AERODYNAMIC PERFORMANCE OF AN AXIAL COMPRESSOR WITH A CASING GROOVE COMBINED WITH INJECTION

Transcription

1 AERODYNAMIC PERFORMANCE OF AN AXIAL COMPRESSOR WITH A CASING GROOVE COMBINED WITH INJECTION Dae-Woong Kim 1, Jin-Hyuk Kim 2 and Kwang-Yong Kim 1 1 Department of Mechanical Engineering, Inha University, Yonghyun-Dong, Nam-Gu, Incheon, Republic of Korea 2 Green Energy System Technology Center, Korea Institute of Industrial Technology, Ipjang-Myeon, Seobuk-Gu, Cheonan-Si, Chungnam, Republic of Korea kdw@inha.edu; jinhyuk.kim@hanmail.net; kykim@inha.ac.kr ICETI 2012-J1163_SCI No. 13-CSME-72, E.I.C. Accession 3530 ABSTRACT Aerodynamic performance of a transonic axial compressor with a casing groove combined with injection has been investigated in this work. Three-dimensional Reynolds-averaged Navier Stokes equations with k-ε turbulence model are discretized by finite volume approximations and solved on hexahedral grids for the flow analyses. For parametric study, the front and rear lengths and height of the casing groove are selected as the geometric parameters and are changed with constant injection to investigate their effects on the stall margin and peak adiabatic efficiency. As a result of the parametric study, the maximum stall margin and peak adiabatic efficiency are found to be obtained in the axial compressor having 70% height of the reference groove. The results show that the application of the casing groove combined with injection to an axial compressor is effective for the simultaneous improvement of both the stall margin and peak adiabatic efficiency of the compressor. Keywords: axial compressor; tip injection; casing groove; stall margin; adiabatic efficiency. LES PERFORMANCES AÉRODYNAMIQUES D UN COMPRESSEUR AXIAL AVEC UNE ENVELOPPE NERVURÉE COMBINÉ À L INJECTION RÉSUMÉ Les performances aérodynamiques d un compresseur transaxial avec une enveloppe nervurée combiné à l injection ont été étudiées dans ce papier. L équation Navier Stokes tridimensionnelle sous la forme moyenne de Reynolds avec le modèle de turbulences k-ε discrétisées par approximation des volumes finis et résolue sur une grille hexahedrique pour l analyse de flux. Pour l étude paramétrique, les longueurs et hauteurs du devant et de l arrière des nervures de l enveloppe ont été sélectionnées comme paramètres géométriques et ont été modifiés avec une injection constante pour analyser leur effet sur la marge de décrochage et l efficacité adiabatique de pic. Le résultat de l étude paramétrique donne une marge de décrochage et une efficacité adiabatique de pic maximum dans le compresseur axial ayant 70% de la hauteur de la nervure de référence. Le résultat montre que l application d une enveloppe nervurée combinée à l injection dans un compresseur axial permet d en augmenter efficacement et simultanément la marge de décrochage et l efficacité adiabatique de pic. Mots-clés : compresseur axial ; injection ; enveloppe nervurée ; marge ; efficacité adiabatique. Transactions of the Canadian Society for Mechanical Engineering, Vol. 37, No. 3,

2 Fig. 1. Geometry of casing groove and location of injection. 1. INTRODUCTION In a compressor, flow phenomena such as stall and surge cause a problem of operational stability and reduce the efficiency [1, 2]. In order to control these phenomena, both passive and active methods have been applied to axial compressors. Many studies have been conducted to prevent surge and stall in axial compressors through experimental and computational methods. Huang et al. [3] suggested that variations of the width and depth of the circumferential casing grooves affect the stall margin and peak adiabatic efficiency of an axial compressor, and also that the stall mechanism is substantially influenced by the blade tip clearance. Houghton and Day [4] studied on the effects of the location of the groove in axial direction on the operating stability and efficiency of an axial compressor. Rabe and Hah [5] investigated the fundamental stall flow mechanism with circumferential grooves installed over the casing of a transonic axial compressor. Cassina et al. [6] studied on the aspect ratio of injector and the injection angle in an axial compressor stage equipped with the discrete tip injections. Suder et al. [7] reported the relationship between the compressor s stability and the axial tip velocity through numerical and experimental studies. Weichert et al. [8] suggested a new design of casing treatment of an axial compressor using injection. Recently, some studies on the stall mechanism of an axial compressor combined with injection and casing groove have been presented [9 11]. Beheshti et al. [9] reported on the performance enhancement of transonic compressors using the blade tip injection coupled with casing treatment. Khaleghi et al. [10] analyzed the effects of various injected angles on the stability enhancement of an axial compressor combined with tip injection and casing groove. Beheshti et al. [11] evaluated the stability margin with various mass flow rates and angles of injection in an axial compressor. In this study, a parametric study on aerodynamic performance of a transonic axial compressor with various groove geometries combined with injection has been performed numerically based on three-dimensional Reynolds-averaged Navier Stokes (RANS) equations. The front and rear lengths and height of the groove are selected as geometric variables and are changed with constant injection. The objectives of this work are to find the effects of groove geometry on the compressor stability and performance and to understand the stall flow mechanism. 2. NUMERICAL DETAILS 2.1. Design Specifications The compressor model investigated in this study is a transonic axial compressor with NASA Rotor 37 [12]. The rotor operates at a speed of 17,188.7 rpm, the total pressure ratio is 2.106, and the adiabatic efficiency is 88.9% at design mass flow rate kg/s. The tip clearance is mm (0.47% span), the choking mass flow is kg/s, and the near stall point is of the choke flow. Flow parameters such as total pressure and total temperature in relation to mass flow rate are measured at the inlet and outlet. 284 Transactions of the Canadian Society for Mechanical Engineering, Vol. 37, No. 3, 2013

