Efficient use of energy is one of the major objectives in designing modern energy conversion systems. Entropy generation and exergy loss analysis are useful means of investigating the sources of irreversibility and improving the efficiency of combustion systems. The amount of entropy generated in a system is directly connected with the amount of available work and hence, the energy efficiency of the system.

A new methodology is being developed for LES of turbulent reacting flows, incorporating the second law of thermodynamics. The entropy transport equation is introduced in LES. The filtered form of this equation includes the effect of unclosed subgrid scale entropy generation. The filtered density function (FDF) methodology provides an effective means to close these terms. The FDF includes the complete statistical information about the joint entropy, turbulent frequency, velocity and scalar within the SGS. An exact transport equation is developed for the FDF, which includes the effects of chemical reaction in closed forms. The unclosed terms in this equation are modeled by considering a system of stochastic differential equations. The modeled FDF transport equation is solved by a Lagrangian Monte Carlo method. LES/FDF is employed to simulate a 3D turbulent temporally developing mixing layer, involving transport of scalars and entropy as shown in the figure above.

Figures below show some of the comparisons with the DNS data. The one on the rightÂ shows the variation of filtered entropy across the layer. The one on the left, shows the average filtered entropy generation rate which includes the contributions due to heat and mass transfer and turbulent dissipation.

]]>

The main objective of the this research is to study the fundamentals of turbulent reacting flows by conducting high-fidelity simulation of turbulent flames. Our main objective is to investigate detailed phenomena as related to energy conversion and combustion-based systems. The unsteady physics of such flows are optimally captured by large eddy simulation (LES). The closure is based on filtered density function (FDF), proven to be particularly effective for reacting flows due to its capacity to account for chemical reaction effect in an exact manner. Our past efforts have shown the effectiveness of this methodology in predicting turbulence-chemistry interactions. Figures below, show some of the FDF predictions of Sandia jet flames. The present work deals with consideration of more complex chemical kinetics. This is a very important issue in quantitative prediction of turbulent combustion in which detailed phenomena depends strongly on key chemical reactions and on the accuracy with which predictions are made.

Turbulence influenced phenomena such as local extinction/reignition in turbulent reacting flows can be studied by simulation of close to extinction flames such as Sandia Flame F. This flame shows significant level of local extinction, accurate prediction of which requires consideration of detailed chemistry. In figure below, the local extinction in flame F is evident from scatter plot of mass fraction of CO2 against mixture fraction. The results show favorable agreements between LES/FDF predictions and experimental data.

]]>