Axial Compressor Fouling and its Effect on Gas Turbine Fuel Consumption and Emissions

Abstract

The literature contains many studies relating gas turbine axial compressor degradation with impurities entering the inlet flow path. The collision and adhering of these impurities onto compressor blades can produce both recoverable and non-recoverable degradation in the compressor, leading to lost overall efficiency. More recently, the topic of how to optimally schedule when compressor washes should occur has been investigated. In many of the scheduling optimization studies, the cost associated with the change in fuel flow for a given fouled state is characterized indirectly as part of an overall fuel calculation for a given fouled state. In this paper, we aim to quantitatively answer the question of what this small quantity of fuel due to compressor fouling actually is. Additionally, we show how this can be included in solving for optimal schedules of compressor washes and estimate the CO2 emissions attributable to compressor fouling. We begin with a look at the literature on compressor fouling to identify common methods and assumptions of how fouling can manifest or be imposed. We then derive a first principles, steady state equation to relate compressor performance parameters with fuel flow. We validate this equation at nominal conditions with data obtained from Solar Turbine’s test cells. Additionally, we develop a generic industrial gas turbine model using the Numerical Propulsion System Simulation software and apply the fouling constraints found from the literature review to run a fouling simulation. The results further validate our equation and provide insight into the quantity of fuel resulting from compressor fouling. We end the paper with a look at how the derived model can be used in the problem of optimizing compressor wash schedules as well as give an estimate on increases in CO2 emissions due to compressor fouling.

Application of ROM-based methods to aeroengines health monitoring, performance, and control

“…Additionally, the application between both sets, the model of the engine (provided by PROOSIS ® ), is indicated with dotted arrows between sets (also the norm). Those sets could be considered, for the sake of simplicity and without lack of generality, to be inside ℝ 11 .…”

“…Customers usually focus on EGT increases or power losses, magnitudes easy to measure, to decide when a water wash must be done. Planning an effective on-line water wash program needs to consider potential fuel losses, materials, and labor required as well as lost production (see [198] and [11]). The proposed methodology could help in this sense as the degradations provided (clear physical meaning) are directly related with the increments of TSFC, which could be calculated by PROOSIS ® .…”