Experimental study of jet fire radiation and a new approach for optimizing the weighted multi-point source model by inverse methods

Abstract

The present study brings new experimental data on radiation from subsonic jet fires and proposes a new method for employing the inverse analysis technique to optimize the weighting coefficients of the multi-point model. First, detailed measurements of radiative heat flux distributions are obtained for different flame powers, injector diameters and sensor distances. All results fall onto a single curve with an adequate non-dimensionalization. A new correlation for the radiative heat transfer rate from these flames was obtained with a better performance compared to other options from literature. Then, an inverse method is employed to obtain the coefficients of the weighted multi-point model for predicting the radiative heat transfer from flames. By using the non-dimensional heat flux distributions, the inverse method became independent of the flame radiated fraction. A new optimization parameter that emerges from the non-dimensional model showed the consistency of the inverse method and leads to a correlation for the maximum heat flux as a function of the radial distance from the flame. Finally, the new weighting coefficients and the radiative heat transfer rate correlation were used to compute the dimensional radiative heat flux distributions. The comparison with the experimental data showed expressive improvements of the model prediction.