Journal Of Applied Fluid Mechanics
Volume:15 Issue: 3, May-Jun 2022

The flow field in a two-dimensional hypersonic mixed-compression inlet in a freestream Mach numbers of M∞ =2.0, 3.0, and 5.0 are numerically solved to understand the effect of throat area variation. The exit area ratio variation is simulated by placing a plug insert at different axial locations at the exit of the model. The flow field is achieved computationally by solving the Reynolds Averaged Navier-Stokes equations in a finite volume framework. For each flow condition, the variation in shock structure is analyzed and the variation of the oblique shock wave angle with the mass flow rate is calculated theoretically and compared with the present CFD analysis. The variation in oblique shock angle is calculated in terms of the mass flow rate by considering the capture area and spillage flow through the inlet. The theoretical results suggest that the method can predict the inlet operating conditions at different freestream Mach numbers and area ratios. This method can quantify the reduction in mass flow rate due to the throttling effect by analyzing the flow field shock pattern. The effects of various important performance parameters such as free stream Mach number, total pressure recovery, and mass flow ratio were then numerically investigated. As the Mach number is increased, the total pressure recovery is reduced, but the maximum value of the mass flow rate is increased. The analysis is also focused on the effect of throat area variation on performance parameters at each Mach number. The characteristic curve of the inlet is then obtained for each free stream Mach number.

Casing treatment is a powerful method for improving the stability of aircraft compressors. An optimized slot-type casing treatment was tested on the first rotor of a highly-loaded two-stage compressor, and the results showed that the casing treatment could not increase the compressor stability at the design and off-design speeds. A coupled casing treatment (CCT), which is built with an injector, a bridge, a plenum chamber, and several slots for a recirculating loop, is proposed and optimized to enhance the stability of the compressor in the present study. The optimized CCT improves the compressor stability and efficiency under the design condition by 75.8% and 0.71%, respectively. The coupling effect, which is established with an inner circulation in the slots and an outer circulation from the slots to the injectors, accounts for the excellent stability enhancement. The coupling effect reduces the amount of tip leakage flow, depresses the development of the tip leakage vortex (TLV), and greatly decreases the blockage in the rotor tip which is primarily induced by the interaction of the shock-wave and boundary-layer at the blade suction surface. The parametric study shows that improving the coupling effect has a positive effect on reducing the rotor tip blockage, but a negative effect on the stability of the compressor stage. This is because the inflow condition of the stator is tremendously distorted while the coupling effect is excessively strong, which can cause a stall in the stator rather than in the rotor. The compressor stability can be maximally enhanced by adjusting the strength of the coupling effect to make a compromise of the improved rotor tip flow and the deteriorated stator flow.

In this study spread of smoke from a possible fire in a University building and the evacuation time of occupants were simulated. Fire dynamic simulations (FDS) have been done for natural and forced smoke evacuation with different scenarios; at the same time, evacuation simulations have also been done for various scenarios for different exits at the building. While occupants move through changing CO, CO2, and O2 concentrations, Fractional Effective Dose (FED) was gathered to obtain results from both simulations. FED results were evaluated for poisoning risk of occupants. According to comparative results, the combination of scenarios that forced smoke evacuation by fan and evacuation of occupants from all exits at the basement of the building has the lowest FED value. On the other hand, depending on the fire source and smoke movement, sometimes occupants cannot use all exits. Therefore, evacuation simulation has been done separately from each exit and evaluated with all FDS results.

Accurate calculation of pressure distribution in the impeller cover side cavity is the key to predict the axial force of centrifugal pump. The existing calculation models almost does not involve the prediction of cavity pressure when the radial clearances of different sealing rings are matched with the diameter of different balance holes. On the basis of the original prediction model of pump cavity pressure, a mathematical model of pressure distribution of impeller cover side cavity with different radial clearance of sealing ring and the diameter of balance hole was established by introducing potential head correction coefficient and flow proportional coefficient. In order to improve the calculation accuracy of rotation coefficient for rear pump cavity, the balance aperture length ratio and the rotation undetermined coefficient were introduced in the calculation equation of original rotation coefficient. A test bed for pressure and leakage was designed and established, and the pressure of impeller cover side cavity and balance hole leakage was systematically tested when the radial clearance of sealing ring and the diameter of balance hole were different. Experimental results showed that the radial clearance of rear sealing ring and the diameter of balance hole had different effects on the radial pressure gradient of pump cavity. The diameter of balance hole had little effects on the pressure of the front pump cavity. When the clearance of the front and rear sealing rings were the same, the pressure of rear pump cavity was generally higher than that of the front pump cavity. For the equilibrium chamber liquid, increasing the diameter of balance hole could relieve the pressure, and increasing the radial clearance of sealing ring could increase the pressure. Combined with the test data, the potential head correction coefficient, the rotation undetermined coefficient and the flow proportional coefficient of different specific areas were calibrated, with a specific solution equation. In this study, the reliability of the proposed pressure mathematical model for impeller cover side cavity was verified by three cases. The results showed that the theoretical prediction value was more consistent with the actual measured value, proving that the proposed mathematical model had high accuracy and universality.

