In order to meet the urgent demand of variable sweep aircraft in wide speed range, a kind of driving mechanism suitable for distributed driving of locally rotating variable sweep wing was designed. Taking the transmission performance as the index, the dimension of the unit driving mechanism was preliminarily designed, and then the SQP (sequence quadratic program) algorithm was combined to optimize the dimension of the mechanism with the energy conversion rate of the whole process of mechanism deformation as the optimization objective. The output work of the optimized driving mechanism under constant force was increased by 44.3%, the energy conversion rate was increased by 37.5%, and the driving distance was shortened by 9.7%. To solve the problem of how to distribute the driving force of multiple driving chains, a quasi-static driving force solution method was proposed by combining the force method of analyzing the internal force of statically indeterminate structure with the traditional force analysis method of mechanism. The driving force of the four wing beam models under certain load and component material conditions was calculated, and the dynamic simulation experiment was carried out based on ADAMS (automatic dynamic analysis of mechanical system) to verify the accuracy of the driving force distribution model. The error analysis showed that the calculation error of the quasi-static driving force of the model was less than 5.5%. Finally, based on the simulation results and the weight of the driving mechanism, the number of driving chains was optimized, and the optimal number of driving chains was determined to be 3.

Because of the insufficiency of numerical simulation and test verification of high temperature non-equilibrium flow status, the high temperature flow field ground test simulation technology and display technology in high enthalpy shock wave wind tunnel were developed respectively. The ground tests of high temperature non-equilibrium flows were carried out in FD-21 high enthalpy shock wave wind tunnel and FD-20 normal shock wave wind tunnel, then the spatial flow field structure and the aerodynamic heating results of the test hemisphere were obtained. Meanwhile, the numerical simulation technology of high altitude and hypersonic thermochemical non-equilibrium flow was established for the supersonic flight environment, and the reliability of the calculation method was verified by the ground test. The test models were of a hemisphere structure with radius of 20 mm and 60 mm. The calculation results showed that: (1) The inlet flow parameters in the wind tunnel were verified by numerical simulation and the pressure and species can be taken as input parameters for the below numerical simulation. (2) Numerical simulations were carried out on the above typical hemispheres, and the test and calculation results of the hemisphere heat flux were compared and analyzed. The results showed that the two-temperature model (thermal chemical non-equilibrium model-2T) was in good agreement with the test. Under the condition of non-catalytic wall, the relative error between the calculated heat flux result of 2T model and the test result was 11.6% when the total temperature was 2700 K. The relative error was 17.5% when the total temperature was 4050 K. (3) The test schlieren (hemisphere shock wave off distance) obtained from the calculated results was in good agreement with each other. Under the condition of non-catalytic wall, the relative error between the shock wave off distance and schlieren of 2T model was −1.9%−0.86%.

The magnitude orders of the terms of the linearized perturbation equations (LPEs) in body-fitted three-dimensional curvilinear orthogonal coordinates were analyzed. Results showed that most of the curvature-related terms were represented by high-order small quantities. The LPEs in body-fitted curvilinear orthogonal coordinates were greatly simplified by neglecting the high-order small quantities, which resulted in the simplified LPEs. The tedious derivation of the fully LPEs in curvilinear orthogonal coordinates was no longer needed. The simplified LPEs can be easily obtained by extending the Cartesian LPEs to body-fitted curvilinear orthogonal coordinates for the convenience of beginners. The simplified linear stability theory (LST) suitable for boundary layers with surface curvatures was developed based on the simplified LPEs. The accuracy of the simplified LST was validated in the boundary layers on the curved surfaces of a subsonic sweep wing and a hypersonic sharp-nose cone. For the former, the growth rate of crossflow instability predicted by the simplified LST deviated from the result predicted by the full form LST less than 4×10⁻⁵; for the later, the simplified LST accurately predicted the effect of circumferential curvature on the neutral line of the Mack mode. Numerical results showed that the simplified LST had almost the same accuracy with the full form LST.

