Numerical study of effects of solid particle erosion on compressor and engine performance

Abstract

Solid particle erosion (SPE) is a major source of damage to, and cause of failure of, turbomachinery, including centrifugal and axial compressors and gas turbines. There is considerable interest within the turbomachinery industry to develop the means to enhance durability and improve prediction of likely breakdowns of machinery operating in locations and conditions where ingestion of, for example, dust and sand grains is unavoidable during operation. The present study examines the implications of degradation of the compressor on gas turbine cycle performance, caused by operating in eroding environments, linking compressor aerodynamics with the thermodynamic cycle. The research engine employed in the study comprised a fan, bypass, and two stages of the low-pressure compressor (booster). A numerical simulation using Computational Fluid Dynamics (CFD) software was performed to predict the likely erosion patterns due to solid particles (SPs) impacting on the blades of the first two stages of an axial compressor. This study introduces a method that provides the freedom to apply roughness to a particular region of the blade following the SPE patterns identified as part of the CFD simulations so as to represent the effects of the erosion spots on the overall flow field. The localised roughness method employs a perturbation of the geometry of the blade in the regions corresponding to the location predicted by the SPE model. The intensity of the perturbation, which is governed by a “fraction factor” coded in a Matlab script, was calibrated by reference to a roughness case, applied over the entire blade surface, with equivalent sand-grain roughness (ks) of 60 μm. Turbomatch, the Cranfield in-house gas turbine performance simulation software, was employed to model the degradation of engine performance. The research confirmed that SPE is linked with significant decreases in engine performance parameters such as isentropic efficiency (ηis) and pressure ratio (PR). These quantities showed, for the SPE cases considered, a drop of 8.8% and 5.2%, respectively. These findings are useful to link local SPE events to global GTE cycle performance.