In a bifurcation including a mother artery and two daughter arteries, the energy drop is minimum, if, the cube of the radius of the mother artery equals the sum of the cube of the radii of daughter arteries. This is the expression of Murray’s law (or cubic law) assuming the flow is steady. In this paper, an extension of Murray’s law is investigated using the minimum energy hypothesis, totally analytical for pulsating flow. In addition to the two terms that Murray considered in his calculations, there is additional energy to move fluid toward and back in the pulsating flow. This additional energy is calculated and added to two other parts of energy in Murray’s analysis, and then optimized. The relationships for diameters and the angle between daughter arteries are extended. The effect of frequency and Womersley number have appeared as coefficients in the relations. According to the results, the most difference between Murray’s law for both diameters and the angle between daughter arteries, and the relationship derived in the present paper, occurs in Womersley number between 2 and 5. For a special case which in the daughter arteries have the same diameter, the power of diameters varies up from 3 to 3.2. Also, for this special case, there is maximum 6 degrees difference with Murray’s law for the angle between daughter arteries. In short, the obtained relations, assuming pulsating flow, do not yield very different results from Murray's law assuming steady flow.

Centrifugal pumps often deviate from its design condi-tion during its operation and work at low mass flow conditions. Under such circumstances, unstable flow phenomena may be generated, affecting the efficient and stable operation of pumps. In this paper, a self-circulating casing treatment in U-tube shape is employed on a centrifugal pump to study its effects on the pump’s performance by computational and experi-mental studies. CFD results show that as the flow rate decreases, the back-flow in the inlet pipe of the studied pump without casing treatment increases in intensity and spreads over an growing distance, interfering with the main flow. CFD results also reveal that the casing treatment has a sucking function to the back-flow due to the blade loading of the pump, and when the inlet bleed of the U-tube is placed above (in front of) the leading edge of the blades, the sucking is the strongest, and the control of the back-flow and the improvement to the head coefficient under low mass flow conditions is the best, as the vortex blockage caused by the sucked back-flow in the U-tube is the smallest; when the bleed is under (after) the leading edge of the blades, the effect of the casing treatment is the second best; and when the bleed is across the leading edge of the blades, the blockage in the U-tube is most severe, and the sucking function is the weakest, so there is little improvement to the back-flow and head coefficient. Finally, the relia-bility of this study was demonstrated employing an open pump experimental system with the original pump and the same pump with the casing treatment whose bleed is located above the leading edge of the impeller.

The intake port of a CA4DD diesel engine was investigated by a computational fluid dynamics (CFD) numerical simulation to enhance the intake swirl and improve the flow characteristics in the engine. The uniform design method was applied to study the influence of guide vanes with different parameters on the intake process at various valve lifts. Nine guide vane models were established and compared to the base model using CATIA™ 3D CAD software. Numerical simulations were conduct with Xflow software based on the lattice Boltzmann method (LBM). The distributions of the streamlines, vorticity, velocity and turbulent intensity in each cylinder were simulated and analyzed. The results show that the influence of the guide vanes on the swirl ratio was greater than 37%, and the flow coefficient was less than 5% compared to the base model. Scheme 5, H7.5-L50-θ20 (guide vane height of 7.5 mm, length of 50 mm and angle of 20°), provided good performance. The flow characteristics of the optimal guide vane model were verified through a steady flow test. When the guide vane aspect ratio and angle were within the ranges of 3.5-6.9 and 12.2°-20.2°, respectively, the swirl ratio had the best effect at maximum valve lift. This study provides a theoretical basis for improving the performance of dual-intake diesel engines.