In order to establish a thermodynamic model of aero-engine in an approximate real operating environment, a thermodynamic model construction method based on QAR (quick access recorder) data of aero-engine operation was proposed. According to the thermodynamic equation of the traditional design point of the engine, and using the least square identification principle, the equation for solving the pressure at the 25 station was corrected, and the dry air thermodynamic model was obtained; according to the additivity of the entropy value of the mixed gas, a thermodynamic model of wet air under rainfall conditions was constructed; finally, combined with particle swarm optimization optimized by genetic algorithm, the parameters of dry air and wet air thermodynamic model were calculated and verified according to QAR data. The results showed that the maximum error between the dry air and wet air thermodynamic parameters calculated by the thermodynamic model and the QAR data was less than 13%, which was closer to the actual adiabatic parameters. The feasibility of constructing a thermodynamic model of an engine based on QAR data and the effectiveness of the thermodynamic model construction method for solving thermodynamic parameters were proved.

In order to carry out aerodynamic/flight/control integration research at the early stage of design, shorten the development cycle and reduce the development cost, a model free-flight test system was established in FL-10 open wind tunnel. The system was used to design the scale verification machine model according to the similarity criteria of size, mass and moment of inertia, invent the flight control system, on this basis, the system was modeled and simulated, and the closed-loop stability increasing control law was designed to realize the stable flight at the trim angle of attack. The test results showed that the structure and parameters of the flight control law were designed according to the similarity criterion, the simulated trim angle of attack was 5°, the trim elevator deviation was −4.5°, and the wind tunnel test angle of attack was controlled within (5±0.2)°. Because the power supply and signal cables gave the test model a positive pitch moment, the elevator deviation was controlled at about −3.2°, which was slightly less than the simulated trim elevator deviation.

The effect of nacelle strake on stall was studied by numerical simulation. Numerical simulation and wind tunnel test showed that in the state of zero thrust, after a large four engine propeller aircraft exceeded the stall angle of attack, the wings between the internal and external engines first separated and quickly advanced to the leading edge of the wing, and the lift loss reached about 30%. In order to limit the separation speed of the wing, a nacelle strake was installed inside the outboard engine nacelle. The simulation results showed that the stall characteristics were significantly improved at the optimal design position, the lift loss after stall was reduced by about 50% compared with the state without nacelle strake, and the stall angle of attack did not change significantly. Different installation positions of nacelle strakes had obvious differences in improving stall; movement from the best position to the circumferential direction and propeller direction caused the failure of nacelle strake. Movement to the wing direction wa effective to improve the wing separation speed, but it significantly reduced the stall angle of attack. After moving to the wing direction, the influence of circumferential position was reduced. A wind tunnel test was carried out to verify the landing configuration in zero-thrust condition according to the optimal design position. After adding the nacelle strake, the lift loss of the airfoil after stall was reduced from 0.92 to about 0.42, and the lift loss was reduced by 54%, which was basically consistent with the numerical simulation conclusion.

To improve the simulation accuracy of transient gas turbine overall performance, a refined secondary air system (SAS) model was established and perfected in a modular fashion. A coupled simulation method for transient performance and secondary air system was further proposed. A two-spool gas turbine was modeled to analyze the dynamic variations of the SAS bleeding and returning as well as their effects on the overall engine performance parameters. Results showed that the effect of engine operating conditions on mass fractions of cooling air can be evaluated by coupled simulation method. A variation of 0.18% in the mass fractions of cooling air was observed during the sudden load change process. By comparison, this value increased to 0.55% for the gas turbine with a sudden load shedding. The mutual interactions between the main flow path and the SAS had more effect on the imbalanced local response of the SAS than the overall engine performance parameters. This phenomenon cannot be neglected in the elaborate simulation of modern gas turbines.

