The Study Of Gas Turbine Combustor Using Chemical Reactor Network Approach

The design of low NOx emission gas turbine combustor is an important part of gas turbine design. This paper focused on the defect of traditional CFD methods of combustor design in computational time and precision, aiming to develop Chemical Reactor Networks (CRN) method, a low computational effort modeling approach, which can quickly assess an industrial gas turbine NOx emission.First of all, it takes a review of the emergence and the development of the CRN method. It analyzes the normal configuration of combustor with a diffusion flame type, and raises a proposal that divides the inlet into three fluids based on the mixture combustible limit.Moreover, it takes this method to analyze a combustor chamber of a gas turbine with multiple-can combustor with a diffusion flame type. The combustor is fueled with syngas gasified from the coal. It takes this method to predict NOx emissions of combustion chamber outlet with variable steam dilution range and variable pressures. Simulation results of variable steam dilution range have been compared with those obtained from combustion experimental campaign. Results predicted with CRN approach closely matches with experimental results. The simulation results state that the NOx emission increases linearly with the P^0.67, according with the literature. From this simulation, it confirm that the CRN construction proposal is reliable in variable working conditions.And then, it takes this method to predict NOx emission of a gas turbine fueled with medium-heating-value syngas changing from nature gas. It is a multiple-can combustor with multiple diffusion flame type nozzle. The structure of combustor is combined can combustor with annulus combustor in series. It take the simulation of combustor in rating condition and three variable steam dilution range conditions. The simulation results and the emission change trend in variable pressure are reasonable.Finally, it takes this method to predict NOx emission of a gas turbine with lean-premixed combustor. According to the analysis of the normal fluid filed conformation of lean-premixed combustor and CFD results, the combustor space is divided into several main character sub-spaces, and the CRN model is built base on those character sub-spaces. With the consideration of unsteady in lean-premixed combustion, the CRN model is taken to consider the influence of the different fuel flux, the predicted results and the trend of variable state appear reasonably.After comparing the simulation results with the experimental results and analyzing the predicted results, this paper draws the conclusion that CRN approach has high accuracy in predicting gas turbine combustor NOx emission and it can be helpful in combustor design and optimization.