2020 Vol. 35, No. 6

Display Method:
Topology optimization of gas turbine engine fan disk
LIU Xiaogang, DAI Sicheng, WU Zhenhao
2020, 35(6): 1121-1130. doi: 10.13224/j.cnki.jasp.2020.06.001
Abstract:
Based on the mathematical model of continuum structure topology optimization, a 1/18 cyclic symmetric simplified fan structure model was established based on a gas turbine engine design. The topological optimization design of the structure based on solid isotropic material with penalization(SIMP) interpolation variable density method was proposed. In order to obtain a better topology optimization result of the fan disk structure, a method of combining the expanded optimization solution domain and the blisk structure was proposed, and with this method, three control groups were set correspondingly based on the design of the original fan disk. All simulation results were appropriately modified according to the actual structure and manufacturing process, and the dangerous working conditions were verified. Under the premise of meeting the performance requirements, the compressor quality was reduced by 349%.
Surge detection method based on rate of change of compressor discharge static pressure
WANG Yudong
2020, 35(6): 1131-1139. doi: 10.13224/j.cnki.jasp.2020.06.002
Abstract:
The parameters of the compressor’s discharge static pressure and rotor speed were used to establish the surge detection method based on the first-order and the second-order rate of change. According to dozens of surge detection verifications in core-engine and whole-engine tests, when the threshold was exceeded, the stall or surge could be detected by the first-order change rate of compressor discharge static pressure, and then the possibility of occurrence was confirmed by the second-order change rate and the follow-up exceeding value of the first-order change rate of static pressure or speed. Those tests proved that this method had short response time, and the detection time was less than half of the surge period. The successful detection rate was as high as 100%, and the false alarm rate was as low as 0. and. With this method, no false or missed detection had been found.
Effect of non-uniform growth of TGO on stress development and failure mechanism of thermal barrier coatings
LIU Yang, QUAN Chanfbiao, YANG Xiaoguang
2020, 35(6): 1140-1148. doi: 10.13224/j.cnki.jasp.2020.06.003
Abstract:
The cohesive force model and the thermal growth oxide (TGO) non-uniform growth subroutine were used to numerically simulate the stress development and cracking behavior of the thermal barrier coating (TBC) under thermal cycling loading. The cracking process was firstly caused by the mixed Ⅰ and Ⅱ cracks due to the tensile and shear stress in the near peak position in the top coat (TC); it turned to the mode Ⅱ crack due to the shear stress near the peak of the TC and the mode Ⅰ crack due to the tensile stress in the thickness direction in the middle of the peak and trough with the increasing number of thermal cycles. The maximum tensile stress in the overall non-uniform growth and the valley uniform growth mode almost did not increase with the number of cycles after some cycles; while in the peak uniform growth and the overall uniform growth mode, the maximum tensile stress increased continuously. In the global non-uniform growth mode and the valley non-uniform growth mode, the maximum shear stress of -16241 MPa and -15428 MPa appeared at the position near the peak after 20 cycles; while in the global uniform wave growth and global uniform growth mode, the maximum shear stress were -11382 MPa and -11198 MPa, respectively. For the uniform growth of the valley and the overall non-uniform growth mode, interface cracks appeared after 9 cycles; while for the uniform peak growth and overall non-uniform growth mode, interface cracks appeared in the 17th cycle.
Measurement method for axial force of aero-engine special-shaped thrust bearing
OUYANG Yunfang
2020, 35(6): 1149-1156. doi: 10.13224/j.cnki.jasp.2020.06.004
Abstract:
Thrust bearings of an aero-engine integrate outer ring of bearing and squirrel-cage as one part. For this special-shaped bearing, a new method for thrust bearing structure modification design and measurement method of bidirectional axial force was given. The modification principle of axial force measurement, equal bending moment structural design idea, structural mechanical model, strength calculation contrast analysis, strain gauge and full-bridge leading wire adhesion were introduced, and test result analysis of force-measuring ring was made. Results proved that the error between theoretical calculations and calibration test was less than 5%. In the field test run, bidirectional force-measuring rings could accurately identify the turning point for direction change of axial force simultaneously. This means the proposed axial force measurement method based on force-measuring ring has high value of engineering practicability.