The influences of different numbers of oil supply holes, supply pressure and piston ring gap angle on leakage were considered by orthogonal excitation method, and the damping coefficients of squeeze film damper (SFD) were identified. The results showed that the maximum leakage occurred when single oil supply hole existed and the piston ring gap coincided with the oil supply hole; under multiple influences of the number of oil supply holes, supply pressure and piston ring gap angle, the difference between the principal damping coefficients C_xx and C_yy was minimum when the piston ring gap angle with low oil pressure and single oil supply hole was 0°; the damping performance of SFD was excellent when the piston ring gap angle was 0° and the lowest single oil supply hole existed, and the average principal damping coefficient amounted to 9.04×10⁴ N·s/m, which was about 5.7 times of the maximum value of theoretical short bearing solution. It is recommended that the two piston ring gap angles should be all 0° during installation to ensure well sealing performance and improve the performance of SFD during operation.

Combined with theoretical simulation of the beat vibration, analysis of the measured vibration signal and summary of the actual vibration elimination experience, three multi-source beat vibration modes of aero-engines were discussed. The identification process and the elimination method of the beat vibration fault were established. Experimental verification showed that for the three common multi-source coupling beat modes, the multi-source beat vibration and the vibration fluctuation could be eliminated when the multi-source excitation frequency differed by more than 3%. The proposed process and method were implemented and verified in engine beat vibration and fluctuation trouble-shooting, and the low-pressure rotor 2 times frequency-high-pressure rotor fundamental frequency coupled beat vibration was accurately identified. The low-pressure rotor speed was well-tuned through the control schedule to separate the above two frequencies by 3%. The vibration fluctuations caused by beat vibration was eliminated and the proposed method was validated.

The fluid-solid-thermal multi-physics coupling theory of graphite seal was analyzed, and a numerical solution model of graphite seal fluid-solid-thermal multi-physics coupling considering the deformation of graphite ring was established. On the basis of verifying the accuracy of the model, carbon graphite and antimony-impregnated graphite materials were studied. The flow field characteristics, the structural mechanical characteristics and leakage characteristics of the graphite seal at different pressure ratios and temperatures were compared and analyzed. The leakage of the graphite ring before and after the deformation was compared and analyzed, and based on the traditional circumferential graphite seal leakage formula, a theoretical formula for graphite seal leakage considering the deformation of the graphite ring was constructed. Results showed that the established graphite seal fluid-solid-thermal multi-physics coupling model considered the deformation of the graphite ring, and can accurately calculate the leakage flow characteristics and mechanical characteristics. The fluid entered the shallow groove under the action of the circumferential shear flow, and formed a local high pressure area by extrusion in the shallow groove. The deformation of the graphite ring on the high pressure side was obvious, of which the carbon-graphite material deformed significantly under the action of high pressure ratio, and the antimony impregnated material deformed significantly under the action of high temperature. When the temperature was 400 K and the pressure ratio was 2—4, the deformation of the carbon-graphite sealing ring was reduced by 12.99% on average compared with the antimony-impregnated material, and the seal leakage was reduced by 6.89% compared with the antimony-impregnated graphite material. The constructed leakage theoretical formula can accurately calculate the leakage considering the deformation of the graphite ring, thus providing a theoretical basis for the analysis of the leakage characteristics of the graphite seal.

A refined method for determining the basis values of tensile failure strain for ZT7H/5429 composite laminates in hygrothermal environment was proposed. Through the tension tests of the composite laminates with open hole under different environmental conditions, the influence of temperature and humidity on the tensile failure strain was quantitatively analyzed, and a tensile failure strain prediction model was established based on small sample holistic inference technology. The results showed that the tensile failure strain decreased with the increase of temperature, and the decline became flat gradually, while the influence of humidity was not so significant. Based on the tensile failure strain prediction surface, the B-basis value of tensile failure strain was further determined by introducing one-sided tolerance-limit factor. Compared with the traditional single-point method of only analyzing the test results in the single state, the proposed method fully took into account the connection between the experimental data in different states and the B-basis values increased by 2% to 25% in most cases, providing a theoretical basis for the refined design of composite structures.

