Imaging based chemiluminescence characterisation of partially premixed syngas flames through DFCD technique

Digital image processing technique for characterising the radiation emission features of partially premixed syngas flames was explored in this investigation. In particular, the DFCD-based (Digital Flame Colour Discrimination) processing methodology was applied. The normalised values obtained from an empirical relation using DFCD-derived variables, which accounts for the image colour, intensity and spatial presence functions, was found to be able to model the normalised spectroscopic derived CO-O* radiation profiles along with changes in equivalence ratios for the partially premixed syngas flames of varying H2/CO proportions, as well as with CO2 addition. Then, the prediction derived from the empirical relation can be correlated by a power factor of y = x1.6 to approximate the normalised spectrometry intensity levels of H2/CO flames. Further correlation for the H2/CO/CO2 flame can be made by introducing a multiplier factor of e−α to the power correlated empirical expression where α denotes the volumetric proportionality ratio of CO2/CO in the syngas composition. In this form, the newly derived expression of y = e−αx1.6 can be applied to predict the relative CO-O* intensity levels for all syngas flames considered in this study. The obtained correlation from DFCD-derived CO-O* intensity variation is highly proportional to the spectroscopic-determined intensity values with R2 of 0.996. Thus, the DFCD-derived parameters demonstrated potential applicability to provide abstract correlations that is linked to the well known physical chemiluminescence emission variation along with changes in combustion conditions. The ability to mimic the 1D spectrometry results is significant as the imaging approach preserves the two-dimensional flame details for further qualitative and quantitative correlations that may yield further useful relationships to better convey the multi-dimensional aspect of the combustion phenomenon.