The unsteady vortex evolution of NACA 0012 airfoil is numerically investigated at low angles of attack ranging from 0° to 10° where the separation is performed from the trailing edge. The Reynolds number ranges between 1000 and 4000. The laminar separation bubble at the trailing edge is observed and the main flow features are presented. It is found that, the increase of the angle of attack and Reynolds number result higher lift to drag ratio by an extensive decrease of the drag coefficient below 8° angle of attack. The transition from the steady condition to periodic force evolution has been revealed with a detailed flow field analysis at different angles of attack and Reynolds numbers. The critical angle is defined as the angle of attack where the onset of oscillations starts with a dominant fundamental frequency of oscillation. The angle of attack where the laminar separation bubble (LSB) is observed is also revealed in the current study. Both the LSB formation angle and the critical angle of attack is found to decrease with the increase of the Reynolds number from 1000 to 4000.

The paper presents the multiscale analysis for the hydrodynamic step bearing with ultra low surface clearances where only the physical adsorbed layer is present in the outlet zone and the continuum fluid flow mainly occurs in the inlet zone. This bearing can occur under heavy loads. The flow in the outlet zone is described by the nanoscale flow equation, while the flow in the inlet zone is described by the multiscale flow equation incorporating both the adsorbed layer flow and the intermediate continuum fluid flow. The pressure and carried load of the bearing were derived. Exemplary calculations show that the fluid-bearing surface interaction has the strongest influence on the pressure and carried load of this bearing when the bearing surface clearance is as small as possible, the bearing step size is close to the surface clearance in the outlet zone and the value of the geometrical parameter  is the optimum one, which depends on the fluid-bearing surface interaction. For the strong fluid-bearing surface interaction, the carried load of the bearing can be 10 times higher than that calculated from the classical hydrodynamic lubrication theory.

Misalignment effects increase the risk of scratch or seizure failure for the journal bearings-rotor system. To investigate the performances of misaligned journal bearings, a full-scale journal bearings-rotor system test rig is designed and journal orbits are tested under different speeds. Then, a THD lubrication model is established to simulate lubrication performances under different misalignment conditions. The results show that for a misaligned journal bearings-rotor system, the center of the orbits of the bearing closing to coupling is fixed around the coupling. The orbits center of the bearing far away coupling is floating gradually with an increasing speed. The vibration amplitude of bearing closing to coupling is larger than that of bearing far away from the coupling. When the minimum oil film thickness is decreasing, the maximum film temperature is increasing except for the bearing closing to coupling. For different misalignment directions, maximum film temperature is determined by the position of the high pressure zone and high temperature zone.

The present study pertains to the experimental work on the characteristic of a rotating liquid jet under various conditions of the nozzle diameter, volumetric flow rate and rotating speed. With emphasis on the important phenomena of a liquid jet, the effects of breakup length, the transition between dripping and jetting, breakup categories, droplet sizes from the breakup, and the time interval between two successive droplets are investigated systematically. The results reveal that the breakup length of a jet increases with flow rate and decreases with imposed rotation. The hysteresis behavior only occurs for larger nozzles, and the transition from jetting to dripping is affected by the imposed rotation. Depending on the imposed rotation, three different breakup patterns are found and named single droplet, satellite droplet, and multi-position necking. An empirical correlation is also proposed to predict the boundary of satellite and multi-droplets formation. The main droplet, satellite droplet, and merged droplet are about 1.8, 0.8, and 2.2 times than the nozzle diameters, respectively, no matter what the rotating speed is. Moreover, the non-dimensional time interval between two main droplets has an ascending tendency with either We number or the imposed rotation.

The effect of leading-edge tubercles on the aerodynamic performance of E216 airfoil is studied by steady 3D numerical simulations using Transition γ−Reθ turbulance model. The investigation is carried out for the various angles of attack in the pre-stall region at Reynolds number of 100,000. Various tubercle configurations with different combinations of amplitude ranging from 2 mm to 8 mm and wavelength varying from 15.5mm to 62 mm are studied. The effect of tubercle parameters on the laminar separation bubble (LSB) is extensively studied. Improvement in the coefficient of lift (Cl) is observed for most of the tubercled models and is significant at high angles of attack. But the simultaneous increase in the drag coefficient resulted in a marginal improvement in the coefficient of lift to drag ratio (Cl/Cd) for most of the cases except for A2W62, which produced a peak value of 46.91 at AOA 6◦ which is higher than that for the baseline by 7.37%. Compared to the baseline, the magnitude of suction peak is higher along the trough and lower along the peak. The low amplitude and low wavelength tubercle model exhibited smooth surface pressure coefficient (Cp) distribution without any sign of strong LSB formation. The LSB moves upstream with the increase in amplitude and wavelength. The LSB along the trough is formed ahead of that at peak inducing three-dimensional wavy shaped LSB unlike the straight LSB as in baseline. Two pairs of counter rotating vortices are formed on the airfoil surface between the adjacent peaks at two different chord-wise locations which strongly alter the flow pattern over it.

