| Abstrakt: |
Combustion phenomenon in combustion chambers used in gas turbines is a complex process in which various factors are involved. Energy loss in these combustion chambers due to chemical reaction factors, internal heat transfer, mass transfer and viscous losses reduces the efficiency of these units and ultimately greatly reduces the overall efficiency of a gas turbine unit. Therefore, it will be very useful to provide a method by which the combustion process and the type of flame can be modelled. Since the fuel used in combustion chambers as an energy carrier may change, it is both time-consuming and costly to perform testing processes under the conditions of using new fuels to determine operating point parameters, so the novelty of this paper presents a general approach. It can be used for any type of fuel only by changing the environmental parameters. The β-PDF approach is proposed to model the thermochemical scalars (i.e. temperature, species mass fractions, and density) as functions of mixture fraction by assuming the fast chemistry. Next, the entropy generation analysis is applied to quantify the contribution of each irreversible process (i.e. heat transfer, mass transfer, chemical reaction, and viscous dissipation) to the total exergy destruction, and theoretical findings are finally presented to relate the exergy losses to the design parameters of the combustor. In this regard, a well-known non-premixed jet flame is considered as the case study in this paper, named the Sandia/ETH H2/He flame. A comparison of the obtained results based on the proposed modelling method and the experimental results confirms the effectiveness of the proposed estimation strategy. Also, entropy generation analysis of the flame demonstrates that the chemical reaction is the dominant irreversible process in the total exergy destruction in turbulent non-premixed flames (85.13%), followed by mass transfer (8.05%) and heat transfer (6.80%). By modelling any desired flame based on the proposed PDF method, two main goals are achieved: first, all the necessary parameters to determine the physical model of the flame can be estimated, thus avoiding multiple experimental tests. Secondly, it is possible to identify the factors affecting the entropy production and ultimately the effect on heat loss by estimating the model for each flame, and by optimizing these factors, heat loss in combustion chambers can be prevented. [ABSTRACT FROM AUTHOR] |
