investigation of the flame temperature for some gaseous fuels using artificial neural network

in the present study the adiabatic temperature of gaseous fuels were calculated and the influence of effective parameters of flame temperature was discussed. firstly, a new computational program named ftc (flame temperature calculations) was prepared to calculate the adiabatic flame temperature and then the effect of initial temperatures of combustion air and fuel, excess air content and oxygen enrichment on these temperatures was evaluated. the obtained results show that the oxygen enrichment influences the adiabatic temperature, remarkably. also, ftc is able to estimate the concentration of combustion components such as: carbon dioxide, steam, oxygen, nitrogen, carbon mono oxide and nitrogen oxide. finally artificial neural networks were presented for estimation of adiabatic temperature. the proper neural networks were trained and tested using obtained data by ftc. the neural network prediction results were compared with those calculated by thermodynamic and chemical equilibrium - based method. it was shown that trained neural networks can provide the adiabatic temperature with reliable accuracy over a wide range of operating conditions.