2017 Vol. 32, No. 3

Display Method:
High precision automatic measurement system for aero engine blade
2017, 32(3): 513-518. doi: 10.13224/j.cnki.jasp.2017.03.001
Abstract:
In order to solve the current problems of the blade surface automatic measurement, a high precision automatic measurement fusion system was put forward. The self-developed measuring equipment and the electronic controlled turntable were combined into one complete measurement system. Using high precision calibration plane rotation, the high precision quaternion and the center position coordinates could be obtained for the system location through the plane fitting, virtual axis and angle calculation. The real time measurement and fusion of the blade could be realized by using the center coordinate and the four element number operation. The convergence of the threshold iterative closest point(ICP) to the fusion data was used to eliminate the error of mechanical rotation. Results show that the integrated precision of the system is 0.03-0.04mm, which can realize the high precision measurement of the engine blade automatically, efficiently and stably.
Model description of rotating stall boundary in low-speed axial compressor
2017, 32(3): 519-527. doi: 10.13224/j.cnki.jasp.2017.03.002
Abstract:
Starting from the typical dynamic characteristics of the complete rotating stall process of low-speed axial compressor, such as catastrophe and hysteresis, by comparing and analyzing the topological properties, the rotating stall and cusp catastrophic model of catastrophe theory were proved similar in spatial topology; then, based on the topological invariance theorem through developing the effective topological mapping method, the topological mapping relationship of the catastrophe point set between the rotating stall and the cusp catastrophic model was established. The applied effect of the model in two low-speed axial compressors showed that cusp catastrophic model can qualitatively describe the catastrophe, hysteresis and bifurcation of the complete rotating stall process of low-speed compressor with the characteristic of “abrupt stall” under different speed conditions; through the mapping method developed, the stall points and recovery points obtained by experiment can be entirely mapped to the model line, which means the cusp catastrophic model can accurately describe the instability boundary of rotating stall.
Comparison of computation models for simulation of turbomachinery rotor-stator interactions
2017, 32(3): 528-537. doi: 10.13224/j.cnki.jasp.2017.03.003
Abstract:
Comparative study on three Fourier-based time-domain unsteady computation methods, including the shape-correction phase shift method, space-time gradient method and harmonic balance method, was carried out by simulation of two typical kinds of rotor-stator interactions, i.e.:wake-blade interactions and shock-blade interactions. Differences and relations between these computation methods were investigated in detail, and their computational accuracy and speed were also analyzed. The results indicate that three methods can resolve the unsteady flow details well. Compared with the conventional time-marching method, the computational speed of these methods can be enhanced from several to dozens of times, making these methods applicable to daily design process.
Numerical simulations on circumferential propagating characteristics of tip leakage flow oscillation in compressor
2017, 32(3): 538-548. doi: 10.13224/j.cnki.jasp.2017.03.004
Abstract:
Through numerical investigations on the characteristics of tip leakage flow oscillation in a low speed axial compressor, it was found that at large flow rate conditions, the tip flow field did not vary with time. As flow rate decreased, the tip flow field gradually started to fluctuate, meanwhile the tip leakage flow oscillated periodically, showing that the origin position, vortex structure pattern and impinging position on the pressure surface of neighboring blade changed with time. The tip leakage vortex and its detached vortex induced low pressure regions on the neighboring blade surface, leading to the oscillation of pressure distribution in blade tip region. Then this also affected the generation of the tip leakage flow in the neighboring passages. The interaction effect of tip leakage flow and the neighboring blade was the main cause of the tip flow field fluctuation. Furthermore, the tip leakage flow oscillations had phase lag between neighboring blade passages, which induced a flow disturbance with non-blade-number periodicity propagating circumferentially around the rotor.