Design for anti-bird impact of aero-engine fan blade
ZHANG Haiyang, WANG Xiangping, DU Shaohui
2020, 35(6): 1157-1168. doi: 10.13224/j.cnki.jasp.2020.06.005
Abstract:
The first stage fan rotor blade is a key component of the engine against bird impact, and the leading edge of the blade is a key part of the anti-bird impact design, by investigating the structural characteristics of the engine and the damage characteristics after bird impact. Therefore, a theoretical analysis method of fan blade bird impact was established to study the relationship among impact conditions, structural parameters and bird impact process, and damage mode as well as damage degree. Results show that the leading edge angle is the key structural parameter of anti-bird impact. The leading edge angle determined the impact process, damage mode and damage degree, when the bird impact conditions were determined. A model with leading edge characteristics was designed and the influence rule of leading edge angle was verified using the method of display dynamics simulation analysis. Based on the results, an improved design of the actual fan blade was carried out. The deformation of improved blade was reduced by a minimum of 33% after being hit by a bird, and the anti-bird impact ability of the improved blade was obviously improved.
Test of vibration fatigue for the TiAl alloy aeroengine blade at high temperature and high cycle
ZHANG Busheng, ZHU Jizhi, SHI Jian
2020, 35(6): 1169-1175. doi: 10.13224/j.cnki.jasp.2020.06.006
Abstract:
Taking TiAl alloy aeroengine blade as the research object, a series of problems likely to happen during the test of the vibration fatigue for aeroengine blade at high temperature and high cycle were studied, including fatigue stress monitoring at high temperature, and high frequency excitation. The closed loop of maximum stress was used to solve the high-temperature fatigue stress monitoring. The high frequency excitation was realized through design of fixture amplification. The temperature load and vibration load were applied synthetically by radiation heating and electromagnetic vibration shaker. Using the method of the vibration fatigue at high temperature and high cycle, the life test was carried out on the blade. The control accuracy of high temperature fatigue stress was better than ±2%. The median fatigue limit value of the blade with 50% reliability was 444 MPa. The life curve of the blade was obtained effectively. This test method is suitable for the high temperature and high cycle vibration fatigue test of the aeroengine blade, providing a reference to the high temperature and high cycle vibration fatigue test of the other aeroengine parts.
Finite element analysis of vibration frequency and damping of ,cylindrical shell partially covered viscoelastic damping layer
CHEN Zhongshi, SUN Wei
2020, 35(6): 1176-1185. doi: 10.13224/j.cnki.jasp.2020.06.007
Abstract:
On the basis of introducing the complex modulus model of viscoelastic materials and considering the elastic boundary of the cylindrical shell, the finite element program was developed to solve the nonlinear calculation problem of vibration frequency and damping for the composite cylindrical shell in view of the frequency dependence. A composite shell element with 4 nodes and 24 degrees of freedom was created to simulate the mechanical behavior of the cylindrical shell with partially covered viscoelastic damping layer, and the stiffness and mass matrices of the element were derived. A circumference variable stiffness elastic constraint model with 6 spring groups was proposed to simulate the elastic boundaries at the bottom of the cylindrical shell. The dynamic finite element equations of the composite cylindrical shell were determined, and the iterative calculation process of solving its vibration frequency and damping by eigenvector increment method was described. A case study for the cylindrical shell with ZN-1 free damping layer was performed and the results showed that the maximum error between the calculation results of the proposed algorithm and the experiment values was 369%. In addition, the frequency dependence of the viscoelastic material had an effect on the natural frequency of the composite structure of less than 001%, but the maximum effect on the modal loss factor was 10947%.
Oil atomization analysis of jet lubrication for high speed angular contact ball bearing
LIU Hongbin, HAO Jinhua, JIA Qisu
2020, 35(6): 1186-1194. doi: 10.13224/j.cnki.jasp.2020.06.008
Abstract:
Angular contact ball bearing was taken as investigation object and FLUENT fluid calculation software was used to establish a simulation model of gas-liquid two-phase flow between bearing rings, so as to analyze the atomization of injected lubricating oil liquid under the action of high-speed airflow between rings. The change of oil particle diameter was discussed under the conditions of different rotating speeds and different injection angles at the same time. And the change trend of Sauter mean diameter (SMD) at different times was also discussed. Results showed that with the penetration of lubricating oil, the proportion of oil with large particle size in the cavity decreased gradually, while the proportion of oil with small particle size increased. With the increase of the rotating speed, the amount of oil entering the cavity would also decrease. Influenced by the air vortex between the rings, the oil droplets would disintegrate rapidly, which would intensify the atomization and reduce the particle diameter, thus reducing SMD and affecting the bearing lubrication. Under the condition of different injection angles, the proportion of large particle size in the bearing cavity with the injection angle of 15° was relatively large.