3 Fig. 2. Structure of the grid system. Table 1. Dimensions of the reference casing groove [9]. X1 X2 X3 Dimension (mm) Figure 1 shows the casing groove geometry with injection. The annular injection is located at front surface of the casing groove. The dimensions of the reference groove are given in Table 1 [9] Numerical Analysis The computational domain for numerical analysis consists of one passage of the axial compressor rotors including a casing groove. The commercial computational fluid dynamics (CFD) code, ANSYS-CFX 11.0 [13] was used for the flow analysis. Blade profile creation and computational mesh generation were performed using ANSYS Blade-Gen, and Turbo-Grid, respectively. And, CATIA and ICEM-CFD were employed to design the groove and to generate the mesh, respectively. CFX-Pre, CFX-solver, and CFX-Post were employed for defining boundary conditions, solving governing equations, and post processing the results, respectively. A structured grid system was constructed in the computational domain, which has O-type grids near the blade surfaces and H/J/C/L grids in other regions. Through a grid dependency test, the optimum numbers of grids for the passage and groove were found to be 330,000 and 90,000, respectively. Figure 2 shows an example of the structured grid system. Total pressure and total temperature are set to 1.0 atm and K, respectively, as the inlet boundary conditions. The air is considered as an ideal gas and the average static pressure is set at the outlet boundary condition for steady-state simulation. The solid boundaries are considered as adiabatic and hydraulically smooth walls. The periodic boundaries are set at the blade passage interfaces, and tip clearance is modeled along with the passage. The general grid interface (GGI) method is used for the connection between the passage and the groove. Mass flow rate of the injection at front of the casing groove is set to 2.4% of the choking mass flow rate. The k-ε model is employed as a turbulence closure. The numerical stall inception point was identified from the last converged point. The present work employed the convergence criterion suggested by Chen et al. [14] as follows: 1. The inlet mass flow rate variation is less than kg/s for 300 steps. 2. The difference between inlet and outlet mass flow rate is less than 0.5%. 3. At that time the adiabatic efficiency variation is less than 0.03% per 100 steps. Transactions of the Canadian Society for Mechanical Engineering, Vol. 37, No. 3,

4 Fig. 3. Validation of CFD results Performance Parameters To evaluate the performance of the axial compressor, the stall margin and peak adiabatic efficiency are selected as the performance parameters, which are defined as follows: ( mpeak SM = PR ) stall 1 100% (1) m stall PR peak η = ( Pt,out P t,in ) γ 1/γ 1 ( Tt,out T t,in ) 1 (2) where m peak and m stall are mass flow rate at peak efficiency and stall point, respectively. PR peak and PR stall are total pressure ratio at peak efficiency and stall point, respectively. γ, P t, and T t indicate the specific heat ratio, total pressure, and total temperature, respectively. 3. RESULTS AND DISCUSSION To validate the numerical results, the results of the RANS analysis have been compared with experimental data [15] for the total pressure ratio and adiabatic efficiency. As shown in Fig. 3, the numerical results are in good agreements with experimental data through the whole mass flow range. Also, the near-stall point of the experimental data is 0.925, which is the mass flow normalized by the chocking mass flow, while the near-stall point predicted by the numerical simulation is Figure 4 shows results of the parametric study for the stall margin and peak adiabatic efficiency with variations of the front and rear lengths (X2 and X3, respectively) and height of the groove (X1) in the axial compressor combined with casing groove and injection. The values on the horizontal axis indicate the percentages on the basis of the reference values in Table 1. The predicted stall margins for the smooth casing and reference groove are and 15.81%, respectively, and their peak adiabatic efficiency values are and 0.868, respectively. In comparison with the smooth casing, the stall margin and peak adiabatic efficiency for the reference groove are increased by 1.24 and 0.20%, respectively, as shown in Fig. 4. In Figs. 4a and 4b, the rear groove length is varied in the range, % of the reference length with fixed front groove location. The stall margin and peak adiabatic efficiency values with the variation of the rear groove length are higher than those of the smooth casing. In particular, the stall margin has the highest value at 80% rear groove length of the reference groove. However, the peak adiabatic efficiency increases only slightly with decrease of the rear groove length as shown in Fig. 4b. Figures 4c and 4d show the results 286 Transactions of the Canadian Society for Mechanical Engineering, Vol. 37, No. 3, 2013

