A coupled flow and chemical reactor network model for predicting gas turbine combustor performance

Abstract

Gas turbine combustor performance was explored by utilizing a 1-D flow network model. To obtain the preliminary performance of combustion chamber, three different flow network solvers were coupled with a chemical reactor network scheme. These flow solvers were developed via simplified, segregated and direct solutions of the nodal equations. Flow models were utilized to predict the flow field, pressure, density and temperature distribution inside the chamber network. The network model followed a segregated flow and chemical network scheme, and could supply information about the pressure drop, nodal pressure, average temperature, species distribution, and flow split. For the verification of the model's results, analyses were performed using CFD on a seven-stage annular test combustor from TUSAS Engine Industries, and the results were then compared with actual performance tests of the combustor. The results showed that the preliminary performance predictor code accurately estimated the flow distribution. Pressure distribution was also consistent with the CFD results, but with varying levels of conformity. The same was true for the average temperature predictions of the inner combustor at the dilution and exit zones. However, the reactor network predicted higher elemental temperatures at the entry zones.

Developing a Turbulence-Inspired Zonal Combustion Modeling Approach, Implementing the CRNs in Turbulent Diffusion Flames

“…[44], Hataysal, and Yozgatligil, [45], propose a hybrid CFD-CRN approach for the prediction of NOx and CO emissions and gas turbine combustor performance, respectively. The real-time monitoring approach based on CRNs is explored by [ [46], [47]].…”