PERFORMANCE STUDY OF A 1 MW GAS TURBINE USING VARIABLE GEOMETRY COMPRESSOR AND TURBINE BLADE COOLING Cleverson Bringhenti (+55-12-3947 6951, cleverson@ita.br) Jesuino Takachi Tomita (+55-12-3947 6951, jtakachi@ita.br) João Roberto Barbosa (+55-12-3947 6952, barbosa@ita.br) ABSTRACT This work presents the performance study of a 1 MW gas turbine including the effects of blade cooling and compressor variable geometry. The axial flow compressor, with Variable Inlet Guide Vane (VIGV), was designed for this application and its performance maps synthesized using own high technological contents computer programs. The performance study was performed using a specially developed computer program, which is able to numerically simulate gas turbine engines performance with high confidence, in all possible operating conditions. The effects of turbine blades cooling were calculated for different turbine inlet temperatures (TIT) and the influence of the amount of compressor-bled cooling air was studied, aiming at efficiency maximization, for a specified blade life and cooling technology. Details of compressor maps generation, cycle analysis and blade cooling are discussed. Keywords: Gas Turbine Performance, Axial Flow Compressor, Turbine Cooling, Compressible Flows. INTRODUCTION Gas turbines are optimized at design point (DP). Their performance is poor at part-load. Due to the physics of the flow in its internal passages, choking and surge may cause performance deterioration. It is important therefore, that at an early design phase, the engine behavior be predicted in order to take actions to improve performance. One technique to improve overall performance is to improve the engine components efficiency at part-load, mainly compressor and turbine. This may be accomplished by better components matching. When the engine speed is reduced, the flow in the rear stages of an axial compressor is accelerated due to the density decrease. This may cause choking of the blade passages. Surge may occur at the front stages as consequence. Variable inlet guide vanes (VIGV) installed in front of the first rotor row have been effectively used as means of improving the compressor part-load operation. Sometimes, only the VIGV is sufficient to guarantee a good performance at low speeds. High performance compressors usually have VIGV and variables stators vanes at the front stages. VIGV moves upwards the surge line, increasing the stability area while the effect of bleed before the last stator is to lower the operation line at a given mass flow and to move upwards the stability line. To change the operation line other resources can also be used: variable geometry at the turbine stator blades (industrial gas turbines) and variable nozzle in the engine exit (aero gas turbines). It can be inferred then that a good matching between compressor and turbine is necessary to evaluate the gas turbine behavior. To increase the engine efficiency requires increase in TIT. Special materials and manufacturing processes are also 1
required. Turbine blades must be cooled to withstand mechanical and thermal loads and endure a prescribed life. NGVs and rotor blades may require cooling. Different types of cooling are available using air bled from the compressor. In this work was simulated a gas turbine, running at steady state, with a 5-stage axial flow compressor, VIGV and turbine blade cooling. Blade life cycles of 10,000 and 20,000 hours and two cooling techniques for the NGV and rotor blades were considered. The results are presented and discussed in detail. The axial flow compressor design and performance characteristics, as well as the gas turbine performance calculation were carried out using computer codes developed by the authors. NOMENCLATURE AFCC Axial Flow Compressor Code COD Constant Outer Diameter DP Design-Point L.C. Life Cycle m Mass Flow NGVs Nozzle Guide Vanes ODP Off-Design Point Pr Pressure Ratio Pt Total Pressure S.P Shaft Power TIT Turbine Inlet Temperature Tt exh Exhaust Gas Turbine Stagnation Temperature Tt Total Temperature VIGV Variable Inlet Guide Vanes VG Variable Geometry η TGG Gas Generator Turbine Isentropic Efficiency η th Cycle Efficiency sfc Specific Fuel Consumption DESIGN AND PERFORMANCE CALCULATIONS OF A 5-STAGE AXIAL-FLOW COMPRESSOR WITH VARIABLE GEOMETRY The axial flow compressor is used in aero and industrial gas turbines due to its high efficiency, high mass flow per unit frontal area, high pressure ratio, simple mechanical design and high reliability. It is recognized that, the aerodynamic