Optimization design of multi-section airfoils for heavy-duty gas turbine compressor
2017, 32(3): 549-557. doi: 10.13224/j.cnki.jasp.2017.03.005
Abstract:
Considering the spanwise difference of both airfoil and aerodynamics boundary, an optimization design process of multi-section airfoils was developed. Inlet guide vane and large camber angle stator of a single-stage highly loaded compressor were taken as optimization subjects, and multi-section associated performance was taken as optimization target. MISES considering boundary layer transition was adopted to analyze the aerodynamic performance of airfoils, and genetic algorithm was also adopted to optimize the multi-section airfoils. The results show that, within the valid operating range, the front loading of multi-section airfoils increases after optimization, while the central and rear loading decreases, bring about performance improvement. Three-dimensional numerical simulation results show that, without large scale secondary flow, the optimization method of multi-section airfoils is practicable.
Design and optimization of end zone of large meridional expansion adjustable blades on variable geometry turbine
2017, 32(3): 558-567. doi: 10.13224/j.cnki.jasp.2017.03.006
Abstract:
The high load design was adopted to reduce blades to increase diameter of disc while reducing the circumferential leakage area. And aft-loaded cascades were employed to retrofit design to minimize the secondary flow loss brought about by high load design. For the fixed geometry turbine, only the flow field of the variable geometry turbine of driving axis and disc type crown variable geometry turbine was numerically calculated. Three kinds of turbine performance under all conditions were analyzed. Result shows that, under the big clearance, intensity of the clearance leakage vortex was big. If merged with the passage vortex, the leakage loss became bigger. The high load design brought the secondary flow loss. The method reduced the circumferential leakage area, restraining greatly the merges between the leakage flow and the secondary flow to reduce the clearance leakage loss. High load design can effectively restrain the development trend of passage vortex, thereby reducing the secondary flow loss. With two measures, the variable geometry turbine has high performance under overall conditions.
Numerical simulation of flat baffle inlet distortion coupled with compressor
2017, 32(3): 568-576. doi: 10.13224/j.cnki.jasp.2017.03.007
Abstract:
In order to explore the integrated design method of inlet/engine,a simulation method was developed to analyze the coupled calculation of inlet distortion and compressor. First, a full-passage body force model for compressor was built by using blade force extracted from Reynolds average Navier-Stokes (RANS) simulation results. Second, by adding the compressor body force model into the CFX software, the flat baffle inlet distortion coupled with compressor was simulated and analyzed. Comparing the simulation results with experimental results, the compressor body force model could simulate the strong coupling interactions between the compressor and the upstream distortion flow field. And the body force model could reduce almost two magnitudes of mesh numbers compared with the compressor full-passage RANS model.
Influence of fuel stage under idle condition on combustion efficiency for high-temperature rise combustor
2017, 32(3): 577-583. doi: 10.13224/j.cnki.jasp.2017.03.008
Abstract:
The combustion efficiency test was done on the single dome test piece of twin annular multi-point direct injection mixing(TAMDIM) high-temperature rise combustor, the idle combustion efficiency of two stages-fueled together mode and the pilot-fueled only mode with two different throat spacing sizes were compared, and the influence of different main and pilot equivalents (the pilot fuel ratio of 40%, 56%, 65%, 100%) and throat spacing size on the combustion efficiency were analyzed. Test results show that, the throat spacing size has a great influence on the combustion efficiency when taking the two stages-fueled together mode, compared with the pilot-fueled only mode, better combustion efficiency has been obtained when the spacing of throat is 19.3 mm and the pilot fuel ratio is within the range of 40%~56%.Validated test has been done on the sector test piece finally, ulteriorly verifying the feasibility of fuel stage under idle condition in the TAMDIM high-temperature rise combustor.