The aim of this study was to investigate the condensation of HFC-134a vapor on a shock tube wall behind shock waves. The time-dependent thickness of the condensed liquid film was measured using an optical interference method based on multiple reflections of a laser beam. The condensation on the wall was accompanied by an instantaneous increase in the pressure behind the incident shock wave, and when the reflected shock wave reached the observation window, condensation occurred again. In this experimental study, the characteristics of the diaphragmless vertical shock tube were verified. Reliable experimental data could be obtained using the shock tube. The shock waves could be visualized to study their behaviors in different time periods. The experimental results confirmed the formation of a liquid film on the cold wall of the shock tube after the passing of incident and reflected shock waves, with the liquid film behind the incident shock wave exhibiting a faster formation.

The cylindrical volute intake structure possesses some advantages including convenient processing, convenient installation & uninstallation and high machining efficiency. The helium turboexpander with this novel intake structure in a superconducting cryogenic device is investigated deeply in this study. Based on the established mathematical model and the corresponding numerical computation methods, the whole flow passage internal flow of the helium turboexpander is numerically simulated. And then the distribution characteristics of total pressure, static pressure, static temperature, relative velocity and total enthalpy in the cylindrical volute, nozzle, impeller and diffuser are explored, and loss mechanism of the internal flow is analyzed. The results indicate that the novel cylindrical volute intake structure has obvious pre-rotation effect on the inlet flow field of the nozzle, and this intake structure has little loss. In addition, the expansion effect in downstream components including nozzle and impeller is obvious, and the flow field changes uniformly. The overall efficiency of the turboexpander is up to 84.8%, which indicates that it is reasonable that the novel cylindrical volute is used as the intake structure of turboexpander.

An optimal Kriging surrogate model based on a 5-fold cross-validation method and improved artificial fish swarm optimization is developed for improving the aerodynamic optimization efficiency of a high-speed train running in the open air. The developed optimal Kriging model is compared with the original Kriging model in two test sample points, and the prediction errors are all reduced to within 5%. Thus, the optimal Kriging model is selected for use in each iteration to approximate the CFD simulation model of a high-speed train in subsequent optimization. After that, the strong Pareto evolutionary algorithm II (SPEA2) is adopted to obtain a series of Pareto-optimal solutions. Based on the above work, a multi-objective aerodynamic optimization design for the head shape of a high-speed train is performed using a free-form deformation (FFD) parameterization approach. After optimization, the aerodynamic drag coefficient of the head car and the aerodynamic lift coefficient of the tail car are reduced by 5.2% and 32.6%, respectively. The results demonstrate that the optimization framework developed in this paper can effectively improve optimization efficiency.

Flow optimization and drag reduction are of great importance in industrial applications. However, most of the structural optimization and drag reduction in pipe flows are based on industrial experience or a large number of experiments, and there is a lack of general theoretical guidance. In the present work, a general approach for flow optimization and drag reduction in turbulent pipe flows is developed based on the irreversibility of flow process and the principle of minimum mechanical energy dissipation. Considering that the effective viscosity coefficient is related to the space coordinates, the field synergy equation of turbulent flow is derived. The reliability and performance of the field synergy principle of turbulent flow as well as the general approach are then evaluated and validated in a turbulent parallel flow conduit, and finally applied to industrial pipe flows. It demonstrates that the present approach is able to optimize flow field for different purposes by adding speed splitter or deflector as an interface at proper locations to alter the interactions between fluid and wall. It is robust and easy to implement, which provides general theoretical guidance for flow optimization and drag reduction in turbulent pipe flows.