Optimization design and experiment for cylindrical ,roller bearings to 3 million DN
LI Hongliang, ZHENG Yanwei, GU Yunlong
2020, 35(6): 1195-1203. doi: 10.13224/j.cnki.jasp.2020.06.009
Abstract:
In order to resolve the failures of cylindrical roller bearings under the condition of high DN(bearing inner diameter×inner speed), high DN cylindrical roller bearings were developed. Based on the design criteria of bearings, the structural parameters optimization design and material selection for cylindrical roller bearings to 3 million DN were carried out. Based on the dynamic theory of bearings, the dynamic model of bearings was established and simulated. The dynamic behaviour for cylindrical roller bearing to 3 million DN was investigated; finally, the bearing test was carried out on the self-developed testing machine to verify the rationality of the structural parameters and material selection for cylindrical roller bearing to 3 million DN. During the test, the temperature rose and vibration acceleration of the bearing was monitored. Result showed that when DN was 3 million, the temperature was lower than 130 ℃ and vibration acceleration was lower than 20g. Through optimized design, the DN of cylindrical roller bearing has increased from 2 million to 3 million.
Thermal characteristics for bearing combined EHL theory with CFD method
LU Fengxia, WANG Tao, ZHAO Zhiqiang
2020, 35(6): 1204-1211. doi: 10.13224/j.cnki.jasp.2020.06.010
Abstract:
In order to accurately predict the thermal characteristics of angular contact ball bearing in the transmission system under spray lubrication, the bearing thermal elastohydrodynamic lubrication (EHL) model was established, and the micro-contact interface load and friction coefficient between the raceway and ball were obtained. On the basis, heat generation was calculated by using the local generation. Computational fluid dynamics (CFD) method was used to establish a coupled oil-gas two-phase flow model for angular contact ball bearings, including the choice of turbulence model and fluid domain boundary conditions. Furthermore, fuel injection speed, nozzle position and gas fraction were analyzed. Results showed that the application of CFD method with accurate heat generation calculated by the thermal EHL theory could predict the optimal injection speed, the nozzle position and gas fraction of the bearing under various operating conditions, lubrication parameters and geometric parameters. When the injection speed was 5 m/s, the maximum temperature of the bearing chamber was decreased by 440% compared with other injection speeds. When the nozzle position was below the bearing axis, the maximum temperature of the bearing chamber decreased by 430% compared with other positions. When gas fraction was 15%, the maximum temperature of the bearing chamber was improved by 157% compared with other gas fraction.
Remaining useful life prediction of rolling bearings using, InfoLSGAN and AC algorithm
YU Guangbin, ZHUO Shi, YU Jun
2020, 35(6): 1212-1221. doi: 10.13224/j.cnki.jasp.2020.06.011
Abstract:
In order to solve the problem of low remaining useful life (RUL) accuracy of rolling bearings under small samples and noise interference,a RUL prediction method of rolling bearings using information least squares generative adversarial network (InfoLSGAN) and actor-critic (AC) algorithm was proposed. Stacked denoising autoencoder,information generative adversarial network and least squares generative adversarial network were integrated to construct InfoLSGAN,which can automatically extract interpretable and robust features from noisy data,and solve the problem of vanishing gradients. The training algorithm based on AC was utilized to train the InfoLSGAN to reduce the training time and accelerate the convergence. According to the InfoLSGAN after training,a softmax classifier was used to predict the rolling bearing RUL in test samples. The effectiveness of the proposed method was validated through an accelerated fatigue life experiment of rolling bearings. The experimental results demonstrated that when the signal-to-noise ratio was equal to 0,the proposed method increased the RUL prediction accuracy of rolling bearing test samples by at least 10%. In the case of small samples,the average accuracy of the RUL prediction of rolling bearings was 9584%.
Failure analysis on central drive bevel gear of turbo-shaft engine with complex excitation
CHEN Xueqi, MA Yanhong, WANG Yongfeng
2020, 35(6): 1222-1227. doi: 10.13224/j.cnki.jasp.2020.06.012
Abstract:
A fatigue failure of the central driven bevel gear was found at the front end of the high pressure rotor of a small turbo-shaft engine. Taking into consideration of both the gear teeth meshing and rotor-gear coupling vibration, the vibration response of the driven bevel gear under the combined influences of tooth surface meshing excitation and rotor lateral vibration excitation was investigated. The analysis results showed that the gear teeth meshing caused the 5th nodal diameter shape vibration, while the coupled vibration of the rotor-driven gear system led to a 2× excitation and then excited the swing shape vibration of the driven bevel gear. Meanwhile, the additional constraints arousing from the gear teeth meshing changed the vibration stress distribution of the swing shape vibration, which then forced the fatigue crack to propagate radially through the gear. The distribution of vibration stress is consistent with the phenomenon of the driven gear fatigue failure, which proves the correctness of the failure mechanism analysis.
