A Lagrangian Analysis of Combustion Regimes Using Multi-modal Turbulent Combustion Model

High Reynolds number turbulent reacting flows poses a modeling challenge due to the multi-regime, mixed-mode nature of the combustion processes. This study presents a novel unified index that includes the description of both premixed and nonpremixed combustion regimes and the occurrences of local extinction and re-ignition. This classifier is applied to large eddy simulations of the Darmstadt multi-regime burner for the nominal cases 18b and 26b. Lagrangian particles are transported along with the flow in order to monitor the evolution of the local flow-chemistry interaction. The simulations are validated against experimental data, and the Lagrangian properties are compared against the traditional premixed model in progress variable space and a generalized multi-modal manifold model in mixture fraction and generalized progress variable space. The comparison reveals that minor radical species are sensitive to the generalized progress variable dissipation rates, and the multi-modal manifold model is more suitable to reproduce the multi-regime flame structure. Using the multi-modal framework, the regime index is simply defined by the slope of the Lagrangian particle trajectory in the phase space and is able to detect the evolution of the combustion regimes and the occurrence of extinction events. Consistent with the previous experimental findings, the statistics of the regimes reveals that the leaner case 18b is more susceptible to extinction and re-ignition and the regime is predominantly premixed, while the richer 26b case is more pronounced in the nonpremixed regime.