and thermodynamic design is not simple, mainly to guarantee good off-design operation conditions and adequate stall-to-choke mass flow range. In this work, the methodology applied considers the streamline at blade mid height. The design and performance analyses of the axial flow compressor are calculated as indicated in [1, 2, 3]. The compressor map was generated to incorporate variable geometry to improve part-load compressor efficiency. Details of the compressor design that would lead to high performance and low frontal area can be seen in [1, 4, 5, 6]. This computer program has been developed and validated using data and published data [1, 2]. The axial flow compressor design parameters are: Inlet stagnation pressure: 101,325 Pa; Inlet stagnation temperature: 288.15 K; Mass flow: 8.1 kg/s; Inlet Mach number: 0.50; Outlet Mach number: 0.26; Pressure ratio: 5.0; Polytropic efficiency: 89 %; Number of stages: 5; Rotational speed: 25,650 rpm; Inlet hub-tip-ratio: 0.40; Axial channel: COD. The designed compressor has 5 stages, fixed stators and VIGV was incorporated to improve the compressor ODP operation. The VIGV schedules for each compressor rotational speed are presented in the Table 1. Table 1. VIGV s angles for each rotational speed % of rpm from DP VIGV Angle ( º) 100 0 95 10 90 20 85 25 82 25 80 30 75 30 72 30 70 30 65 30 60 30 55 30 The compressor maps with and without VIGV and its efficiencies are presented in Figure 1 and Figure 2. pressure ratio 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 map without VIGV map with VIGV 0.0 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 corrected mass flow Figure 1 - Pressure ratio versus corrected mass flow 2
isentropic efficiency 0.95 0.90 0.85 0.80 0.75 0.70 0.65 0.60 0.55 0.50 0.45 0.40 0.35 map without VIGV map with VIGV 0.30 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 corrected mass flow Figure 2 - Efficiency versus corrected mass flow Figure 3b - Gas generator view GAS TURBINE CONFIGURATION AND PERFORMANCE CALCULATIONS The study was carried out using a computer program specially developed that numerically simulate all possible gas turbine performance configurations, based on engine functional blocks build-up. For this study a single-shaft free power turbine gas generator was chosen, as sketched in Figure 3a. Figures 3b, 3c, 3d and 3e show details of gas generator (design and test). Engine configuration and performance parameters at DP were chosen as described by Bringhenti and Barbosa [7, 8, 9] and shown in Table 2. This engine has been designed by the Gas Turbine Group and is being built in association with TGM Turbinas, a local industry. Presently, the gas generator (GG) is undergoing vibrational and lubrification tests and it is expected that it will run at full load by the end of the year. A propelling nozzle is used in place of the power turbine for the sake of simplicity during the gas generator tests. The power section will be added after the GG tests are completed. This study aims at future project improvement having in mind distributed power generation. Figure 3c - Gas generator cutaway Figure 3d Gas Generator vibration and lubrication test stand mass flow 12 12 load Figure 3e Gas Generator vibration and lubrication data acquisition display 1 2 3 4 5 6 7 8 9 10 11 ambient intake compressor bleed combustion chamber turbine mixer power turbine Figure 3a - Sketch of a single shaft free power turbine duct exhaust pipe The main engine functional blocks used for this analysis were: ambient, intake, compressor, combustion chamber, turbine, power turbine, and exhaust. The compressor spool turns at N1 rpm at design. At part-load N1 is lowered from 100% down to the point at which the surge margin vanishes. The power turbine is coupled to the generator by a gearbox, so that speed is fixed at 60 Hz or 3600 rpm. For each chosen offdesign point all thermodynamic parameters were calculated, 3