Experiment of extremely low wind velocity measurement based on laser Doppler
XU Dachuan, LONG Yanzhi, LI Yudong
2020, 35(6): 1228-1237. doi: 10.13224/j.cnki.jasp.2020.06.013
Abstract:
The principles of laser Doppler effect and telemetry were introduced, and the requirement for the size of aerosol particles was analyzed. A low wind velocity measurement system was established, and through which the measurement accuracy and influence of installation angle were analyzed. The experiments of rotary table, linear guideway and wind tunnel were carried out. The results verified the analysis and proved that the measurement accuracy was better than 001 m/s by experimental data. The measurement system can obtain stable and reliable signal when the medium in wind tunnel was conventional air, and the measurement results were very close to the high-performance instrument.
A method of propeller design with given thrust distribution and it’s application
GUO Jiahao, ZHOU Zhou, FAN Zhongyun
2020, 35(6): 1238-1246. doi: 10.13224/j.cnki.jasp.2020.06.014
Abstract:
Starting from the propeller design, a favorable propeller slipstream form was obtained by designing a suitable propeller to improve the aerodynamic characteristics of the wing in the propeller/wing configuration. The propeller slipstream was controlled by changing the radial thrust distribution of the propeller. A method of propeller design based on the given thrust distribution was proposed and the feasibility of the method was verified. Propellers designed with minimum energy loss and different thrust distributions were applied to multi-propeller/wing configuration, so as to study the effect of the propeller on the wing. Results showed that changing the propeller force distribution can change the propeller slipstream, which in turn changed the aerodynamic characteristics of the wing. The propeller with a gentle thrust distribution can increase the lift and reduce the drag on the wing, despite of its efficiency decrease. And compared with the minimum energy loss design results, the lift-drag ratio can be increased by 994%.
Wide stability margin aerodynamic optimization design of supersonic profile
HAN Lu, ZHOU Zhenggui, CHEN Jinfan
2020, 35(6): 1247-1256. doi: 10.13224/j.cnki.jasp.2020.06.015
Abstract:
By comparing with the experimental data of PAV-15 designed by German space agency, the calculation method of flow field in supersonic cascade with high accuracy was determined. The results showed that the consistency between the calculation and the experimental results can be improved by adjusting the thickness of the flow tube according to the position of shock wave. In order to improve the stability margin and ensure the performance at design point of supersonic cascade, the back pressure estimation method and optimization design method based on target margin were established. The multi-objective optimization of the two supersonic blades was carried out. The optimization results show that the optimal cascade can reduce the incident angle of the shock wave at design point, and cut down the shock wave and shock wave boundary layer interference loss. The aerodynamic throat moving forward and the terminal normal shock wave moving backward can improve the shock-holding capability. Under the premise of keeping the total and static pressure ratio unchanged, the stability margins of the two optimal blade profile reached the design goal, and the loss at design point also decreased.
Performance of ducted rocket with typical fuel rich propellant
LI Xintian, ZHAO Xiaoning, CAI Qiang
2020, 35(6): 1257-1265. doi: 10.13224/j.cnki.jasp.2020.06.016
Abstract:
Performance prediction model of ducted rocket was developed based on thermodynamic calculation. Research on performance characteristics of ducted rocket with three kinds of fuel rich propellants was conducted. The study shows that under the same height and Mach number, as excess air coefficient increases, inlet stability margin increases and thrust coefficient decreases, while specific impulse increases first and then decreases. Under the same height and excess air coefficient, as incoming Mach number increases, the inlet stability margin increases, while thrust coefficient and specific impulse decrease. Under the same incoming Mach number and excess air coefficient, the changes of inlet stability margin, thrust coefficient, and specific impulse caused by height variation are not obvious. The variations of characteristic velocity and ducted rocket performances are obvious with the increasing of air total temperature, which should not be neglected in performance analysis. Compared with hydrocarbon fuel rich propellant and aluminum magnesium fuel rich propellant, boron based fuel rich propellant has moderate equivalent air-fuel ratio, higher specific impulse, and higher density, which revealings advantages in engineering application.
