, T. Lucchini, G. D’Errico, and C. Mounaïm-Rousselle. Experimental and numerical analysis of nitric oxide effect on the ignition of iso-octane in a single cylinder hcci engine. Combust. Flame, 160(8):1476–1483,  2013. link [3] F. Contino, J.-B. Masurier, F. Foucher, T. Lucchini, G. D’Errico, and P. Dagaut. CFD simulations using the TDAC method to model iso-octane combustion for a large range of ozone seeding and temperature conditions in a single cylinder HCCI engine. Fuel, 137(0):179–184, 2014. link [4] N. Bourgeois, S. S. Goldsborough, G. Vanhove, M. Duponcheel, H. Jeanmart, and F. Contino. Cfd simulations of rapid compression machines using detailed chemistry: Impact of multi-dimensional effects on the auto-ignition of the iso-octane. Proceedings of the Combustion Institute, 36(1):383–391, 2017. link

Principal Component Analysis

PrinciPCA_scheme.0011pal component analysis (PCA) has recently received strong attention  for its use in combustion modelling. Several advantages of PCA include: its ability to identify orthogonal variables which are the best linear representation of the system; its ability to reduce in dimensionality requiring fewer coordinates; and the ability to generate the models using canonical systems, such as laminar premised flames, counter diffusion flames or empirical data-sets. The PCA modelling framework has been demonstrated a priori and a posteriori for a wide range of configurations, including simple batch and perfectly stirred reactors, one-dimensional laminar flames and complex cases such as flame-vortex interaction as well as plasma flows. Results indicated that PCA-based models are able to provide very accurate results when compared to full size simulations. Moreover, it was shown that the principal components from simple systems could be employed for simulating more complex ones, indicating a relative invariance of the chemical manifold. This represents a very attracting feature of the models, which can be built from inexpensive simulations. Finally, the development of a new methodology for the identification of chemical time scales in reacting flows showed that PCA allows focusing on the slow scales of the process, leading to a stiffness reduction.

More details on PCA models models can be found here: [1] Isaac, B., Thornock, J., Sutherland, J., Smith, P., & Parente, A. (2015). Advanced regression methods for combustion modelling using principal components. Combustion and flame, in press. link [2] Isaac, B., Coussement, A., Gicquel, O., Smith, P., & Parente, A. (2014). Reduced order PCA models for chemical reacting flows. Combustion and flame, 161, 2785–2800.  link [3] Sutherland, J., & Parente, A. (2009). Combustion modeling using Principal Component Analysis. Proceedings of the Combustion Institute, 32, 1563-1570. link