The original (delayed) detached-eddy simulation ((D)DES), a widely used and efficient hybrid turbulence method, is confronted with some flaws containing grid-induced separation (GIS), log-layer mismatch (LLM), and slow RANS-LES transition. A novel hybrid turbulence model, namely production-limited eddy simulation (PLES), depleting the production through introducing the subarid-scale eddy viscosity is proposed. The simulation data of the zero-pressure gradient boundary layer proves that a good performance in mitigating the GIS issue is obtained from the PLES model. The results of the channel flows reveal that the PLES model has eliminated the LLM of the velocity. A good conformity is given for the backward-facing step flow in the PLES simulation, which proves that the PLES model is validated for complex flow. More turbulent scales in the shear layer are captured by the PLES model, which testifies that the PLES model has a faster RANS/LES switch than the IDDES model. The PLES model has a good performance in predicting the cylinder flow with coarser grid resolution. Due to the new LES mode, the PLES model behaves better than the IDDES model in simulating the cylinder flow. Furthermore, the PLES model allows one to use different LES model in the LES portion for other complex flows.

The gas-liquid two-phase flow with interfacial behaviors and bubble-liquid interactions is widely encountered in industrial processes such as that in gas-liquid reactors. The complicated phase structure makes it difficult to be modeled. The present work proposes a multi-scale mathematical model to simulate the bubbly flow in a square column. The volume of fluid (VOF) method is applied to treat the separated interface, and the discrete bubble model (DBM) is incorporated to handle the dynamics of dispersed bubbles. The hybrid model is validated against the benchmark experimental data to study the accuracy and suitability of the modeling framework for bubbly flows. And the influence of interphase forces on bubbly flow patterns and velocity profiles is investigated. It is found that the employment of both pressure gradient force and Ishii-Zuber drag model provides fairly good agreements with experimental data for velocity profiles.

To investigate the dynamics of droplet-vortex interactions in particle-laden Karman vortex street flows, the simulations were carried out by using Euler-Lagrange approach, which was validated by the available experiments and numerical results. Then, the particle dispersion and the dimensionless frequency (Strouhal number) of the wake flow were analyzed to evaluate the particle-vortex interactions. The particle dispersion was statistically analyzed from both time and space dimensions and the different instantaneous dispersion patterns were explained by the relative slip velocity. Two independent scaling parameters, Stokes number StL and particle-fluid mass loading ratio Φ were revealed, and the particle mean square displacement and the Strouhal number were modelled by using these two scaling parameters, respectively. Finally, the characteristic lengths of the particle-laden wake flow were researched, and the Strouhal number physical model was developed based on the oscillating fishtail model. The results indicated that, firstly, StL and Φ, which constitute a dominant scaling group, can characterize the dynamics of droplet-vortex interactions in wake flow. Particles gradually separate from the vortex with the increase of StL due to the centrifugal effect, and the vortex intensity and regularity get worse with the increase of Φ, which further disperses the droplets for their momentum exchange with irregular vortex structures. Secondly, the length of the formation region and the width of the free shear layer diffuse are the two simultaneous characteristic lengths of the Strouhal number in oscillating wake. The proposed Strouhal number model gives a physical basis for the frequency determination, and the predicted errors are within ±1.5% error bands with mean absolute percentage error of 0.67%.

Three-dimensional numerical simulations are performed to examine the effects of dynamic wing morphing of a hummingbird-inspired flexible flapping wing on its aerodynamic performance in hovering flight. The range analysis and variation analysis in the orthogonal experiment are conducted to assess the significance level of various deformations observed in the hummingbird wings on wing aerodynamic performance. It has been found that both camber and twist significantly can affect lift, and twist has an even higher significant impact on lift efficiency. Spanwise bending, whether out-of-stroke-plane or in-stroke-plane, has a negligible impact on lift and efficiency, and the in-stroke-plane bending can cause lift to decrease to an extent. Optimal parameters for determining the wing deformations are selected and tested to validate the conclusions drawn in the analysis for the results in orthogonal experiment. Through a comparison study between the optimized wings and the rigid wing, it is found that although the wing flexibility can cause the net force to decrease, the flexible wing used less energy to bring the net force closer to the vertical direction, thereby improving the lift efficiency. This study provides an aerodynamics understanding of the efficiency improvement of the hummingbird-inspired flexible flapping wing.

