Analysis of Some Sources of Numerical Uncertainty Applied to a Transonic Compressor Stage

Author(s):  
Cristian Ferrari ◽  
Mirko Morini ◽  
Michele Pinelli ◽  
Pier Ruggero Spina

In recent years, a constantly growing interest in CFD accuracy assessment has been noticed in the open literature. In particular, this interest has grown considerably in the field of technical and industrial applications, since CFD is nowadays routinely used by many engineers, especially in research and development. In this paper, a contribution to this topic is presented. The most common issues about CFD uncertainty are reviewed and commented. Then, some findings originating from the application of three-dimensional numerical calculations to a model of the NASA Stage 37 axial compressor are reported. Particular attention is devoted to multistage turbomachinery modelling uncertainty in terms of rotor/stator interface models and rotor and stator gridding issues.

2008 ◽  
Vol 2008 ◽  
pp. 1-10 ◽  
Author(s):  
M. Hembera ◽  
H.-P. Kau ◽  
E. Johann

This article presents the study of casing treatments on an axial compressor stage for improving stability and enhancing stall margin. So far, many simulations of casing treatments on single rotor or rotor-stator configurations were performed. But as the application of casing treatments in engines will be in a multistage compressor, in this study, the axial slots are applied to a typical transonic first stage of a high-pressure 4.5-stage compressor including an upstream IGV, rotor, and stator. The unsteady simulations are performed with a three-dimensional time accurate Favre-averaged Navier-stokes flow solver. In order to resolve all important flow mechanisms appearing through the use of casing treatments, a computational multiblock grid consisting of approximately 2.4 million nodes was used for the simulations. The configurations include axial slots in 4 different variations with an axial extension ranging into the blade passage of the IGV. Their shape is semicircular with no inclination in circumferential direction. The simulations proved the effectiveness of casing treatments with an upstream stator. However, the results also showed that the slots have to be carefully positioned relative to the stator location.


Author(s):  
Andreas Lesser ◽  
Jens Iseler ◽  
Reinhard Niehuis

This paper deals with the numerical flow-simulation of a transonic compressor stage, which has been investigated for baseline as well as distorted inflow conditions at the Institute of Propulsion Technology of the DLR in Cologne (Dunker [1] and Lecht [2]). The inlet distortions are generated in the experiment upstream of the compressor stage by non-rotating steel bars, while in the numerical calculations the total pressure and inflow angle distribution measured downstream of the bars are taken as inflow boundary conditions. The circumferential extent of the generated total pressure and inflow angle distortion is 120 degrees. Numerical simulations were performed for uniform inflow conditions at 85% and 100% rotational speed. For disturbed inflow conditions, a full-annulus calculation has been carried out for an operational point at peak efficiency. The object of the investigations is to validate the flow solver for compressor flow with distorted inflow. The results from time-averaged numerical and experimental data are compared extensively. The experimental trends are qualitativly and in the most part also quantitativly well reproduced in the numerical calculations.


1992 ◽  
Vol 114 (3) ◽  
pp. 510-516 ◽  
Author(s):  
J. Paulon ◽  
Zhifang Zhang ◽  
Pingfang Jia ◽  
Jingfei Meng

Interaction phenomena between rotor and stator are unavoidable in advanced compressors and their effects increase with the performance of the turbomachines. Until now, it was not possible to quantify the interaction effects, but with the development of three-dimensional unsteady computation codes in a complete stage, it is possible to know, in detail, the flow field through the machine and to make evident and to explain the difficulties encountered in measuring the flow parameters. A study has been conducted in this way at ONERA, on an axial transonic compressor stage. The computations have been made with a simulation of the losses; in this manner, the overall computed and measured performances of the compressor are the same. A detailed analysis of the unsteady computation results makes evident, between rotor and stator, large variations of some parameters of the flow as a function of time, but also as a function of the axial and tangential relative position of steady probes and stator blades. Unsteady measurements made on another transonic machine confirm the indications given by these computations.