from which a selection of appropriate data was taken to produce the graphs shown. Compressed air to cool the turbine is bled from the compressor exit (station 4). In this study, the cooling air returns to the gas stream after the turbine rotor (station 7), therefore not participating in the rotor expansion process. The main performance parameters that impose restrictions on engine operations, considered in this work, were: maximum cycle temperature or combustion chamber outlet temperature (due to material limits), surge margin (safe operation or surgefree operation) and compressor speed (mechanical integrity due to the stresses caused by rotational speed). The data on Table 2 describe chosen parameters at DP without bleed and maximum cycle temperature 1173 K. Table 2. Reference engine design point characteristics parameters values mass flow (kg/s) 8.10 compressor pressure ratio 5.0 maximum cycle temperature (K) 1173.0 power output (MW) 1.388 25 compressor isentropic efficiency 0.863 combustor chamber pressure loss 0.05 combustion chamber efficiency 0.99 gas generator turbine isentropic efficiency 0.87 mechanical efficiency - gas generator shaft 0.99 exhaust gas temperature (K) 857 isentropic efficiency of free power turbine 0.87 mechanical efficiency - free power turbine shaft 0.99 In this study surge margin at design point was set to 15%. Walsh and Fletcher [10] suggest that required surge margin for power generation application is 15-20%, and is dependent upon acceleration and deceleration times required, engine configuration, whether centrifugal or axial compressors are applied, whether bleed valves or variable stator vanes are employed at part-load, etc. Design speed control acts on blowoff valves to avoid surge at low rotational speeds. Action of blow-off valves are not considered in this study. Part-load conditions were obtained varying N1 from 100% (relative to design) down to the point at which the surge margin vanishes, maintaining the power turbine speed constant. Figure 4 and Figure 5 show the simulation results at several off-design conditions, varying gas generator speed N1. In this study the main cycle parameters were analyzed, such as: stagnation temperatures, pressure ratio, surge margin, air mass flow, power output, compressor isentropic efficiency and cycle efficiency. Figure 4 shows the designed compressor map, pressure ratio versus corrected mass flow for each corrected speed and surge line. Corrected mass flow is calculated by: Pt m Pr ef Tref Tt. Simulation was carried out at off-design and results obtained in all possible operation conditions are shown in Figure 4, where it is possible to see the running line, in red. As surge margin was set as 15% at DP, simulation results indicate that engine corrected speed can be lowered to 82% before it vanishes. The surge margin definition used in this work is defined by equation 1. pressure ratio PR _ DP 1 sm (1) PR _ stall 1 6.0 compressor map PR_stall 5.5 running line PR_DP 5.0 4.5 surge line 4.0 3.5 3.0 2.5 2.0 1.5 1.0 65% 70% 75% 80% 85% 90% 95% 100% 0.5 72% 82% 0.0 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 corrected mass flow Figure 4 Compressor characteristics pressure ratio Figure 5 shows the isentropic efficiency of the developed compressor versus correct mass flow for each corrected speed. The isentropic efficiency at DP was set as 86.3%. Due to the fact that the designed compressor was developed using variable IGV, the isentropic efficiency for each constant corrected speed can be kept almost constant at part-load. This characteristic could not be observed without IGV, where the isentropic efficiency for each constant corrected speed generally falls significantly, indicating low part-load performance. For this particular application and engine configuration the isentropic efficiency of compressor remained practically unaffected, as shown by running line in Figure 4, indicating that the use of variable IGV was very important in improving cycle effiency. 4
isentropic efficiency 0.95 0.90 0.85 0.80 0.75 0.70 0.65 0.60 0.55 0.50 0.45 0.40 compressor map running line 65% 70% 75% 80% 82% 72% 0.35 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 85% corrected mass flow 90% DP 95% 100% Figure 5 Compressor characteristics - efficiency THE STUDY OF THE INFLUENCE OF TURBINE BLADE COOLING ON THE ENGINE EFFICIENCY For each turbine blade geometry there is an appropriate cooling system (convection, convection with coating, film with convection, full coverage film and others). Cooling has influence on the turbine efficiency, therefore on the overall machine characteristics. The problem is to estimate the drop of uncooled turbine efficiency when a cooling system is required. A modular computational routine was implemented in the engine simulation program. The cooling characteristics are calculated based on the work [11]. Detailed information of the mathematical formulation can be found in [12]. These models are compatible with the characteristics of the engine under study and fabrication capability. Cooling technology was selected based on the technology of year 2000. The input data are: Total temperature at