5 Fig. 4. Results of parametric study on the stall margin and peak adiabatic efficiency: (a, b) stall margin and peak adiabatic efficiency with the rear groove length; (c, d) stall margin and peak adiabatic efficiency with the front groove length; (e, f) stall margin and peak adiabatic efficiency with the groove height. Transactions of the Canadian Society for Mechanical Engineering, Vol. 37, No. 3,

6 Fig. 5. Performance curves: (a) total pressure ratio; (b) adiabatic efficiency. with the variation of the front groove length. The peak adiabatic efficiency discretely increases as the front groove length decreases. The stall margin reaches the maximum value, 20.65% at 80% of the front groove length, which is the value improved by 6.08 and 4.84% in comparison with the smooth casing and reference groove, respectively. And, it is noted that the stall margin is suddenly decreased to 9.29% as the front groove length decreased from 70 to 60%. The results with the variation of the groove height retaining the reference groove lengths are shown in Figs. 4e and 4f. The peak adiabatic efficiency increases steadily with decrease of the groove height. On the other hand, the stall margin at 70% of the reference groove height has the highest value of 20.4%. Figure 5 shows performance curves for the smooth casing, reference groove, and the cases which have the most effective stall margin values in Fig. 4. The mass flow rate is normalized by the chocking mass flow rate which does not include the injected mass flow rate. The predicted near-stall flows for the cases with smooth casing, reference, 80% of the reference rear groove length, 80% of the reference front groove length, and 70% of the reference groove height are 0.909, 0.883, 0.859, 0.863, and 0.859, respectively. These results and the results shown in Fig. 5a indicate that the application of the casing groove combined with injection substantially contributes to the extension of the operating range without losses of the pressure ratio and peak adiabatic efficiency. Figure 6 indicates the Mach number contours at 98% span for the smooth casing and reference groove, the aforementioned cases at the near-stall point of the smooth casing (m/m max = 0.909). In case with the smooth casing, the low speed region is observed on the pressure surface near the leading edge. And, this low energy region coincides with the vortex stagnation region. When the groove and injection are installed over the casing of the compressor, the low speed region is moved from the upstream to downstream of the blade. In Fig. 6d, the low speed region is remarkably suppressed in comparison with other cases, thus the stall inception point happened by shock-vortex interaction is relatively moved to downstream region compared to other cases. Figure 7 shows the static entropy contours at the 98% span for the smooth casing, reference groove, and aforementioned cases at the near-stall point of the smooth casing. As shown in Fig. 7, application of the casing groove combined with injection to the compressor significantly reduces the entropy in the entire region of the blade passage compared to the smooth casing. Consequently, the reduction in the entropy contributes to the recovery of efficiency, as shown in Fig. 5b. Figure 8 shows the trajectories of the tip leakage vortices for the smooth casing, reference groove, and aforementioned cases at the near-stall point of the smooth casing. The tip leakage vortex is produced by 288 Transactions of the Canadian Society for Mechanical Engineering, Vol. 37, No. 3, 2013

7 Fig. 6. Mach number contours at 98% span at near-stall point of the smooth casing case: (a) smooth casing; (b) reference groove; (c) 80% rear groove length; (d) 80% front groove length; (e) 70% groove height. Fig. 7. Static entropy contours at 98% span at near-stall point of the smooth casing case (unit J/kg-K): (a) smooth casing; (b) reference groove; (c) 80% rear groove length; (d) 80% front groove length; (e) 70% groove height. Transactions of the Canadian Society for Mechanical Engineering, Vol. 37, No. 3,

8 Fig. 8. Trajectory of tip leakage vortex at near-stall point of the smooth casing case: (a) smooth casing; (b) reference groove; (c) 80% rear groove length; (d) 80% front groove length; (e) 70% groove height. Fig. 9. Static entropy contours on meridional plane at each peak adiabatic efficiency point (unit: J/kg-K): (a) smooth casing; (b) reference groove; (c) 80% rear groove length; (d) 80% front groove length; (e) 70% groove height. the interaction between tip leakage flow and incoming flow. In the smooth casing, the tip leakage vortex is mainly driven by strong tip leakage flow, and proceeds along the pressure surface downstream of the blade, as shown in Fig. 8a. Meanwhile, the trajectories of the tip leakage vortices for the axial compressors with various casing grooves combined with injection are mainly driven along the suction surface upstream of the blade by the injected mass flow and incoming flow, resulting in the trajectory of the tip leakage vortex close to the passage center, as shown in Figs. 8b 8e. 290 Transactions of the Canadian Society for Mechanical Engineering, Vol. 37, No. 3, 2013