Composite cooling structure on rotating turbine blade
2017, 32(3): 584-591. doi: 10.13224/j.cnki.jasp.2017.03.009
Abstract:
Composite cooling performance of a rotating turbine blade cooling structure was investigated through heat transfer simulation with three-dimensional fluid-solid coupling, and the effects of radiation heat transfer and rotational speed on overall cooling effectiveness were discussed. Results showed that a local high-temperature area around the blade hub and hence a large scope of low cooling effectiveness occurred, leading to a nonuniform distribution of overall temperature for blade structure 1. The cooling effectiveness adjacent to the leading edge was improved, lessening the low effectiveness area as a result of more reasonable film flow distribution of blade structure 2. The cooling effectiveness on blade suction side was markedly enhanced through internal serpentine passages and a uniform distribution of integral cooling effectiveness was obtained for blade structure 3. The blade temperature increased with a regional temperature increment over 50K for the surface emissivity of 1, while the effect of radiation heat transfer cant be ignored. The overall cooling effectiveness on pressure side improved with the augmentation of rotational speed, resulting in a highest regional increase percentage of 15.6%, 13.4% and 16.4% for the three structures, respectively. Nevertheless, the cooling effectiveness on suction side produced little change except a reduction at mid-chord region with the increase of rotational speed.
Impingement heat transfer inside wedge-shaped concave by jets from impingment plate with front-extended port
2017, 32(3): 592-598. doi: 10.13224/j.cnki.jasp.2017.03.010
Abstract:
Experimental analysis was conducted to investigate the heat transfer inside a wedge-shaped concave by a row of jets from impingment plate with front-extended port. The effect of front-extended port extension spacing ratio (5~11), front-extended port width spacing ratio (2.5~8) and jet Reynolds number (7900~31700)on the concave surface temperature distribution, spanwised-averaged Nusselt numbers,area-averaged Nusselt numbers and jet pressure loss were experimentally tested. The results indicated that the convective heat transfer of impinging plate with front-extended port was obviously enhanced compared with baseline plate, the front-extended port caused high jet pressure loss. The convective heat transfer of concave surface was significantly improved with the increase of front-extended port extension spacing ratio, yielding little effect on jet pressure loss. Convective heat transfer of concave surface was enhanced with increase of front-extended port width spacing ratio, however, the total pressure loss increased also.
Experiment on influences of inlet temperature and pressure on vortex tubes performance
2017, 32(3): 599-606. doi: 10.13224/j.cnki.jasp.2017.03.011
Abstract:
Experiments were carried out to study the influences of the inlet/outlet expansion ratio (2~5), the throttle setting value (1~11) and the inlet temperature (313~373K) on vortex tubes performance. The variation of refrigeration and heating effects was obtained along with above parameters. It was found in experiments that the performance of vortex tube was enhanced with the increasing inlet expansion ratio, promoting the in-tube vortex intensity and increasing inlet the angular velocity gradient at different radiuses. The refrigeration effect increased by 10.9% and heating effect increased by 46.5% separately. With the rise of throttle setting value, the degree of energy exchange between the cold/heating flow increased, the eddy effect strengthened and cold flow rate increased by 47.3%. The performance of the vortex tube was improved with the increase of the inlet temperature. All the results show that as the inlet temperature increases by every 20K, the refrigeration effect at the cold side is enhanced by about 12%, while the heating effect at hot side is improved by about 5%.
Three-dimensional large eddy simulation of strut flame-holder with cavity
2017, 32(3): 607-613. doi: 10.13224/j.cnki.jasp.2017.03.012
Abstract:
In order to study the cold flow field of strut, three-dimensional LES (large eddy simulation) of strut flame-holder with cavity at Ma=0.06 and Ma=0.09 was carried out respectively. The simulation results were compared with experiment data. It showed that the flow field structure of three-dimensional LES was similar with that of experiment. Whats more, two quantitative comparisons in the frequency of wake vortex and time-averaged stream-wise velocity on strut rear section were made. And the maximum relative errors were 9.4% and 7.7% respectively, proving that the model of numerical calculation at boundary conditions was correct and believable. Then by analyzing the three-dimensional flow field of strut, a conclusion that the faster velocity means the bigger size and intensity of wake vortex. Besides, the result of three-dimensional LES shows an obvious phenomenon of three-dimensional effect which gets stronger when the velocity gets faster.
