Validating Numerical CFD Simulations With Experimental Data for Turbulence Phenomena in Axial Flow Gas Turbine Diffusers

Scale models give engineers an excellent understanding of the aerodynamic behavior behind their design; nevertheless, scale models are time consuming and expensive. Therefore computer simulations such as Computational Fluid Dynamics (CFD) are an excellent alternative to scale models. One must ask the question, how close are the CFD results to the actual fluid behavior of the scale model? In order to answer this question the engineering team investigated the performance of a large industrial Gas Turbine (GT) exhaust diffuser scale model with performance predicted by commercially available CFD software. The experimental results were obtained from a 1:12 scale model of a GT exhaust diffuser with a fixed row of blades to simulate the swirl generated by the last row of turbine blades five blade configurations. This work is to validate the effect of the turbulent inlet conditions on an axial diffuser, both on the experimental front and on the numerical analysis approach. The object of this work is to bring forward a better understanding of velocity and static pressure profiles along the gas turbine diffusers and to provide an accurate experimental data set to validate the CFD prediction. For the CFD aspect, ANSYS CFX software was chosen as the solver. Two different types of mesh (hexagonal and tetrahedral) will be compared to the experimental results. It is understood that hexagonal (HEX) meshes are more time consuming and more computationally demanding, they are less prone to mesh sensitivity and have the tendancy to converge at a faster rate than the tetrahedral (TET) mesh. It was found that the HEX mesh was able to generate more consistent results and had less error than TET mesh.

Fully Parametric High-Fidelity CFD Model for the Design Optimisation of the Cyclic Stagger Pattern of a Set of Fan Outlet Guide Vanes

The present paper deals with the redesign of cyclic variation of a set of fan outlet guide vanes by means of high-fidelity full-annulus CFD. The necessity for the aerodynamic redesign originated from a change to the original project requirement, when the customer requested an increase in specific thrust above the original engine specification. The main objectives of this paper are: 1) make use of 3D CFD simulations to accurately model the flow field and identify high-loss regions; 2) elaborate an effective optimisation strategy using engineering judgement in order to define realistic objectives, constraints and design variables; 3) emphasise the importance of parametric geometry modelling and meshing for automatic design optimisation of complex turbomachinery configurations; 4) illustrate that the combination of advanced optimisation algorithms and aerodynamic expertise can lead to successful optimisations of complex turbomachinery components within practical time and costs constrains. The current design optimisation exercise was carried out using an in-house set of software tools to mesh, resolve, analyse and optimise turbomachinery components by means of Reynolds-averaged Navier-Stokes simulations. The original configuration was analysed using the 3D CFD model and thereafter assessed against experimental data and flow visualisations. The main objective of this phase was to acquire a deep insight of the aerodynamics and the loss mechanisms. This was important to appropriately limit the design scope and to drive the optimisation in the desirable direction with a limited number of design variables. A mesh sensitivity study was performed in order to minimise computational costs. Partially converged CFD solutions with restart and response surface models were used to speed up the optimisation loop. Finally, the single-point optimised circumferential stagger pattern was manually adjusted to increase the robustness of the design at other flight operating conditions. Overall, the optimisation resulted in a major loss reduction and increased operating range. Most important, it provided the project with an alternative and improved design within the time schedule requested and demonstrated that CFD tools can be used effectively not only for the analysis but also to provide new design solutions as a matter of routine even for very complex geometry configurations.

The Application of Non-Axisymmetric Endwall Contouring in a Single Stage, Rotating Turbine

As turbine manufacturers strive to develop machines that are more efficient, one area of focus has been the control of secondary flows. To a large extent these methods have been developed through the use of computational fluid dynamics and detailed measurements in linear and annular cascades and proven in full scale engine tests. This study utilises 5-hole probe measurements in a low speed, model turbine in conjunction with computational fluid dynamics to gain a more detailed understanding of the influence of a generic endwall design on the structure of secondary flows within the rotor. This work is aimed at understanding the influence of such endwalls on the structure of secondary flows in the presence of inlet skew, unsteadiness and rotational forces. Results indicate a 0.4% improvement in rotor efficiency as a result of the application of the generic non-axisymmetric endwall contouring. CFD results indicate a clear weakening of the cross passage pressure gradient, but there are also indications that custom endwalls could further improve the gains. Evidence of the influence of endwall contouring on tip clearance flows is also presented.

A New Concept of a Two-Dimensional Supersonic Relative Inlet Mach Number Compressor Cascade

In this study, a numerical shape optimization method based on evolutionary algorithms coupled with a verified CFD solver has been applied to the ambitious target of a shock free 2-D supersonic inlet Mach number compressor cascade. The study is based on the DLR-PAV-1.5 supersonic compressor cascade designed by the pre-compression blading concept. The DLR cascade airfoil has been optimized using a verified CFD code. A superior performance of the optimized supersonic cascade with about 24% reduction of the total pressure loss coefficient compared to the original cascade has been realized. The flow mechanisms observable around the blade with improved performance and the resulting design concept are discussed in this paper.

Influence of Labyrinth Seal Leakage on Centrifugal Compressor Performance

The focus of this paper lies on the leakage flow across the shroud of a centrifugal compressor impeller. It is common practice to use shrouded impellers in multi stage compressors featuring a single shaft. The rotating impeller then has to be sealed against the higher pressure in the downstream diffuser by means of labyrinths. The relative amount of leakage is higher for stages designed for low flow, meaning that the associated losses gain in relevance. In addition to this loss source, the injection of the leakage flow has a serious influence on the main flow in a region where it is prone to separation, i.e. at the suction side of the impeller blades close to the shroud, where the highest relative velocities are found. The present paper discusses the numerical results of several geometrical arrangements where the leakage flow was mixed with the main flow in different ways. The distance between the location of injection and the leading edge of the impeller as well as the orientation of the injected flow showed a distinct influence on the performance of the entire stage, mainly on stability.

