Parametric study for reducing emission characteristics using different computational schemes | ||||
International Conference on Aerospace Sciences and Aviation Technology | ||||
Article 43, Volume 18, Issue 18, April 2019, Page 1-17 PDF (1.2 MB) | ||||
Document Type: Original Article | ||||
DOI: 10.1088/1757-899X/610/1/012041 | ||||
View on SCiNiTO | ||||
Author | ||||
Hatem Kayed | ||||
Department of Mechanical Power Engineering, Cairo University, Egypt. | ||||
Abstract | ||||
The present study numerically predicts emissions (NOx & CO) characteristics issued from swirl burner using different computational schemes adopted in ANSYS 17.2. Various turbulence models were tested and compared in a structured grid having an average skewness of 0.04. RSM model shows the best results in predicting the central toroidal recirculation zone for such severe swirling flow (S>0.6). Three different schemes of combustion modeling were also tested and compared. Flamelet PDF model was preferred among other combustion models for its potential to consider non-equilibrium chemistry, where its choice was recommended form many studies in the literature. A radiation heat transfer analysis was carried out using P1 radiation model to account radiation effects in estimating NOx and CO concentrations. The effect of varying swirl ratio, excess air, diluting fuel with N2 and CO2, oxidizer preheating, and oxidizer composition on emissions (NOx and CO) reduction strategy concluded that increasing the swirl number, excess air factor, and (N2 and CO2) mixing ratios have a considerable influence on NOx and CO emissions reduction. While increasing the preheating temperature, and the oxidizer Oxygen concentration leads to an increase the NOx and CO emissions. Hence, a considerable reduction in NOx and CO formation was apparently observed by lowering the peak flame temperature less than 1800 K. These predicted outcomes are in agreement with measurements data published in the literature. | ||||
Keywords | ||||
NOx Emission; Turbulent Combustion; Modelling | ||||
Statistics Article View: 30 PDF Download: 51 |
||||