compressor discharge = 502 K Compressor inlet mass flow = 8.1 kg/s Turbine total inlet temperature = 1350 K; 1375 K; 1400 K; 1425 K; 1450 K Fuel mass flow = 75 kg/s Turbine expansion ratio = 2.26 Number of turbine stages = 1 Uncooled turbine efficiency = 87 % Full coverage film in the NGV and film with convection for the rotor are shown in Figure 6 and Figure 7 that indicate the aspects of the NGV and rotor blades, showing the cooling holes. Figure 6. Film with convection cooling system. Figure 7. Full coverage film cooling system. The allowable metal temperature depends on the year of material technology, desired life, mission, atmospheric conditions and power settings [7]. Figure 8 shows the allowable metal temperature based on the assumption of historical increase of 5.56 K/year from 1985 material technology for the life cycle equal to 10,000 h and 20,000 h. Another assumption is that the compressor air bleed flow is extracted from the compressor last stage. temperature (K) 1380 1360 1340 1320 1300 1280 1260 1240 1220 NGV; 10,000h rotor; 10,000h NGV; 20,000h rotor; 20,000h 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 year of material technology Figure 8. Allowable metal temperature of NGV and rotor blades The calculated required compressor air bleed and the drop of the turbine efficiency are presented in Figure 9. From Figure 9 it is possible to observe that a higher air bleed mass flow is necessary to guarantee the life cycle and that the drop of turbine efficiency is highest for 20,000 h due to the increase of the turbine blade profile losses. 5
compressor bleed mass flow (kg/s) 0,28 0,26 0,24 0,22 0,20 0,18 bleed;10,000h bleed; 20,000h 0,16 0,7 eff drop;10,000h 0,14 eff drop;20,000h 0,6 1340 1360 1380 1400 1420 1440 1460 TIT (K) Figure 9. Compressor bleed mass flow and turbine efficiency drop vs TIT GAS TURBINE PERFORMANCE RESULTS CONSIDERING TURBINE BLADE COOLING In order to consider the effects of turbine blade cooling and compare cycle performance at all possible operating points the gas generator tubine and bleed parameters at design point were altered but maintaining fixed all other parameters. Therefore, turbine inlet temperature, TIT, gas generator turbine isentropic efficiency, η TGG, and percentage of bleed were altered. Table 3 shows five of the cases studied. Case 1 is the same shown in Table 2. In and 3 TIT, η TGG and bleed for life cycle equal to 10,000 h were modified. In and alterations occured at TIT, η TGG and bleed for life cycle equal to 20,000 h. For all cases, air mass flow, pressure ratio and surge margin are: 8.1 kg/s, 5:1 and 15% respectively.. Table 3 shows values of some engine parameters for 10,000 and 20,000 h life cycles and temperatures of 1375 K and 1425 K. For cases 2 to 5 the cooling system used was full coverage film in the rotor and film with convection in the stator. Table 3 Parameters settings for the studied cases TIT (K) 1173 1300 1425 1300 1425 η TGG (%) 87 86.3 85.8 86.2 85.7 bleed (%) 0 1.9 3.2 2.1 3.4 L.C. (h) 10,000 10,000 20,000 20,000 Table 4 shows results obtained at design point for the five different cases studied. This study was carried out to identify directions for future product improvement aiming cooled blades for cycle lives of 10,000 and 20,000 h. Table 4 shows that turbine blade cooling improves 1% in when comparing cases 1, 3 and 5. Case 3 and show similar results when the comparison is made in cycle effiency and power output, but is better because its life cycle is 20,000 h while in is 10,000 h. 1,3 1,2 1,1 1,0 0,9 0,8 Turbine efficiency drop (%) Table 4 Comparison between different studied cases η th (%) 21.5 22.0 22.4 21.9 22.4 S.P (MW) 1.3877 1.6714 1.9618 1.6646 1.9544 sfc (kg/s/mw) 0.1075 0.1052 0.1034 0.1054 0.1036 fuel flow (kg/s) 93 59 28 55 24 Tt exh 850 946 1034 945 1033 Figure 10 shows part-load simulation results for the five cases considered. At left hand side it can be seen the values for (continuous lines), while at the right hand side the values are for power output (dashed lines). From all the cases studied the, at all possible operating points, are higher for cases 3 and 5, for a given corrected speed. This is because power output absorved by compressor is constant, due to the fact that mass flow and pressure ratio are equal for all cases studied, at the same corrected speed. The higher the maximum temperature higher the fuel flow as indicated in Figure 11 and Figure 12. 0.6 0.4 0.2 0.80 0.82 0.84 0.86 0.88 0.90 0.92 0.94 0.96 0.98 1.00 1.02 corrected speed Figure 10. Cycle efficiency and power output versus corrected speed 1100 1050 1000 0.80 0.82 0.84 0.86 0.88 0.90 0.92 0.94 0.96 0.98 1.00 1.02 corrected speed Figure 11. Cycle efficiency and maximum cycle temperature versus corrected speed 2.0 1.8 1.6 1.4 1.2 1.0 0.8 1450 1400 1350 1300 1250 1200 1150 power output (MW) maximum cycle temperature (K) 6