Investigation on aerodynamic and infrared characteristics of spherical convergent two-dimensional vectoring nozzle: simulating ground status
2017, 32(3): 614-620. doi: 10.13224/j.cnki.jasp.2017.03.013
Abstract:
Numerical investigation was performed to study the aerodynamic and infrared performances of spherical convergent two-dimensional vectoring nozzle adopted in turbofan engine exhaust system, only the ground status simulation and pitch deflection were taken into consideration. Results showed that, the pitch deflection had a very weak impact on the thrust coefficient and total pressure recovery coefficient of exhaust system when the pitch deflection angle was less than 20 degree. The aerodynamic thrust vectoring angle was nearly the same as the pitch deflection angle. Due to the feature of volumetric thermal radiation of gas, the variation of nozzle pitch deflection angle resulted in significant change of infrared radiation directionality of hot plume. The infrared radiation of hot plume in 3-5μm band was also increased within a certain scope in relation to the no-vectoring case. With respect to the total infrared radiation of exhaust system in 3-5μm band, the peak infrared radiation intensity decreased as the pitch deflection angle increased. The detecting angle corresponding to the peak infrared radiation intensity was smaller than the pitch deflection angle. Under a big pitch deflection angle, the total infrared radiation of exhaust system viewed from top in the vertical plane was lower than that of no-vectoring case. However, the total infrared radiation of exhaust system viewed from bottom in the vertical plane was higher than that of no-vectoring case, resulting in a secondary peak value.
Influence of exterior hot-film on droplet impingement characteristics over aero-engine inlet strut
2017, 32(3): 621-629. doi: 10.13224/j.cnki.jasp.2017.03.014
Abstract:
Four ice-protection structures with different film-slot angles were designed, an Eulerian-framework based droplet impingement computation method was developed and verified to study the influences of film-slot angle and blowing ratio on droplet impingement characteristics for 20μm diameter. The results indicated that the exterior hot-film can blow the droplet away to decrease local collection efficiency and impingement limitation significantly, and the influence was more evident when the position of film-slot was closer to strut leading edge, the average local collection efficiencies of four structures declined 82%, 8%, 1% and 0.5%, respectively, compared with no-film situation. Besides, the maximum local collection efficiency and impingement limitation decreased with the increasing blowing ratio. The impingement characteristics of film-slot angle 5°was most sensitive to blowing ratio. Similar local collection efficiency distribution features were presented at leading edge. For rear area, when blowing ratio was big enough, no droplet impinged on the surface.
Effects of heat soakage on transient performance of gas turbine engine
2017, 32(3): 630-636. doi: 10.13224/j.cnki.jasp.2017.03.015
Abstract:
With an unsteady heat transfer model, an assumption of neglecting the effects of heat transfer on the work of rotating components and a simplified heat transfer model of cooled turbine were put forward, and a transient engine model was built by taking the heat soakage into account. The numerical simulation results of a turbofan engine transient performance showed that,the main effects of heat transfer soakage on the engine performance were attributable to the response delay in each component, which was more serious in low pressure components. The thermal time constants of the compressor were smaller than those of the turbine. For the thermal time constants of the same component, the blade was the smallest, the hub was larger, and the casing was the largest.
Analysis of hammer shock load characteristics in S-duct
2017, 32(3): 637-647. doi: 10.13224/j.cnki.jasp.2017.03.016
Abstract:
To analyze the effect of instantaneous high pressure profile caused by engine surge on the hammer shock propagation, an S-duct model was simulated under conditions of both subsonic and supersonic inflow, based on improved delayed detached eddy simulation (IDDES) method. Results showed that when hammer shock was generated, it propagated upward along the inlet rapidly with approximate constant speed and the strength of hammer shock increased. Meanwhile, the aerodynamic load of S-duct inner wall was uneven under the influence of centrifugal force field; the aerodynamic load on side wall with the larger radius was greater. Increasing the pressure loading gradient led to quicker hammer shock propagation speed and enhanced strength, however different unloading pressure profiles had no obvious effect on the hammer shock propagation speed and strength. Under both subsonic and supersonic inflow conditions, the intensity of hammer shock increased with the growth of maximum value of instantaneous high pressure, approximately matching the quadratic function distribution law. The hammer shock intensity of supersonic inflow was greater than the subsonic inflow.