Gas Temperature Profile Attenuation Through a Multistage Axial-Flow Turbine

Nowadays 2D through-flow models are widespread for designing and analysis of a turbine. Unlike 2D calculations, the measurements show that a non-uniform inlet gas temperature profile is strongly attenuated to the outlet of a turbine. This attenuation can be taken into account in through-flow models only using some corrective coefficients. The objective of this work was to find such an empirical coefficient. The results of full-scale tests of several power GTUs and aviation GTEs were employed to obtain values of the temperature profile attenuation coefficient in the through-flow model for various span locations of airfoil rows. During the tests detailed radial-circumferential distributions of the gas temperature upstream of each of rows and downstream of the turbines were measured (in absolute motion for stator and in relative motion for rotor). The values of the attenuation coefficient for airfoil rows of the three front stages were obtained by means comparison of experimental and computed results. The experience shows that the attenuation coefficient is easily incorporated into the 2D gas-dynamic codes. This incorporation allows spanwise distributions of flow parameters to be predicted and the airfoil geometry and cooling mass flow to be chosen more correctly.

Multistage Compressor and Turbine Modeling for the Prediction of the Maximum Turbine Speed Resulting From Shaft Breakage

A mathematical model for the prediction of the maximum speed of a high pressure turbine following a shaft failure event was developed. The model predicts the high pressure compressor and ducting system pre- and post-stall behavior like rotating stall and surge after the shaft breakage. The corresponding time-dependent high pressure turbine inlet conditions are used to calculate the turbine maximum speed, taking into account friction and blade&vane tip clearance variations as a result of the rearward movement of the turbine and destruction of the turbine blading. The compressor and ducting system is modeled by a 1-dimensional, stage-by-stage approach. The approach uses a finite-difference numerical technique to solve the nonlinear system of equations for continuity, momentum and energy including source terms for the compressible flow through inlet ducting, compressor and combustor. The compressor blade forces and shaft work are provided by a set of quasi steady state stage characteristics being valid for pre-stall and post-stall operations. The maximum turbine speed is calculated from a thermodynamic turbine stand-alone model, derived from a performance synthesis program. Friction and blade&vane tip clearance variations are determined iteratively from graphical data depending on the axial rearward movement of the turbine. The compressor and ducting system model was validated in pre-stall and post-stall operation mode with measured high pressure compressor data of a modern 2-shaft engine. The turbine model was validated with measured intermediate pressure shaft failure data of a 3-shaft engine. The shaft failure model was applied on a modern 2-shaft engine. The model was used to carry out a sensitivity study to demonstrate the impact of control system reactions on the resulting maximum high pressure turbine speed following a shaft failure event.

Effects of Endwall Profiling on Axial Flow Compressor Stage

In recent years endwall profiling has been well validated as a major new engineering design tool for the reduction of secondary loss in turbines. However, its application on compressors have been rarely performed and reported. This paper documents the findings of the analysis for diminishing compressor stator corner separation using endwall profiling; In the study, novel profiled endwalls were designed and numerically studied on a subsonic axial-flow compressor stage. The compressor stator endwalls were profiled on both axial and azimuthal directions. The results showed, the stator corner separation was significantly suppressed under all the operating conditions by implementing this profiled endwall. Significant improvements on stage pressure ratios and stage efficiency were observed. Detailed flow field changes, as well as endwall profiling methods are provided in the paper, so that the results of this research can be referenced to other compressor designs.

A Hybrid Unstructured/Spectral Method for the Resolution of Navier-Stokes Equations

A novel high-order finite volume method for the resolution of the Navier-Stokes equations is presented. The approach combines a third order finite volume method in an unstructured two-dimensional grid, with a spectral approximation in the third dimension. The method is suitable for the resolution of complex two-dimensional geometries that require the third dimension to capture three-dimensional non-linear unsteady effects, such as those for instance present in linear cascades with separated bubbles. Its main advantage is the reduction in the computational cost, for a given accuracy, with respect standard finite volume methods due to the inexpensive high-order discretization that may be obtained in the third direction using fast Fourier transforms. The method has been applied to the resolution of transitional bubbles in flat plates with adverse pressure gradients and realistic two-dimensional airfoils.

Validation of Turbine Airfoil Leading Edge Pressure and Skin Friction Modeling and Measurement

New pressure and skin friction measurements on a cylinder were made and compared with analytical and computational modeling. The measurements were made in what we call the stagnation region, such as exists on the leading edge of a turbine airfoil. The new skin friction measurements were made with an oil film interferometry technique that has been used in earlier turbine cascade studies, and the accuracy of the technique is demonstrated by comparisons with the analytical and computational modeling. The analytical model is an exact solution for laminar stagnation point or Hiemenz flow, and it is based solely on a reference velocity and a length scale, both which describe the free stream acceleration and pressure gradient in the stagnation region. Comparisons were made between the analytical model and measurements of preasure and skin friction from other cylinder studies in the open literature. These comparisons were used to demonstrate that the analytical model accounts for variations in the free stream pressure gradient and acceleration of the flow due to blockage, separation location, and compressibility. The simplicity of the analytical model allows one to quickly characterize stagnation region aerodynamics based only on a pressure profile, which can be used, for example, to assess the computational grid density in the stagnation region of a turbine airfoil. This basic cylinder experiment along with the analytical modeling are shown to be particularly effective for the validation of skin friction techniques, and this study serves to validate the OFI method used in earlier cascade studies.