Investigation of an Industrial Gas Turbine Combustor and Pollutant Formation Using LES

Abstract

An experimental variant of a commercial swirl burner for industrial gas turbine combustors operating at 3 bar is numerically investigated using high-fidelity Computational Fluid Dynamics models using STAR-CCM+ v11.06. This work presents the computational results of the SGT-100 Dry Low Emission gas turbine provided by Siemens Industrial Turbomachinery Ltd. The related experimental study was performed at the DLR Institute of Combustion Technology, Stuttgart, Germany. The objective of this work is to compare the performance of the Flamelet Generated Manifold model, which is the widely accepted combustion model in Gas Turbines with the Complex Chemistry model. In particular this work examines the flame shape and position, pollutant formation predicted by the aforementioned models with Large Eddy Simulations. Mean and RMS quantities of the flow field, flame temperatures and major species are presented and compared with the experiments. The results show that the predictions are insensitive on the meshing strategy and at the evaluated mesh sizes of ∼10 million and ∼44 million cells. The mean and RMS errors are ∼8% compared to the reported experiments and these differences are within the measurement errors. The results show that the calculated flame positions are in very good agreement with the reported measurements and the typical M-shape flame is reproduced independent of the combustion model. Pollutant formation in the combustor predicted by two combustion models is scrutinised. The predicted NO and CO emissions levels are in agreement with the literature.

Reduced Methane Combustion Mechanism and Verification, Validation, and Accreditation (VV&A) in CFD for NO Emission Prediction

“…Considering the simulation accuracy ability and cost, this simulation of the industrial gas turbine combustor is based on 3.5 million meshes. For the experiment comparison with simulation, this paper shows the temperature, CH 4 and CO 2 mass fraction measured by Stopper et al [34] and the outlet NO mass fraction given by Mallouppas et al [35] and Jaravel et al [36]. It should be noted that the experimental errors associated with the temperature measurement are estimated up to 13% for fully reacting products, and the experimental errors associated with species concentration are estimated 4% for CH 4 and 20% for CO 2 [9].…”

“…Therefore, controlling the maximum temperature is an effective way to control the NO emission. For the prediction of pollutant emissions from the combustor exit, this paper compared the NO emission with the measurement given by Mallouppas et al [35] and Jaravel et al [36]. It should be noted that the measurement of pollutants is carried out at 30 mm downstream, thus causing approximately 7.5% of air leakage [35], which is not considered in the simulation.…”

“…For the prediction of pollutant emissions from the combustor exit, this paper compared the NO emission with the measurement given by Mallouppas et al [35] and Jaravel et al [36]. It should be noted that the measurement of pollutants is carried out at 30 mm downstream, thus causing approximately 7.5% of air leakage [35], which is not considered in the simulation. It can be seen from Table 4 that the simulation can well predict the NO concentration and emission index, but the predicted value is 1.810 -6 higher than the measured value.…”

Prediction of flow induced vibration of a flat plate located after a bluff wall mounted obstacle

“…being ∆ = C w V 1/3 a length scale parameter dependent of the cell volume V, and S w is a deformation parameter, dependent of the strain rate tensor. Here, the model constant was set to C w = 0.544, given that this value has been shown to provide acceptable results both for homogeneous isotropic decaying turbulence and for channel flows (see, for example, [23] or Malloupas et al [24]).…”

Longitudinal and azimuthal thermo-acoustic instabilities in an industrial gas turbine combustor operating at elevated pressure

“…It should be noted that several other numerical works in the open literature have also simulated the SGT-100 combustor. However, these works have focused almost exclusively on the more acoustically stable operating condition at 3 bar (Case A) with the aim of evaluating the performance of different combustion, chemistry, radiation or adaptive mesh refinement models [48][49][50][51][52][53][54][55][56][57][58][59][60][61][62][63][64] . An isothermal flow case was furthermore simulated by Bulat et al, 65 who identified both a PVC and a central vortex core inside the combustion chamber, whereas Xia et al 66 studied the dispersion of artificially induced entropy perturbations.…”