Low Engine Order Research Articles

Multibladerow Forced Response Modeling in Axial-Flow Core Compressors

This paper describes the formulation and application of an advanced numerical model for the simulation of blade-passing and low-engine order forced response in turbomachinery core compressors. The Reynolds averaged Navier–Stokes equations are used to represent the flow in a nonlinear time-accurate fashion on unstructured meshes of mixed elements. The structural model is based on a standard finite-element representation. The fluid mesh is moved at each time step according to the structural motion so that changes in blade aerodynamic damping and flow unsteadiness can be accommodated automatically. A whole-annulus 5-bladerow forced response calculation, where three upstream and one downstream bladerows were considered in addition to the rotor bladerow of interest, was undertaken using over 20 million grid points. The results showed not only some potential shortcomings of equivalent 2-bladerow computations for the determination of the main blade-passing forced response, but also revealed the potential importance of low engine-order harmonics. Such harmonics, due to stator blade number differences, or arising from common symmetric sectors, can only be taken into account by including all stator bladerows of interest. The low engine-order excitation that could arise from a blocked passage was investigated next. It was shown that high vibration response could arise in such cases.

Low-Engine-Order Excitation Mechanisms in Axial-Flow Turbomachinery

One particular regime of forced response, occurring at much lower frequencies than the blade-passing frequency, is the so-called low-engine-order (LEO) forced response. The source of LEO excitation is a loss of symmetry in the flow, such as that caused by stator blade throat width variations, flow exit angle variations, perturbations in the passage cooling flow,or temperature distortions. Using a 3D integrated time-domain aeroelasticity code, parametric forced response studies were conducted for a high-pressure turbine stage with 36 stator and 90 rotor blades. Both whole-annulus and sector models were used to investigate the effects of individual and combined LEO parameters. For individual parameters, the amplitude of the excitation was, proportional to the imposed variation. For combined cases, the total excitation was found to be determined by the phasing between the individual excitations. A ballpark comparison suggested that LEO and blade-passing forced response vibration amplitudes were similar for typical variations of the controlling LEO parameters.

Multi-bladerow fan forced response predictions using an integrated three-dimensional time-domain aeroelasticity model

A non-linear integrated aeroelasticity system to predict the forced vibration response of aero-engine fans is presented in this paper. The computational fluid dynamics (CFD) solver, which uses Favre-averaged Navier-Stokes equations on unstructured grids of mixed elements, is coupled to a modal model of the structure so that the effects of blade flexibility can be accommodated. The structural motion due to unsteady fluid forces is computed at every time step and the flow mesh is moved to follow the structure so that the resulting flow unsteadiness is determined in a time-accurate fashion. Two fan forced response case studies are reported in detail. The first one deals with a high-pressure ratio fan, the excitation being due to the upstream variable-angle inlet guide vanes (VIGVs). The unsteady flow analysis with blade motion was conducted using a sector of three VIGVs and four rotor blades. The wake predictions were found to be in good agreement with the corresponding laser measurements. The flow was observed to be completely separated for high VIGV angles and the excitation encompassed several harmonics. The predicted rotor blade vibration levels were generally found to be within 30 per cent of the measured values. The forced response to upstream obstructions was studied in the next fan case study. Three whole bladerows, consisting of 11 struts, 33 VIGVs and 26 rotor blades, were modelled in full. The model also included a prescribed inlet distortion pattern so that the combined effects of stator wakes and inlet distortion on the response of the rotor blades could be studied. The unsteady flow calculations were conducted using a time-accurate non-linear viscous flow representation. Blade motion was also included. Such an undertaking required about 4.2 million grid points to include all three bladerows in a complete stage calculation. To reduce the computational effort, a number of smaller computations were conducted by considering the stator and rotor domains separately: the outflow solution of the stator domain was used as an inflow boundary condition to the rotor domain. The results indicated that such isolated bladerow computations were likely to under-predict the response levels because of neglecting rotor-stator interactions. A number of low engine order (LEO) harmonics were identified from an inspection of the unsteady forcing created by the inlet distortions. Good agreement was obtained for cases where experimental data were available.

Turbine forced response prediction using an integrated non-linear analysis

The forced response due to flow defects caused by the upstream blade rows is predicted for two turbines: intermediate pressure (IP) and low pressure (LP). The prediction method is based on an advanced numerical tool where the compressible viscous flow field is modelled by solving Favre-averaged Navier-Stokes equations with the Baldwin and Barth turbulence model. The flow solution is coupled to a modal model of the structure and information is exchanged every time step between the fluid and the structural domains. The hybrid unstructured mesh is moved at each time step to follow the structural motion using a spring analogy. For the IP turbine, the method was used to rank two different designs of nozzle guide vanes. For the LP turbine, special emphasis was placed on predicting vibration amplitudes due to high and low engine order excitations. Predictions and measurements were found to be in good agreement for both turbines. Due to insufficient experimental data, it was difficult to assess the accuracy of the low engine order computations, although it was shown that the model was capable of undertaking such a task.