Application strategy and improvement of unstructured dynamic grid method based on elasticity analogy
2017, 32(3): 648-656. doi: 10.13224/j.cnki.jasp.2017.03.017
Abstract:
Taking the field of application as a focal point, three classical dynamic grid methods based on elasticity analogy were analyzed in many ways, such as deformation capability, deformation quality and spatial distribution. Combined with engineering demand, the application strategy for grid deformation was presented. Finally, a improvement was proposed for integrating advantages of classical dynamic grid methods based on elasticity analogy, and then verified through pitch oscillation of NACA 0006 airfoil in hypersonic condition. Results show that, when faced with densified grids, the area modification can be controlled better, but faced with uniform grids, the distance modification is more suitable, so this application strategy can provide a guidance for engineering. Secondly, the improvement is accurate in simulation, its all grid quality parameters exceed 10% than classical method based on distance, whats more, its minimum grid quality parameter is about four times the classical method based on area. In conclusion, the improvement is effective in strengthening the robustness of classical elasticity analogy.
Aerodynamic performance of configurations of two-dimension inflatable wings under high Reynolds number
2017, 32(3): 657-665. doi: 10.13224/j.cnki.jasp.2017.03.018
Abstract:
The aerodynamic performance analysis based on the configurations of two-dimension inflatable wings under high Reynolds number was analyzed. Firstly, configuration characteristic of the two-dimension inflatable wing was designed to define an offset parameter describing the degree of approximation between an inflatable wing profile and a standard airfoil, then several models were developed. Furthermore, with the numerical method, aerodynamic performance of the inflatable wing and the sensibility of the offset parameter was studied and compared with the standard airfoil. Numerical result showed that the aerodynamic performance of the inflatable wing had loss to a certain degree under high Reynolds number. Meanwhile, with the flow field characteristic analysis, the reason for the total drag increasing significantly was explained from mechanism that, in those bumpy areas, the pressure distribution changed a lot, so the increase of the local pressure caused significant increase of pressure drag.
Analysis on vibration suppression of mistuned bladed disk via bi-periodic distributed piezoelectric shunt damping
2017, 32(3): 666-676. doi: 10.13224/j.cnki.jasp.2017.03.019
Abstract:
A eletromechanical coupling system was established by introducing piezoelectric shunt circuit into mistuned bladed disk. The eletromechanical coupling system dynamic equations were derived, and the suppression effect on the amplitude magnification were analyzed. By constructing bi-periodic distributed piezoelectric shunt circuit, vibration suppression effect of bi-periodic distributed piezoelectric shunt damping was analyzed. The results indicate that, piezoelectric shunt damping (both in single periodic and in bi-periodic) can reduce the amplitude magnification of the mistuned bladed disk effectively, and 99.7% of the statistical analysis results show that the amplitude magnification factor of bi-periodic eletromechanical coupling system is less than 1, which means the amplitude magnification of mistuned bladed disk can be eliminated almost completely. Bi-periodic distributed piezoelectric shunt circuit outperforms the single one in some proper design. What more, with the presence of the piezoelectric shunt damping, the threshold value of amplitude magnification factor nearly disappears, which means the mistuning robustness of the bladed disk has been improved obviously
High cycle fatigue life model of nickel-based single crystal superalloys based on critical plane approach
2017, 32(3): 677-682. doi: 10.13224/j.cnki.jasp.2017.03.020
Abstract:
The high cycle fatigue life predication of the 〈001〉 orientation of nickel-based single crystal superalloys was studied by foreign scholars with critical plane approach based on octahedral slip system currently. However, without considering the characteristics of slip system parameters when the 〈111〉 orientation was loaded, the predication of high cycle fatigue life of the 〈111〉 orientation was inaccurate. Therefore hexahedral and octahedral slip systems was considered comprehensively when choosing the critical plane. Slid planes of maximum fatigue parameter could determine the position of critical plane, and the life models of the SSR(shear stress rang), CCB(Chu-Conle-Bonnen) and Walls were used for high cycle fatigue life prediction of single crystal superalloys. According to the results of high cycle fatigue test under 800℃ along 〈001〉,〈011〉 and 〈111〉 orientations of DD6 nickel-based single crystal superalloys, the prediction accuracy of the life model was verified. The results indicate that the fitting coefficient of life model can reach to 0.9134 when the high cycle fatigue life of single crystal superalloys is predicted based on two slip systems.