Research on simulation and test of vacuum plume diversion technique based on vehicles taking off outside the earth
YE Qing, SHU Yan, ZHANG Xuhui
2020, 35(6): 1266-1274. doi: 10.13224/j.cnki.jasp.2020.06.017
Abstract:
In view of the problem of limitations on engine plume diversion, based on engineering experiences, four typical molding surfaces of diversion devices (including form of inner groove and form of diversion cone) were proposed to carry out numerical simulation of the aerodynamic and aerothermal effects caused by plume diversion during take-off using the coupling simulation method of computational fluid dynamics (CFD) and direct simulation of Monte Carlo (DSMC), and analyze the plume field shock wave and spacecraft surface pressure and heat flow density distribution law, thereupon the evaluation on diversion effects of the four diversion devices was presented, and finally tests in the form of diversion core were conducted to verify the simulation algorithm. The result show that the engine is stable with the plume diversion, the scheme of the large diversion core is optimal considering the spacecraft plume and engine safety, the shock near the diversion core is the same as the simulation, the variation trend of simulation and test is consistent, the algorithm is credible and the law can be applied in engineering.
Research progress of ionic liquid micropropulsion technology
FAN Yipeng, XIA Guangqing, HAN Yajie
2020, 35(6): 1275-1285. doi: 10.13224/j.cnki.jasp.2020.06.018
Abstract:
The basic structure and working principle of ionic liquid thrusters, including the restrictions on particle emission and the commonly used operating regimes, were introduced. The common classification methods of ionic liquid thruster were summarized. The presently wide-used experimental or simulation methods, and the related progress on emission threshold field strength, beam divergence, polydispersive efficiency, long-term working stability and other issues were reviewed to compare and analyze the working conditions suitable for particle emission and more accurate simulation methods. The future research ideas and reference for the improved design, operating regime setting, and performance evaluation of ionic liquid thruster were also presented. Results show that increasing the flow resistance of propellant and array density of emitter is the appropriate means to improve the efficiency and thrust of the ionic liquid thruster, and changing the emitter voltage polarity and increasing the emitter voltage gradually by using the closed-loop control method is the effective way to maintain the thrust and improve the working stability of the thruster.
High order direct numerical simulation of compressor cascade ,channel separated flow
ZHU Haitao, LI Yan
2020, 35(6): 1286-1295. doi: 10.13224/j.cnki.jasp.2020.06.019
Abstract:
Separated flow of V103 compressor cascade was numerically simulated by directly solving the two-dimensional Navier-Stokes equations using high order finite difference schemes.The numerical results showed that in the transient flowfield, there existed an obviously separated flow at the rear of the suction side with a large-scale separated vortex followed by alternative second vortex and shedding-vortex forming the wake. In the time-averaged flowfield, a short separated bubble was formed and identified by the pressure platform of the pressure distribution on the suction side of the blade. Compared with two-dimensional flat separated flow, both the transient and time-averaged flowfields were similar with the same vortex structure. However, the non-dimensional vortex-shedding frequency of cascade was twice times of the flat flow. The time-averaged pressure distribution on the blade surfaces coincided with the numerical results in the references except the separation area. Compared with the reference results, the present separated flow axial length was 41% bigger than the former. At last, second order statistics of pulse velocity in the separation area were bigger than those in the wake, indicating the unsteadiness of the separated flow.
Improved design for flutter free of fan rotor blade
ZHANG Yaoguang, YANG Lin, LIU Yixiong
2020, 35(6): 1296-1303. doi: 10.13224/j.cnki.jasp.2020.06.020
Abstract:
Through aeroelastic stability prediction by energy method in original scheme of a two-stage transonic fan, the geometric modeling of aeroelastic unstable rotor was modified to improve the aeroelastic stability. Through comparative analysis of the geometric modeling, aerodynamic performance, vibration characteristic and aeroelastic stability of original and modified schemes, it showed that the minimum modal aerodynamic damping ratio of modified scheme increased from -146 to 183, eliminating the flutter risk, while the aerodynamic performance of the fan was kept unchanged and the modification to the original scheme was small. Increasing thickness and chord length can increase aeroelastic stability of blade at corresponding radial position, making it an effective means to improve flutter; however, attention should be paid to the change of aerodynamic performance. With the same reduction of channel blockage margin, adjustment of thickness distribution is the best way to improve aeroelastic stability.