Based on the actual service environment of metal rubber vibration absorber, the transverse and axial quasi-static loading fixtures of vibration absorbers were designed. The corresponding quasi-static mechanical properties of 175 metal shock absorbers, which can be divided into 5 kinds of types, were successfully obtained. Furthermore, the experimental data were fitted with high order polynomial on the basis of commercial software. The average values and performance boundaries of transverse and axial load-displacement curves for the 5 kinds of metal rubber vibration absorbers were successfully determined. The results showed that the dispersions of transverse and axial load-displacement curves of vi-bration absorbers did not obey the common law of Gaussian probability distribution, making difficult to be taken into account at the same time. The dispersion of MR2 was the minimum and MR3 was the maximum under axial loading, but the dispersion of MR3 was the minimum and MR2 the maximum under transverse loading. The corresponding per-formance boundaries can provide important support for the model selection and safety assessment of metal rubber vibra-tion absorber. According to the three-dimensional computed tomography analytical device of μCT, the three-dimensional morphology of metal wire component contained in the metal rubber vibration absorber was established. Furthermore, the volume of metal wire in different height regions were determined successfully.

The development and present situation of aeroengine fan blade structure, forming technology abroad were introduced in details. The research status of titanium alloy wide chord hollow fan blade in China was analyzed from the aspects of structure design, forming process and mechanical properties. Based on the existing problems in the service process of wide-chord hollow fan blades, combining with the diffusion welding/conventional plastic/superplastic forming technology, the development direction of key technologies in the three fields of blade cavity structure design, forming process optimization and mechanical properties of multilayer diffusion welding structure were discussed. The cavity structure design should be based on three objectives: lightweight structure, feasibility of forming and optimal mechanical properties; The optimization of forming process focuses on the coupling optimization of multi-process and multi-objective, the improvement of numerical calculation accuracy, and the precise prediction and control of material shape evolution during cold and hot composite machining; The improvement of mechanical properties of diffusion welded structures should be studied from the aspects of bonding joint crack propagation and fatigue behavior evolution under complex load conditions.

In view of the dual-rotor system with two sets of vibration modes of low-pressure rotor excitation and high-pressure rotor excitation, the orthogonality of the two vibration modes was affected by the speed ratio, so N₁+N₂ plane and N₁+N₂+4 plane mode dynamic balance method of the dual-rotor system was proposed by replacing the rotor mode under the actual operating speed line with the rotor mode under the constant speed ratio. On the basis of the known modal orthogonality theory of the dual-rotor system, the equilibrium conditions of the two-rotor system N₁+N₂ and N₁+N₂+4 plane mode dynamic balance methods were deduced, and two dynamic balance methods of the calculation equations of the unbalanced mass and the orthogonal correction mass group were given. What’s more, taking a certain type of aero-engine dual-rotor system as an example, the process of balancing each order mode was given by the N₁+N₂ mode dynamic balancing method. Finally, the dynamic balance verification experiment was carried out on the dual-rotor test system. The experimental study found that the vibration reduction effect of dynamic balance method can reach up to 72.4% under the condition of constant speed ratio; when the high and low pressure rotor speed control rate was the actual working speed line in the working speed range, the total vibration amplitude at each critical speed was reduced to below 120 μm, meeting the design requirements. The experimental results showed that the proposed mode dynamic balancing method for the dual-rotor aero-engine is feasible.

In order to curb the carbon and NOx emissions of the aeroengine and ground gas turbines, a multi-injector combustor dome configuration was proposed. To explore the micromix non-premix combustion mechanism and the impacts of the critical parameters on the combustor performance, the k-ω SST (shear stress transfer) turbulence model, FGM (flamelet generated manifold) diffusion flamelet method and 9 species-26 steps reactions were adopted for the numerical simulation. Sensitivity analysis was performed with regard to the momentum flux ratio, equivalence ration, air-gate height and hydrogen-gate height. The influencing factors of mixing characteristics, flame structure, temperature distribution and NOx emission were explored. Meanwhile, the NOx emission between the conventional kerosene combustor and the hydrogen micromix was evaluated. Results showed that lower momentum flux ratio was beneficial for enhancing the mixing and reducing the flame length, resulting in 169.6% NOx reduction. There existed a critical value of the air gate height which brought about the highest NOx up to 5×10⁻⁶ under the 15% O₂ content condition. As H₂ gate height increased from 3 mm to 11 mm, NOx fell by 75.9%. Compared with the conventional kerosene combustor, NOx emission dramatically dropped by 85.7%, using the optimal combination of critical parameters.