Figure 11 shows part-load simulation results for five cases considered. At left hand side one can read the values for cycle efficiency (continuous lines) while at the right hand side one can read the values for maximum cycle temperature (dashed lines). Figure 11 shows that, for a given corrected speed maximum is obtained at maximum cycle temperature. 5 0 5 0 5 0 0.135 0.120 0.105 0.090 0.13 0.075 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 power output (MW) Figure 12. Cycle efficiency and fuel flow versus corrected speed Figure 13 shows that for a given power output of less than 1.4 MW, better is achieved for (that simple cycle), without bleed and lower maximum cycle temperature. Therefore, if power output is limited to 1.4 MW, higher maximum cycle temperature will not result in better, for all the five cases. 1500 1450 1400 1350 1300 1250 1200 1150 1100 1050 0.13 1000 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 power output (MW) Figure 13. Cycle efficiency and maximum cycle temperature versus power output Figure 14 shows that for a given power output of less than 1.4 MW, better and surge margin is achieved for. From analyses above for a given power output, less than 1.4 MW, the simple cycle and the respective parameters,, is the best option. fuel flow (kg/s) maximum cycle temperature (K) 0.06 0.04 0.02 0.13 0.00 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 power output (MW) Figure 14. Cycle efficiency and surge margin versus power output From Figure 14 it is seen that surge margin will limit the part-load operation for all cases, so that stators variable geometry would be required in addition to VIGV, not included in this study. CONCLUSIONS Indigenous validated high fidelity computer programs were developed for the design and performance analysis of gas turbines. Application to the study of a single shaft free power turbine has been carried out in order to investigate possible performance improvements if higher temperatures, and therefore blade cooling, were used. From Figure 10 it is seen that cases 3 and 5 are the best when and power are concerned. Figure 13 shows that, for the same level, increase in TIT results in increased power output, what is a significant achievement if plant uprate is considered. ACKNOWLEDGMENTS This research was carried out at the Center for Reference on Gas Turbine, ITA, with support of TGM Turbinas, the partner company that is manufacturing the engine. REFERENCES [1] Tomita, J.T., Numerical Simulation of Axial Flow Compressors. MSc Thesis, ITA, Brazil, April 2003. [2] Tomita, J.T. and Barbosa, J.R., A Model for Numerical Simulation of Variable Stator Axial Flow Compressors. 17th COBEM 2003 International Congress of Mechanical Engineering, São Paulo, Brazil, ID-0239. [3] Barbosa, J.R., A Streamline Curvature Computer Programme for Performance Prediction of Axial Compressors. Ph.D. Thesis, Cranfield Institute of Technology, England 1987. [4] Denton, J. D., Loss Mechanisms in Turbomachines. IGTI 1993, Journal of Turbomachinery vol. 115/621. 0.12 0.10 0.08 surge margin 7
[5] Miller, D.C. and Wasdell D. L., Off-Design Prediction of Compressor Blade Losses. Rolls-Royce, Bristol C279/87. March 1987. [6] Tan, C. S., Three-Dimensional Flow and Tip Clearance Flow in Axial Compressors. VKI Lecture Series, May 18, 2006. [7] Bringhenti, C., Barbosa J. R. Performance Study of A 1 MW Gas Turbine. Proceedings of the 10th Brazilian Congress of Thermal Sciences and Engineering, ENCIT, Brazil Soc. of Mechanical Sciences and Engineering -- ABCM, Rio de Janeiro, Brazil, Nov. 29 -- Dec. 03, Paper CIT04-0089, 2004. [8] Bringhenti, C., Barbosa J. R. In Portuguese: Study of a Gas Generator for Turboshaft and Turbojet. Proceedings CONEM, Brazil, CON04-41042, 2004. [9] Bringhenti, C. 2003, Variable Geometry Gas Turbine Performance Analysis, Ph. D. Thesis, ITA, Brazil. [10] Walsh, P.P., Fletcher, P., Gas Turbine Performance. Blackwell Science, 1998. [11] Plencner, R. M. and Sayder, C. A., The NAVY/NASA Engine Program (NNEP89)-A User s Manual, NASA/TM- 105186, August 1991. [12] Gauntner, J. W., Algorithm for Calculating Turbine cooling flow and the Resulting Decrease in Turbine Efficiency, NASA/TM-81453, February 1980. 8
9