Influence of form of tip gap on aerodynamic performance of centrifugal compressor using mixture of helium and xenon
WANG Guojie, LIU Xuezheng, GAO Jie
2020, 35(6): 1304-1314. doi: 10.13224/j.cnki.jasp.2020.06.021
Abstract:
The tip leakage flow directly affects the performance of the centrifugal compressor. The influence of the independent change of the radial and axial tip gaps and the influence of the distribution of the tip gap on the centrifugal compressor internal flow field and performance were studied by numerical simulation. The influence law of different height and shape of the centrifugal compressor tip gap on its performance was summarized. Result showed that increasing the centrifugal compressor gap would reduce the efficiency at near surge condition and design condition; increasing the radial gap solely, from 01 mm to 04 mm, could broaden the choke mass flow range by 388% and improve the performance of the centrifugal compressor near blockage condition; increasing the axial gap solely had no effect on the performance at near choke condition. The shrinking distribution of tip gap increased the efficiency of the centrifugal compressor by 023%.
Modeling and flow field analysis of large-diameter butterfly valve
ZHANG Song, DAN Zhihong, LI Teng
2020, 35(6): 1315-1325. doi: 10.13224/j.cnki.jasp.2020.06.022
Abstract:
The motion characteristics and numerical simulation method were developed to obtain input-output characteristics of a large-diameter butterfly valve. Under given operating conditions, velocity, pressure, torque and pressure drop were calculated to obtain the flow field of butterfly valve. Then, using some characteristic parameters such as velocity profile, pressure distribution, turbulence kinetic energy and turbulence intensity, the model of butterfly valve was established by fitting method and compared with experimental data. Results illustrated that flow field characteristic differed at different opening degrees, and when the opening degree was equal to or greater than 5365%, the flow velocity of the butterfly valve fluid at the inlet and outlet was relatively full, the flow performance was relatively good and the flow state was stable. The mathematical model of butterfly valve modified by the experimental data has high degree of confidence, making it feasible to analyze the motion characteristics of butterfly valve.
Fault diagnosis of liquid rocket engine based on comprehensive fuzzy clustering algorithm
DONG Zhoujie, GUO Yingqing
2020, 35(6): 1326-1334. doi: 10.13224/j.cnki.jasp.2020.06.023
Abstract:
Based on the completeness of the normal and fault condition data of liquid rocket engine and the improvement of data quality, a data-driven comprehensive fuzzy clustering algorithm was proposed for fault diagnosis. The fuzzy c-means (FCM) algorithm was used to cluster the known normal sample data to obtain the optimal clustering center, and the obtained cluster center was used as a-priori sample data to select the optimal classification result of the closure method to obtain the fault detection result. Only a small amount of normal prior sample data were required to quickly and accurately detect the fault; then the FCM algorithm was used to classify the fault, and the corresponding fault type can be clustered according to the existing fault database, and the range of fault amplitudes can be given. The simulation results showed that the detection rate of the algorithm was up to 968% and the fault isolation rate was 94%. The actual test data of a liquid rocket engine show that the fault diagnosis algorithm can detect and isolate faults accurately and timely.
Experiment on combustion characteristics of micro gas turbine combustor
LIU Aiguo, LI Yuze, YANG Yudong
2020, 35(6): 1335-1344. doi: 10.13224/j.cnki.jasp.2020.06.024
Abstract:
According to the requirements of low pollution emission of gas turbine, the combustion characteristic experiment was carried out on the single head experiment piece of the combustor of low emission micro gas turbine designed for the use of gas fuel. The position and quantity of the fuel nozzle, the fuel distribution ratio between the duty class and the main combustion stage, and the combustion efficiency and pollutant emission characteristics under different air flow distribution modes were compared and analyzed. Results showed that changing the position and quantity of the fuel nozzle can change the mixing characteristics of the fuel and air, yielding a great effect on the combustion characteristics; the change of the fuel flow distribution ratio between the duty class and the main combustion stage led to the change of the equivalence ratio of each combustion zone, and the reduction of the current ratio of the main combustion stage to below 08 was conducive to the reduction of pollutant emissions; the change of air flow distribution ratio of the combustion chamber was conducive to the reduction of pollutant emissions With this method, the equivalence ratio of the main combustion zone can be reduced, the NOx emission can be reduced to 272 mg/m3, and the combustion efficiency can reach 986%. However, the decrease of air flow rate for mixing could increase the outlet temperature distribution coefficient from 021 to 024.