Design and experiment study of a grand scale diameter and small annular cavity combustor were carried out for the first time interiorly, considering the turbine engine with innovated structure. A combustor design scheme based on annular vortex field matching tangential fuel injection under the condition of diffuser with reverse air intake was proposed. Cold and hot performance experiments of combustor under different conditions were carried out, the combustor characteristics including flow resistance characteristics, ignition characteristics at ambient temperature and pressure, lean blow out characteristics, combustion efficiency, as well as the outlet temperature field, were obtained. The results showed that: （1） the design scheme met the design requirements of combustor; （2） compared with conventional combustor, the nozzle spacing ratio of the scheme reached 1.65; （3） the ignition performance of the designed combustor was excellent, as its minimum lean ignition fuel-air-ratio reached 0.016 and its ignition and flame combination time was within 4 s; （4） the total pressure loss, combustion efficiency, out temperature distribution factor （OTDF） and radial temperature distribution factor （RTDF） of the designed combustor had outstanding performance; as the combustion efficiency of idle state reached 98.6%, the OTDF of design point reached 0.16 as well.

In order to verify the applicability of the classical heat transfer correlations mostly based on the experimental data of large tube diameter on heat transfer of small-scale tube bundle, the flow and heat transfer characteristics outside the staggered tube bundle in cross flow were numerically investigated. With a new method to determine the reasonable mesh density of variable geometry, the effects of tube diameter, tube pitch ratio and Reynolds number on Nusselt number and flow characteristics outside the tube bundle were studied within a large parameter range. The following conclusions can be obtained: （1） when flow Mach number was smaller than 0.1 and Reynolds number kept constant, the tube diameter had little effect on the Nusselt number because the flow characteristics outside the tube varied slightly with tube diameter; （2） the tube pitch ratio affected Nusselt number by changing the flow characteristics outside the tube, and there was a peak area of Nusselt number within a specific range of tube pitch ratio when the Reynolds number and tube diameter kept constant; （3） a more specific piecewise function was needed to fit the heat transfer correlation to accurately predict the Nusselt number.

In order to investigate the influence of the flow and heat transfer characteristics on vortex cooling with different finned structures in the intake cavity and vortex cooling chamber, six different models with finned structure were built. The flow characteristics, the heat transfer characteristics, and the comprehensive heat transfer performance of these six models were compared and analyzed under the same boundary conditions. The results showed that the heat transfer intensity of the target in the intake cavity increased obviously when finned structure was added to the intake cavity. When finned structure was added to the vortex chamber target surface, the heat transfer intensity and comprehensive heat transfer factor of the target in the vortex cooling chamber were enhanced a lot, too. Among these six models, the model with 45° oblique fins on the intake cavity and 90° fins on the vortex cooling chamber had the highest total heat transfer and comprehensive heat transfer factor, the total heat transfer of the two targets was 51.8% higher than that of the non-finned model, the comprehensive heat transfer factor was 3.44% higher than that of the non-finned model.

In view of the problems of high missed detection rate and inability to separate adhering bubbles in the extraction of icing microscopic images by traditional image segmentation methods, a method combining deep neural network and traditional segmentation algorithm was proposed. Based on the attention U-Net network, a two-branch fusion prediction strategy was adopted to extract the bubbles in the icing microscopic images. For some bubble adhesion problems, histogram equalization and local minima were introduced, and a watershed algorithm based on distance transformation was used to segment the adhesion bubbles in icing microscopic images twice. The experimental results showed that the two-branch prediction prediction Attention U-Net network was more accurate for the extraction of bubbles in different icing microscopic images, especially the detection rate for smaller bubbles was higher. The pixel accuracy, mean pixel accuracy, mean inter-section over union and frequency weighted intersection over union of the test images reached 0.9767, 0.8916, 0.8188 and 0.9575, respectively. The watershed algorithm based on distance transformation also showed good performance in the segmentation of sticky bubbles, providing quantifiable data support for the subsequent statistics of the number and area of bubbles.