Author(s):  
Marcus Lejon ◽  
Niklas Andersson ◽  
Tomas Grönstedt ◽  
Lars Ellbrant ◽  
Hans Mårtensson

Surface degradation in an axial compressor during its lifetime can have a considerable adverse effect on its performance. The present study investigates how the optimized design of compressor blades in a single compressor stage is affected by considering a high level of surface roughness on a level representative of a long period of in-service use. It is shown that including surface roughness in the optimization process is of relatively little importance, however, matching of compressor stages is shown to require consideration as the rotational speed must be increased to reach the design point as surface quality decrease. An increased surface roughness in itself is shown to have a large effect on performance. Two optimization approaches are compared. The first approach considers the compressor blades to be hydraulically smooth. The designs obtained from this approach are subsequently degraded by increasing the level of surface roughness. The compressor blades from the first approach are compared to designs obtained from a second optimization approach, which considers a high level of surface roughness from the outset. The degraded compressor stages from the first approach are shown to be among the best performing designs in terms of polytropic efficiency and stability when compared to designs obtained with the second approach.


Author(s):  
Yiming Zhong ◽  
WuLi Chu ◽  
HaoGuang Zhang

Abstract Compared to the traditional casing treatment, the self-recirculating casing treatment (SCT) can improve or not decrease the compressor efficiency while achieving the stall margin improvement. For the bleed port, the main design indicator is to reduce the flow loss caused by suction, while providing sufficient jet flow and jet pressure to the injector. In order to gain a better study of the bleed port stabilization mechanisms, the bleed configuration was parameterized with the bleed port inlet width and the bleed port axial position. Five kinds of recirculating casing treatments were applied to a 1.5-stage transonic axial compressor with the method of three-dimensional unsteady numerical simulation. Fifteen identical self-recirculating devices are uniformly mounted around the annulus. The numerical results show that the SCT can improve compressor total pressure ratio and stability, shift the stall margin towards lower mass flows. Furthermore, it has no impact on compressor efficiency. The optimal case presents that stability margin is improved by 6.7% employing 3.1% of the annulus mass flow. Expanding bleed port inlet width to an intermediate level can further enhance compressor stability, but excessive bleed port inlet width will reduce the stabilization effect. The optimal bleed port position is located in the blocked area of the low energy group at the top of the rotor. In the case of solid casing, stall inception was the tip blockage, which was mainly triggered by the interaction of the tip leakage vortex and passage shock. From radial distribution, the casing treatment predominantly affects the above 70% span. The reduction of tip reflux region by suction effect is the main reason for the extension of stable operation range. The SCT also has an obvious stability improvement in tip blockage stall, while delaying the occurrence of compressor stall.


1987 ◽  
Vol 109 (3) ◽  
pp. 388-397 ◽  
Author(s):  
A. J. Wennerstrom

Between 1970 and 1974, ten variants of a supersonic axial compressor stage were designed and tested. These included two rotor configurations, three rotor tip clearances, addition of boundary-layer control consisting of vortex generators on both the outer casing and the rotor, and the introduction of slots in the stator vanes. Design performance objectives were a stage total pressure ratio of 3.0 with an isentropic efficiency of 0.82 at a tip speed of 1600 ft/s (488 m/s). The first configuration passed only 70 percent of design flow at design speed, achieving a stage pressure ratio of 2.25 at a peak stage isentropic efficiency of 0.61. The rotor was grossly separated. The tenth variant passed 91.4 percent of design flow at design speed, producing a stage pressure ratio of 3.03 with an isentropic efficiency of 0.75. The rotor achieved a pressure ratio of 3.59 at an efficiency of 0.87 under the same conditions. Major conclusions were that design tools available today would undoubtedly permit the original goals to be met or exceeded. However, the application for such a design is currently questionable because efficiency goals considered acceptable for most current programs have risen considerably from the level considered acceptable at the inception of this effort. Splitter vanes placed in the rotor permitted very high diffusion levels to be achieved without stalling. However, viscous effects causing three-dimensional flows violating the assumption of flow confined to concentric stream tubes were so strong that a geometry optimization does not appear practical without a three-dimensional, viscous analysis. Passive boundary-layer control in the form of vortex generators and slots does appear to offer some benefit under certain circumstances.


