RANS Simulations of Interaction between Premixed Flame and Turbulence using OpenFOAM Library

Doktorsavhandling, 2015

Unsteady multi-dimensional numerical simulation of turbulent flames is a widely recognized tool for research and development of future internal combustion engines capable of satisfying stringent requirements for ultra-low emission and highly efficient energy conversion. To attain success, such simulations need, in particular, elaborated CFD software and predictive models of turbulent burning. Accordingly, the present work aimed at implementing advanced models of flame turbulence interaction into an open-source CFD package called OpenFOAM, which has been attracting increasing amounts of attention from both commercial and academic organizations, but has yet been rarely assessed in CFD studies of premixed turbulent flames. Subsequently, the extended code was applied to RANS simulations of various experiments with premixed turbulent flames in order to validate the code and the models implemented into it. More specifically, first, the influence of turbulence on combustion was addressed using the so-called Turbulent Flame Closure (TFC) and Flame Speed Closure (FSC) models. The former model has already been tested against a wide range of targets, whereas its extension known as the FSC model has yet been mainly validated against experimental data obtained from expanding flames. In the present study, the two models were implemented into OpenFOAM and the so-extended code was successfully applied to RANS simulations of four widely recognized sets of experiments with substantially different laboratory premixed turbulent flames. Simulations of these four experiments provided detail assessment of the TFC and FSC models by varying flame configuration, turbulence intensity, ambient temperature, mixture composition, etc. Moreover, because all the simulated flames were statistically stationary, the performed tests substantially extended the domain of validation of the FSC model. Numerical results obtained using the FSC model agree both qualitatively and quantitatively with the experimental data, thus, validating both the model and its implementation into OpenFOAM. Second, in order to address the influence of combustion on turbulent transport, a recent simple model of turbulent scalar flux in premixed flames was extended and implemented into OpenFOAM. Subsequently, the so-extended code was applied to RANS simulations of two widely recognized sets of experiments in that countergradient turbulent scalar flux was documented. Obtained numerical results indicate that the model is capable of predicting the countergradient turbulent scalar transport in weakly turbulent flames, as well as transition to gradient scalar transport with an increase in a ratio of the rms turbulent velocity to the laminar flame speed. T hese results are encouraging and justify further joint applications of (i) the FSC model of the influence of turbulence on combustion and (ii) the simple model of turbulent transport in flames to RANS simulations of burning in engines. The joint use of the two models allows researchers to evaluate the mean rate of product creation on a post-processing stage and, subsequently, to use this rate for simulating emissions.