In response to multi-objective and multi-criteria performance evaluation problem for aero piston engine, the analytic hierarchy process (AHP) and entropy weight method (EWM) were used to transform it into single-objective and multi-level evaluation, and establish performance indicators, weight allocation, and evaluation systems under different operating conditions. Genetic algorithm-back propagation (GA-BP) neural network was employed to calculate and estimate its performance degradation state, the correctness of the performance evaluation system was verified by the joint simulation experiment. For the abnormal fuel injection holes leading to abnormal engine performance degradation as an example, this study explored the mechanism of its performance degradation from the perspective of combustion. The results showed that, when there were abnormal components of engine, the hierarchical entropy weight performance evaluation was a better method to reflect current performance status of the engine and make accurate judgments on its safety. The GA-BP computational model had a high accuracy, whose mean absolute percent error (MAPE) decreased by 3.5208%, 0.7027% and 3.7854%, respectively, compared with BP, radial basis function (RBF) and Elman models.

A two-stroke aviation piston engine was taken as the research object. By establishing one-/three-dimensional engine model, injector model and air-fuel ratio control model, the main influencing parameters of intake and injection were discriminated. Based on the recurrent neural network intake prediction and injection model, the air-fuel ratio control of two-stroke engine was studied under variable working conditions. Under the transient condition of throttle opening change, the overshoot of air-fuel ratio control was controlled within 4.6%, which can restore the cylinder mixture to the equivalent ratio within 0.3 s after the variable condition stopped. Under the simulation study of different altitude conditions, it was found that with the increase of altitude, the overshoot and callback time of air-fuel ratio control model appropriately decreased and gradually stabilized.

An aero-engine component level model considering outlet temperature distribution factor (OTDF) and nitrogen oxides (NOx) emission characteristics was established, which provided a simulation platform for the study of engine combustor outlet temperature distribution and emission control. Taking a variable cycle engine as the research object, the three-dimensional combustor model was designed according to its design point parameters. Based on CFD numerical simulation method, the OTDF characteristics and NOx emission characteristics of the three-dimensional combustor model under different working conditions at sea level were obtained. Based on this, a component level model of variable cycle engine was established, which can calculate the characteristic parameters of combustor under full envelope and full state. Compared with the traditional component level model, the model can accurately calculate the combustor outlet temperature distribution and NOx emission under different working conditions and different envelope points. The simulation results showed that there was a negative correlation between the temperature distribution coefficient at the outlet of the combustor and the working state of the engine. The higher engine speed indicated, the smaller OTDF and the better temperature distribution quality at the outlet of the combustor. There was a positive correlation between the NOx emission at the outlet of the combustor and the working state of the engine. The higher engine speed indicated more NOx emission at the outlet of the combustor, which was in line with the basic law of engine combustion.

In order to establish an aero-engine thermal performance model with high confidence, precisely master the engine performance changes and accurately predict performance of the aero-engine at off-design conditions, a new correction method for the aero-engine mathematical model was proposed. On the basis of the translation and scaling of the component characteristic line, the rotation degree of freedom was considered to achieve the purpose of rotation adjustment of the compressor characteristic line. Nonlinear scaling factor function and rotation factor function were introduced to realize adaptive adjustment of compressor characteristic data under different operating conditions. Based on a certain turboshaft engine, the simulation verification of model correction method was carried out. The simulation results showed that, compared with the model correction method only considering translation and scaling two degrees of freedom, the proposed three-degree-of-freedom correction method considering translation, scaling and rotation can reduce the model steady-state average error from 0.901% to 0.344%, the model dynamic average error under high-power changed from 1.295% to 0.889%, and the overall correction effect of model steady state and dynamics was improved by 62.99% and 31.31% , respectively, which can meet the requirements of higher aero-engine model accuracy.