A creep model based on normalized parameters
2017, 32(3): 683-688. doi: 10.13224/j.cnki.jasp.2017.03.021
Abstract:
In view of the deficiency that the traditional creep models, e.g. the Nortons law, cannot be used to simulate the third stage of creep, a creep model based on normalized time as a variable and the normalized stress and normalized temperature as parameters was presented. The developed model could describe all three stages of the creep curves. On the basis of the developed model, an improved model was put forward to describe the first stage of the creep curves better. Then feasibility of the developed model was validated by use of the experimental data of the direct aged GH4169G.The relationship between the parameters of the developed model and the dimensionless temperature,dimensionless stress was established. And the models can be used in finite element software, showing that the model can be applied to simulate the creep behavior of realistic structure.
Vibration characteristics of rotating thin-walled cylindrical shell: comparison of finite element method with analytical solution
2017, 32(3): 689-696. doi: 10.13224/j.cnki.jasp.2017.03.022
Abstract:
Analytical solution and finite element method were used to study the influence of Coriolis force and centrifugal stiffening effect on the vibration characteristics of rotating thin-walled cylinder structure under different constraints. Results show that the finite element method to calculate the natural frequency of the thin-walled cylinder structure, compared with the analytical solution, has deviation under different static boundary conditions. Results of the two methods fit well for simply supported boundary conditions. With the increase of rotating speed, travelling wave frequency curves of backward wave and forward wave were clearly separated, and the trend of separation decreased with the increase of circumference wave number. Because of the influence of centrifugal stiffening effect, compared with to the case of only considering the effect of Coriolis force, the travelling wave frequency of rotating thin-walled cylindrical shell will increase, and when the rotating speed and circumference wave number is larger, the forward and backward traveling wave frequencies are greater than the static natural frequency.
Implementation and validation of creep model based on normalized parameters
2017, 32(3): 697-703. doi: 10.13224/j.cnki.jasp.2017.03.023
Abstract:
In order to numerically simulate creep behaviors and stress relaxation effects of realistic structural components under operating conditions in service, a model based on normalized parameters, which can be used to describe all the features of creep deformation in three stages, was presented and coded into a users subroutine under the platform of a commercial finite element package by use of its user programmable features tools. The difference between the numerical results from the coded subroutine and the experimental data, and the effect of time step on numerical results, was investigated. The computers time consuming for solution of finite element models in different scale was compared and analyzed. For a loading case with stress and/or temperature changing, the time hardening theory and the strain hardening theory, as well as the relative time hardening theory presented with its result between those of them, were implemented in the coded subroutine. Numerical results of a plate with a hole show that the coded subroutine can be used to simulate the creep behaviors and effects of stress relaxation in realistic structural components, and that the feasibility and effectiveness of the coded subroutine are validated.
Variation analysis of sensitivity for satellite momentum wheel
2017, 32(3): 704-713. doi: 10.13224/j.cnki.jasp.2017.03.024
Abstract:
A numerical method for nonlinear sensitivity variation analysis of satellite momentum wheel was proposed based on the fuzzy theory, so as to evaluate the overall variation characteristics of stability system by changing the fuzzy similarity relation of engineering practice into the fuzzy equivalence relation of space vector. Through simulating the time data sequence of the uniform distribution, linear distribution and period distribution, the feasibility of model was verified; according to the actual steady-state running test for the three sets of satellite momentum wheels, this method was practical and effective. Among them, the minimum sensitivity coefficient of momentum wheel A was 0.678,greater than the 0.5 threshold, showing that its sensitivity was very good during the operation; however, the minimum sensitivity coefficient of momentum wheel B was 0.439, less than the 0.5 threshold, so its sensitivity produced a variation; and the minimum sensitivity coefficient of momentum wheel C was equal to the 0.5 threshold, showing the momentum wheel C had the most fuzzy relation between stability and variation. The variation process of sensitivity of the momentum wheel was forecasted and described in time. This model is also suitable for many nonlinear poor information problems in the field of space.