9 Circumferentially averaged static entropy contours on meridional plane at the peak adiabatic efficiency conditions for the smooth casing and the aforementioned cases are shown in Fig. 9. In case with the smooth casing, the high entropy values are widely distributed near the casing, as shown in Fig. 9a. In contrast to the case with smooth casing, the axial compressors combined with the casing groove and injection have mostly lower entropy values, as shown in Figs. 9b e. 4. CONCLUSIONS A parametric study on the aerodynamic performance of a transonic axial compressor combined with a groove and tip injection was performed using three-dimensional RANS equations. The front and rear lengths and height of groove were selected as geometric parameters. The numerical results were validated in comparison with experimental data for the pressure ratio and adiabatic efficiency. With the application of casing groove combined with tip injection to an axial compressor, the low energy region is substantially diminished compared to the case with the smooth casing. Therefore, the casing groove combined with tip injection can be effectively used in an axial compressor for the simultaneous improvement of both the stall margin and peak adiabatic efficiency. The stall margin is similarly sensitive to the three parameters tested, and a maximum stall margin is presented with the variation of each parameter. However, the peak adiabatic efficiency is less sensitive to the parameters, and generally increases with a decrease of each parameter. ACKNOWLEDGEMENT This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korea government (MEST) (No ). REFERENCES 1. Xu, C. and Amano, R.S., Study of the flow in centrifugal compressor, International Journal of Fluid Machinery and Systems, Vol 3, No. 3, pp , Boretti, A.A., Numerical evaluation of the performance of a compression ignition CNG engine for heavy duty trucks with an optimum speed power turbine, International Journal of Engineering and Technology Innovation, Vol. 1, No. 1, pp , Huang, X., Chen, H. and Fu, S., CFD investigation on the circumferential grooves casing treatment of transonic compressor, in Proceedings of ASME Turbo Expo 2008, Berlin, Germany, GT , June 9 13, Houghton, T. and Day, I., Enhancing the stability of subsonic compressor using casing groove, ASME Journal of Turbomachinery, Vol. 133, No 2, Rabe, D.C. and Hah, C., Application of casing circumferential grooves for improved stall margin in a transonic axial compressor, in Proceedings ASME Turbo Expo 2002, Amsterdam, the Netherlands, GT , June 3 6, Cassina, G., Beheshti, B.H., Kammerer, A. and Abhari, R.S., Parametric study of tip injection in axial flow compressor stage, in Proceedings ASME Turbo Expo 2007, Montreal, Canada, GT , May 14 17, Suder, K.L., Hathaway, M.D., Thorp, S.A. and Strazisar, A.J., Compressor stability enhancement using discrete tip injection, ASME Journal of Turbomachinery, Vol No. 1, pp , Weichert, S., Day, I. and Freeman C., Self-regulating casing treatment for axial compressor stability enhancement, in Proceedings ASME Turbo Expo 2011, Vancouver, Canada, GT , June 6 10, Beheshti, B.H., Farhanieh, B., Ghorbanian, K., Teixeira, J.A. and Ivey, P.C., Performance enhancement in transonic axial compressors using blade tip injection coupled with casing treatment, Proceedings of Institution of Mechanical Engineers, Part A Journal of Power and Energy, Vol. 219, No. 5, pp , Khaleghi, H., Texeria, J.A., Tousi, A.M. and Boroomand, M., Parametric study of injection angle effect on stability of axial compressor, AIAA Journal of Propulsion and Power, Vol. 24, No. 5, pp , Beheshti, B.H., Ghorbanian, K., Farhanieh, B., Teixeira, J.A. and Ivey, P.C., A new design for tip injection in Transactions of the Canadian Society for Mechanical Engineering, Vol. 37, No. 3,

10 transonic axial compressor, in Proceedings ASME Turbo Expo 2006, Barcelona, Spain, GT , May 8 11, Reid, L. and Moore, R.D., Design and overall performance of four highly-loaded, high-speed inlet stage for an advanced, high-pressure-ratio core compressor, NASA TP-1337, ANSYS CFX-11.0, ANSYS CFX-Solver Theory Guide, ANSYS Inc., Chen, H., Huang, X. and Fu, S., CFD investigation on stall mechanisms and casing treatment of a transonic compressor, in Proceedings 42nd AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit, Sacramento, USA, AIAA , July 9 12, Dunham. J., CFD validation for propulsion system components, AGARD Advisory Report 355, Transactions of the Canadian Society for Mechanical Engineering, Vol. 37, No. 3, 2013