Emissions and Thermodynamic Performance Simulation of an Industrial Gas Turbine

Abstract

This paper presents the results of a study on the influence of the fuel gas composition on the thermodynamic performance of a typical heavy duty gas turbine, followed by an emissions prediction using a chemical reactor network which, in its turn, uses one of the most developed chemical kinetic mechanism for the natural gas combustion. The first part of this work regards a statistical analysis of a time series which shows the fuel gas composition variation along a certain period, revealing the typical concentrations of the gas species. Based on the curves provided by the statistics, several gas compositions were proposed and applied to a thermodynamic model in order to evaluate the impact of the species composition of the natural gas on the gas turbine heat rate and power. Afterwards, several approaches of chemical reactor network were considered in order to choose the methodology that better predicts the temperature and CO and NOx formation within a gas turbine combustor operated at base load regime. Finally, it was studied the influence of the natural gas composition variation over the aforementioned thermochemical properties at the exit of the combustion chamber when considering both operation regime and combustor’s natural gas composition. HE modeling of the combustion systems in transforming energy equipments, such as gas turbine combustors, has many challenges especially concerning to the limitations of the predictive capacity of the models based on the Computational Fluids Dynamics (CFD) and their associated computational cost. Thus, when compared to these more elaborate techniques, models based on Chemical Reactor Network (CRN) allies the low computational cost with the ability to predict pollutant emissions, however, without the descriptiveness of the dynamics of the reacting flow. So, since the advent of the “jet engine”, the modeling based on the CRN approach has been an essential tool in the analysis, design and optimization of practical combustion systems, particularly, with the purpose to ensure an efficient operation coupled to low emissions of pollutants. The CRN approach consists in the resolution of simplified transport equations in order to describe the operation of trivial combustion models such as Perfectly Stirred Reactor (PSR) and Plug Flow Reactors (PFR), which are systematically arranged into a structured scheme with the purpose to represent the combustion process occurring within combustion equipment such a gas turbine combustor. This approach is very useful in the prediction of relevant thermochemical properties as temperature pollutants emissions, which are the most significant parameters in the operation of those equipments. This is the reason why, the choice of a detailed kinetic mechanism that represents the oxidation of any fuel involved in the combustion of gas turbines requires the previous identification of the associated validity range. The principal aim of this work is the determination of the influence of natural gas composition on energetic and emissions performance of an industrial gas turbine. For this purpose, this paper is composed of four sections. In the second section, a linear regression statistical analysis is applied to a data set corresponding to the natural gas composition that is daily provided to an industrial gas turbine installation, in order to establish the variations of each hydrocarbon and the plausible range to be analyzed. Afterwards, a model was built in thermodynamic software in order to estimate the temperature and the pressure values after the combustor, as