Nonlinear vibration of dual-rotor system with surface waviness in inter-shaft bearing
2017, 32(3): 714-721. doi: 10.13224/j.cnki.jasp.2017.03.025
Abstract:
Considering the effect of dual-rotor model with surface waviness in inter-shaft bearing, the motion equations of a dual-rotor system were formulated by employing rotor dynamics and Langrange equations. The nonlinear vibration responses of the system were derived through numerical calculations. Accordingly, the dynamic behaviors of the system influenced respectively by rotation speed, maximum surface waviness amplitude, surface waviness order and the initial surface waviness amplitude were analyzed. The results showed that the effects of the surface waviness on the high pressure rotor and the low pressure rotor were almost the same. With the increase of the rotation speed, both of high pressure and low pressure rotors showed an alternative variation of periodic motion and non-periodic motion. In particular, the quasi-periodic and even chaotic motions may occur within a relatively low rotation speed region. With the increase of maximum surface waviness amplitude, the dynamic behavior of the system may turn from periodic motion into periodic 2 motion, quasi-periodic motion and periodic 4 motion against the safety and stability of the vibration system. When the surface waviness order was integer multiples of the number of rollers, the motion of the system would might be non-periodic, e.g. quasi-periodic or chaotic. With the increase of the initial surface waviness amplitude, the system showed an alternative variation of quasi-periodic motion and chaotic motion. Nevertheless, for a larger initial surface waviness amplitude, the vibration amplitude of system was even smaller.
Optimization design method of bearing structure parameters based on Kriging model
2017, 32(3): 723-729. doi: 10.13224/j.cnki.jasp.2017.03.026
Abstract:
There is a close relationship between heat generation rate and the life of the bearing. To decrease the heat generation rate of the bearing, associated parameters of the working bearing were obtained based on pseudo statics analysis and raceway control theory of the bearing, and then, the heat generation rate of the bearing was further acquired. The Kriging model and particle swarm optimization algorithm were proposed for optimizing structure parameters of the bearing to obtain the minimum heat generation rate of the bearing. Taking the NSK-7016A5 bearings as an example, the results show that the total heat generation rate of the bearing and inner and outer rings can be decreased after optimization of bearing structure parameters. The combination method of Kriging model and particle swarm optimization for optimizing bearing structure parameters could obtain satisfactory design results, and increase bearing design efficiency as well.
Analysis on dynamic characteristics of high speed angular contact ball bearing based on quasi-dynamic
2017, 32(3): 730-739. doi: 10.13224/j.cnki.jasp.2017.03.027
Abstract:
In order to reflect the real operation of high speed angular contact ball bearings, a quasi-dynamic calculation model for high speed angular contact ball bearing was built. BFGS(Broyden-Fletcher-Goldfarb-Shanno) mutative scale optimization was embedded in quasi-dynamic calculation program for high speed angular contact ball bearing and then the calculation model proposed was validated as compared with the calculation example in quasi-dynamic calculation program AT74Y001 developed by SKF Corporation, and error between the two calculation methods was within 5%. Quasi-dynamic equations were solved to analyze the dynamic performance of high speed angular contact ball bearing under different structural parameters and operating conditions. Results showed that, with the increase of axial load, the contact angle of inner and outer rings increased; with the increase of radial load, the inner contact angle decreased and the outer contact angle increased in the bearing area, while in the non bearing area ,the inner contact angle increased and the outer contact angle decreased; with the increase of rotational speed, the inner contact angle increased, and the outer contact angle decreased; with the increase of the number of steel balls and axial load, the bearing stiffness increased in three directions; with the increase of curvature of inner ring groove of steel balls and radial load and rotational speed, the bearing stiffness decreased in three directions.
