Page 78 - IJAMD-1-3

International Journal of AI for
Materials and Design
Review of gas turbine blade failures by erosion

2.2. CFD and FEA turbulence and particle interactions in high-detail. LES, in

Advancements in CFD and FEA continue to provide particular, has been shown to provide more accurate results
detailed insights into the behavior of solid particles within compared to traditional Reynolds–Averaged Navier–
GTEs. These methods simulate the interaction of fluid flow Stokes (RANS) models, especially in simulating highly
with turbine blades, offering a deeper understanding of turbulent flows around turbine blades. In gas turbines,
how erosion patterns develop under varying operational flow conditions are often turbulent. Models such as LES
conditions. 34,53 When combined with ML, these simulations and RANS are used to model turbulence. LES resolves
can be optimized to predict critical erosion-prone zones, the large-scale turbulent structures while modeling the
providing more effective erosion control measures. smaller scales, which is crucial for accurately predicting
Incorporating such methods helps bridge the gap between particle impacts in highly turbulent flow regions. Rivaz
43
theoretical understanding and practical solutions. The et al. applied LES to model gas-particle interactions in
application of CFD and FEA has become indispensable in a turbine operating at high temperatures, revealing the
understanding the complex fluid-structure interactions, intricate erosion patterns caused by turbulent eddies near
material degradation, and failure modes associated with the blade’s trailing edge. Their research demonstrated that
gas turbine blade erosion. 35,36,54 Both techniques allow more accurate CFD models can significantly improve the
researchers to simulate real-world operational conditions prediction of erosion-prone zones and inform better blade
in a virtual environment, thus offering deeper insights design. Table 8 presents computational models for erosion
into the mechanisms of erosion and enabling more precise prediction and their different aspects.
design optimization. 37,55,56 Recent advancements in CFD 2.2.2. FEA for structural deformation and erosion
and FEA, when combined with ML, have significantly effects
enhanced the ability to predict erosion-prone areas,
optimize blade geometries, and select appropriate materials While CFD primarily focuses on the fluid dynamics aspect,
and coatings for gas turbine blades. FEA is essential for understanding the structural response
of turbine blades to erosion-induced damage. FEA allows
2.2.1. CFD for erosion prediction and flow analysis researchers to simulate the mechanical stresses, strains,
CFDs play a critical role in simulating the flow of gases and and material deformation that occur in turbine blades
particles within GTEs, which is essential for understanding under operational loads. 44,45,59 Erosion, particularly when
the dynamics of erosion. By modeling the airflow combined with high thermal and mechanical stresses,
57
patterns and particle trajectories, CFD provides a detailed can lead to significant structural degradation, and FEA
visualization of how solid particles interact with turbine provides a framework for assessing the extent of this
blades at various velocities, temperatures, and angles of damage.
impingement. 38,39,58 The resulting simulations help identify Wei et al. employed FEA to simulate the impact
46
regions of high erosion risk, enabling engineers to design of repeated particle impingement on turbine blade
blades that minimize erosion damage. surfaces, showing that erosion leads to micro-cracking
In their study, Peng et al. employed CFD to simulate and pitting, which weakens the material over time. Their
40
the behavior of solid particles in the gas flow of a turbine. FEA model incorporated both thermal and mechanical
They found that particles with higher velocities tend loads, providing a more comprehensive understanding of
to follow more erratic trajectories, which significantly how erosion accelerates fatigue failure in turbine blades.
increases erosion near the leading edge of turbine blades. Their study concluded that the combination of erosion
Their research demonstrated that regions experiencing and thermal cycling creates stress concentrations, which
the highest flow velocity and turbulence were the most propagate cracks and lead to premature blade failure.
vulnerable to particle impingement. Alqallaf and Teixeira Similarly, Mortazavi et al. used FEA to analyze the
41
47
expanded on this by integrating CFD with Eulerian– effects of particle size, impact velocity, and angle on the
Lagrangian models to predict erosion rates more accurately, deformation of turbine blades. Their study found that
considering both the fluid flow and the particle interaction larger particles caused deeper pits and more extensive
with the blade surface. Their results highlighted that, while material removal, while higher velocities increased the
particle size and velocity are important factors, the angle of plastic deformation of the blade surface. The integration
particle impact is critical in determining erosion severity. of FEA with CFD, as performed by Prashar and Vasudev,
48
Further advancements in CFD, as shown by Goswami provided even more precise results. By coupling fluid flow
et al., involve the use of large eddy simulations (LES) and data from CFD with the structural analysis capabilities
42
direct numerical simulations (DNS) to capture small-scale of FEA, they were able to model not only the particle

Volume 1 Issue 3 (2024) 72 doi: 10.36922/ijamd.5188