In view of the phenomenon of acoustic correlation frequency of liquid film appearing in hot-fire test of a hydrogen-oxygen preburner with swirl injectors in literature, the change process of average temperature of liquid oxygen film in injector during 30 s hot-fire test was calculated by transient thermal simulation of multiphase flow. In order to reduce the consumption of computational resources and time, multiphase model with Fluent was used to calculate the steady isothermal flow field, then, a two dimensional axisymmetric heat transfer model was used to call the velocity field and phase distribution results of Fluent to simulate the transient temperature. The results of heat transfer calculation showed that the gas temperature at the injector outlet had a significant effect on the liquid oxygen film temperature. The average temperature of liquid oxygen increased by about 2 K when the gas temperature increased by 50 K. The average temperature of liquid oxygen obtained by simulation was in good agreement with the empirical correction value in the literature when the injector outlet temperature was set to 105 K, which verified the reasonable temperature correction method. The calculation results can explain the phenomenon that the acoustic frequency of injector decreased duo to the effect of solid heat capacity on the heating of liquid oxygen.

In order to study the combustion process and aluminum agglomeration characteristics of NEPE propellant, the combustion process of NEPE propellant was recorded by the optical imaging method combined with a high-speed camera and a long distance microscope, the combustion wave flame was recorded by an infrared thermal imager, and the condensed combustion products of the propellant were collected to analyze by scanning electron microscope (SEM) and laser particle size analyzer. Results showed that the splash phenomenon of aluminum agglomerates existed in five combustion stages: initial ignition, flame diffusion, steady combustion, flame attenuation and flame extinction. At the pressure range of 0.5—2.0 MPa, with the increase of environmental pressure, the high temperature combustion region increased, the number of aluminum agglomeration at the burning surface increased, and the particle size of aluminum agglomeration gradually decreased. The condensed combustion products were mainly divided into aluminum agglomerates and oxide particles. The particle size of aluminum agglomerates was within the range of 50—600 μm, and the particle size of alumina particles usually less than 1 μm.

In view of the effective range of film cooling in the thrust chamber of liquid propellant rocket engine, based on the model thrust chamber, the effective length data of liquid and gas films, the temperature rise of regenerative coolant and the temperature of outside wall were obtained by test. Heating length and evaporation length of liquid film, distribution of gas film cooling efficiency, regenerative coolant temperature rise and outside wall temperature were obtained by simulation. The temperature rise of the regenerative coolant was 90.1 K, and the outside wall temperature was about 340 K by test. The regenerative coolant temperature rise and outside wall temperature were 89.7 K and about 330 K separately calculated by simulation. The effective length of liquid film was 74.67 mm and that of gas film was 124.2 mm by test. The effective length of the liquid film was 75.0 mm by simulation, in which the heating length was 66.0 mm and the evaporation length was 9.0 mm. The standard for judging effective cooling efficiency of gas film obtained after calibration was 0.687. By comparing the test results and simulation results, the applicability and accuracy of the simulation program were verified, and the critical value of gas film cooling efficiency was obtained, which can be used as a reference for thermal protection design of thrust chamber.

In view of the requirements of rotor pressure sensitive paint measurement technology, an image enhancement acquisition method with short exposure for rotor based on a charge-coupled device (CCD) camera was proposed, and a corresponding measurement system was established. Two methods (short-exposure image enhancement acquisition and transient image acquisition) of 9 exposure cases were implemented on the rotor blades of a small-scale fan at 2700 r/min. The experimental results showed that the rotor enhanced acquisition method with short exposure, and the measurement system , can effectively improve the rotor image quality; “no motion blur” image can be achieved when the blur length of the rotating image obtained by using the image enhancement acquisition method with short exposure of the rotor was less than 4 pixels; the brightness of the rotating state image was equivalent to that of the static image when the number of acquisitions equalled to the ratio of static image exposure time to rotating state image exposure time.