Modeling simulation and optimization of oil system based on MATLAB/Simulink
2017, 32(3): 740-748. doi: 10.13224/j.cnki.jasp.2017.03.028
Abstract:
In order to analyze the oil systems characteristics and study the coperative working and design problems involving aero-engine, a general simulation block library for its components was developed in MATLAB/Simulink based on flow rate, pressure and heat transfer characteristics. Then a full-system simulation model was established, including supply, return, air and thermal system. Taking a certain turbofan engine and its oil system as example, the main working parameters of the oil system were calculated under high temperature taking off condition. Compared with the calculated values of Flowmaster model, the relative error was less than 3.8%. Through the extension ability of the full-system simulation model, a co-simulation with GasTurb engine model was successfully implemented and the oil systems design parameters were optimized through genetic algorithm. The highest temperature of return oil decreased by 4.12% by changing the oil supply ratio. This method can provided a reference for oil system design and improvement.
Method of compressor characteristic extension combining exponent extrapolation method with support vector machine
2017, 32(3): 749-755. doi: 10.13224/j.cnki.jasp.2017.03.029
Abstract:
Considering the poor accuracy computation of the compressor lower speed characteristic in aero-engine startup modeling, a method of compressor characteristic extension combining exponent extrapolation method with support vector machine (SVM) was proposed. Firstly, the mechanism and limitation of exponent extrapolation method were analyzed. Then the known data of compressor higher speed characteristics were used as SVM training set, and the data of lower speed characteristic obtained by exponent extrapolation method were used as the test set. Meanwhile, in order to reduce the nonlinearity of the original data in SVM, the compressor characteristic was converted into the characteristic represented by outlet flow parameters. The lower speed characteristics were predicted and obtained by using a kind of cross validation algorithm to select the parameters of SVM and train the SVM model. The comparative analysis with the characteristics obtained by using only exponent extrapolation method shows that using the method of compressor characteristic extension combining with exponent extrapolation method with SVM, the maximum relative error is reduced by about 2.8%, and the characteristic extension precision is effectively improved.
Optimization of anode propellant allocation on xenon ion thruster
2017, 32(3): 756-761. doi: 10.13224/j.cnki.jasp.2017.03.030
Abstract:
Approach of micro-scale similar flow field of inert gas was obtained with two numerical methods. It was verified that the flow characteristics of ion thrusters propellant distributor could be better simulated under first-order velocity slip boundary condition by N-S(Navier-Stokes) method. Based on xenon flow features and discharge characteristics of four-ring-cusp discharge chamber, improvement scheme was proposed for decelerating injection velocity and ameliorating circumferential uniformity of propellant. Dual-stage distributor was utilized and extra amounts of orifices were arranged at 45° on both sides. Calculation results demonstrate that after conducting the improvement scheme, circumferential uniformity of xenon has an amelioration of 37% while the speed of flow has a decrease of 32%. Furthermore, experimental results illustrate that ion production cost has dropped from 183W/A to 167W/A at maximum throttle condition, and performance of discharge chamber has improved.
Study on subcooled technology for cryogenic propellants
2017, 32(3): 762-768. doi: 10.13224/j.cnki.jasp.2017.03.031
Abstract:
Three subcooled methods,including the subcooled technology of heat transfer, pumping down decompression and injecting cooled helium for cryogenic propellants were summarized. By comparing advantages and disadvantages of three methods, the subcooled technology of pumping down decompression was recommended to be used on the entire loading process of ground. Based on thermodynamics theory, expressions of consumption, refrigeration capacity, pumping down time and lowest pumping speed were derived for the subcooled technology of pumping down decompression. The researches showed that consumption of cryogenic propellants was mainly used for its temperature drop, then the proportion for offsetting external leakage heat and special heat of tank material was very small, for instance, the relative consumption of its temperature drop, special heat of tank material, and external leakage heat for liquid hydrogen was 10.94%,0.38%,0.098%, respectively. The arithmetic accurately calculating consumption of cryogenic propellants was recommended to reduce launch costs of launch vehicle and enhance utilization of cryogenic propellants. Compared with the exiting formula, its relative error was 18%.