A Taguchi approach for numerical investigation of CO emission from a non-premixed methane–air flame

In this study, carbon monoxide pollutant emission from a non-premixed methane–air flame has been numerically investigated and optimum design condition which emitted the minimum amount of CO pollutant has been obtained using Taguchi method. A L 9 (3 3 ) Taguchi orthogonal array was utilized to design the experimental plan. The signal to noise ( S / N ) ratio was employed to obtain the optimal levels and to analyze the influence of the operation conditions on CO emission values. Furthermore, the analysis of variance was applied to find the predominant factors affecting CO emissions. The design factors and their levels were furnace length at 340, 400 and 460 mm, swirl number of 0, 0.2 and 0.4, air preheat temperature at 300, 350 and 400 K, and percent excess air of 0, 10 and 20 %. The results revealed that air preheat temperature and furnace length were the most effective factors on CO emissions, while swirl number and percent excess air had less effect on this pollutant emission. Confirmation tests with the optimal levels of design parameters were conducted to assess the predictive capability of the Taguchi optimization method. Results of confirmation runs showed good agreement with the predicted values based on Taguchi method. It was, thus, shown that the Taguchi approach is appropriate to study the problems that attempts to achieve minimum CO emissions from diffusion flames with the minimum experimentation.