Author(s):  
Wu Xiaoxiong ◽  
Bo Liu ◽  
Shi Lei ◽  
Zhang Guochen ◽  
Mao Xiaochen

In this paper, an improved streamline curvature (SLC) approach is presented to obtain the internal flow fields and evaluate the performance of transonic axial compressors. The approach includes some semi-empirical correlations established based on previous literatures, such as minimum loss incidence angle model, deviation model and total pressure loss model. Several developments have been made in this paper for the purpose of considering the influences of three-dimensional (3D) flow in high-loaded multistage compressors with high accuracy. A revised deviation model is applied to predict the cascade with large deflection range. The method for predicting the shock loss is also discussed in detail. In order to validate the reliability of the approach, two test cases including a two-stage transonic fan and a three-stage transonic compressor are conducted. The overall performance and distribution of spanwise aerodynamic parameters are illustrated in this paper. Compared with both the experimental and computational fluid dynamic (CFD) data at design and a number of different off-design condition, the SLC results give reasonable characteristic curves. The validation demonstrates that this improved approach can serve as a fast and reliable tool for flow field analysis and performance prediction in preliminary design stage of axial compressors.


1991 ◽  
Author(s):  
Jacques Paulon ◽  
Zhifang Zhang ◽  
Pingfang Jia ◽  
Jingfei Meng

Interaction phenomena between rotor and stator are unavoidable in advanced compressors and their effects increase with the performances of the turbomachines. Until now, it was not possible to quantify the interaction effects, but with the development of 3-D unsteady computation codes in a complete stage, it is possible to know, in detail, the flow field through the machine and to make evident and to explain the difficulties encountered in measuring the flow parameters. A study has been conducted in this way at ONERA, on an axial transonic compressor stage. The computations have been made with a simulation of the losses; in this manner, the overall computed and measured performances of the compressor are the same. A detailed analysis of the unsteady computation results makes evident, between rotor and stator, large variations of some parameters of the flow as a function of time but also as a function of the axial and tangential relative position of steady probes and stator blades. Unsteady measurements made on another transonic machine confirm the indications given by these computations.


Author(s):  
Vaclav Cyrus

A detailed investigation of three-dimensional flow was carried out in a low speed rear axial compressor stage with the change of the stator blade row setting. The stator blade stagger change was in the range of (−14) – (23) degree. Measurements were performed by means of both stationary and rotating pressure probes at seven working points. The origin of large regions of separated flow in blade rows at positive incidence angles was analysed with the use of the spanwise diffusion factor distribution. These areas in the rotor and stator rows originated as the diffusion factor exceeded the critial value D = 0.6 within (1/4 – 1/3) of the blade height near one end-wall. The rotating stall in compressor stage arised when large regions of separated flow occured simultaneously in both rotor and stator blade rows.


Author(s):  
Ali A. Merchant ◽  
Mark Drela ◽  
Jack L. Kerrebrock ◽  
John J. Adamczyk ◽  
Mark Celestina

The pressure ratio of axial compressor stages can be significantly increased by controlling the development of blade and endwall boundary layers in regions of adverse pressure gradient by means of boundary layer suction. This concept is validated and demonstrated through the design and analysis of a unique aspirated compressor stage which achieves a total pressure ratio of 3.5 at a tip speed of 1500 ft/s. The aspirated stage was designed using an axisymmetric through-flow code coupled with a quasi three-dimensional cascade plane code with inverse design capability. Validation of the completed design was carried out with three-dimensional Navier-Stokes calculations. Spanwise slots were used on the rotor and stator suction surfaces to bleed the boundary layer with a total suction requirement of 4% of the inlet mass flow. Additional bleed of 3% was also required on the hub and shroud near shock impingement locations. A three-dimensional viscous evaluation of the design showed good agreement with the quasi three-dimensional design intent, except in the endwall regions. The three-dimensional viscous analysis predicted a mass averaged total pressure ratio of 3.7 at an isentropic efficiency of 93% for the rotor, and a mass averaged total pressure ratio of 3.4 at an isentropic efficiency of 86% for the stage.


Sign in / Sign up

Export Citation Format

Share Document