Estimation of flame temperature of porous radiant burner using artificial neural network.

Artificial neural network (ANN) is employed to estimate the average temperature of the solid- and the gas-phase, in porous radiant burner (PRB). Experimental data is extracted from literature, and used to train the ANN. Levenberg-Marquardt algorithm is used for training of the ANN. Once trained, the ANN is used to generate temperature corresponding to conditions not given in the initial data set. This approach is very useful to understand problems involving PRB, and hence reduce the requirement of costly numerical and experimental methods. [ABSTRACT FROM AUTHOR]