Booster pump system (BPS) can control the number of revolutions through an inverter by combining two or more vertical or horizontal centrifugal pumps in a series. Efficiency and energy savings, the most appealing aspects of booster pump systems, can be improved by controlling the operating conditions of individual pumps by measuring the flow rate of each pump. For improved operation, a booster pump system with a flow sensor to detect individual pump flow rates and a control algorithm to manage each low and high flow rate pump’s revolutions per minute are critical. To achieve this, first, the turbine-type flow sensor was developed through computational fluid dynamics and experimentation. The flow sensor was improved using computational fluid dynamics, and its accuracy was validated through experiments. The resulting flow measurement accuracy of the designed flow sensor was within 4%, with a measurement uncertainty of 0.4%. In addition, an experimental pump facility was built and used to evaluate booster pump system performance to investigate the energy saving rate. Then, after driving one low-flow rate pump at a set pressure, the flow and frequency control operation algorithm was used. This algorithm increased the allowed output of the drive pump by increasing the inverter’s frequency. When the frequency corresponding to the allowed output is achieved in the low-flow rate pump rather than the high flow rate pump, power savings increased due to the low-flow rate pump’s extended drive range. The investigations on the developed system’s energy consumption revealed that the energy savings were approximately 6.2% compared to the conventional system, depending on the system in question. The development of a booster pump system with a flow sensor was tested, and it was found to be effective.

Numerical simulation was used to investigate the effect of an inclined volute tongue on the complex flow characteristics and the aerodynamic performance of multi-blade centrifugal fans. We focused on the effects of the clearance, the inclination angle and the volute tongue’s on controlling the centrifugal fan’s internal flow characteristics and aerodynamic performance. The results showed that the volute tongue’s reasonable clearance ratio and the inclined volute tongue design were beneficial to improving the flow pattern around the volute tongue and volute outlet of the centrifugal fan, as well as reducing the local flow loss. It is of great significance to increase the centrifugal fan’s static pressure and related efficiency by increasing the radius and inclination angle of the volute tongue. Due to the reduced of vortex, the local flow loss was reduced. Numerical results indicated that model C’s static pressure rose to 12.5Pa, and the related static-pressure efficiency of to 3.8% compared with the reference geometry due to the reduced of flow loss.

Aquatic animals usually generate the effective propulsive force via non-sinusoidally flapping their fins. Inspired by the kinematics of fish, the propulsive characteristics of a NACA012 hydrofoil is numerically studied in this paper. The combination of non-sinusoidal heaving and pitching motions is adopted in the two-dimensional hydrofoil kinematics parameters. The elliptic function and the flattening parameter S are introduced to achieve the varieties of non-sinusoidal periodic motions. The numerical model is established by using the commercial computational fluid dynamic solver STAR-CCM+, and the code is verified by comparing with the published experimental results. The Reynolds number is fixed at 40,000 in all the numerical simulations. The results show that the non-sinusoidal trajectories affect the propulsive performance by affecting the angle of attack (AOA), the hydrodynamics of the foil and the flow structure behind the foil. The non-sinusoidal flapping trajectories can improve significantly the thrust coefficient at the same kinematics parameters compared with the sinusoidal motions in most cases. However, they may reduce the propulsive efficiency. When the values of S are greater than 1, the improvement of thrust coefficient acquired with the non-sinusoidal motions is more obvious. The wake pattern is also discussed which indicates that the strong leading-edge vortices results in the decrease of the propulsive efficiency acquired by the non-sinusoidal trajectories. It is possible to apply the non-sinusoidal motions of a flapping foil to improve propulsive performance of the underwater bionic machine.

Stall is a complex flow phenomenon in centrifugal pumps at part load conditions. However, there is no clear description of the stall evolution process and mechanism, which is critical for stall control. Based on a high-frequency Particle Image Velocimetry (PIV) system (10k Hz) and a non-refraction experimental bench, emphasis is laid on the flow structures near the initial stall conditions. The results show that as the flow rate decreased, the flow separation occurred at the middle of the blade suction side and then evolved into a stall vortex which moved to the impeller’s inlet direction and kept growing. Subsequently, it broken into two vortexes when reaching the location where the impeller cross-sectional area is the smallest. One stall vortex continued its motion toward the passage inlet direction, while the other vortex separates to the impeller outlet. As the stall vortex’s size at the impeller inlet enlarged, the flow incident direction at the impeller inlet was directed to the blade suction side, which caused the stall vortex on the suction side to disappear. The stall mechanism is explained in detail using both experiments and numerical simulations. Meanwhile, the Scale Adaptive Simulation-Shear Stress Transport (SAS-SST) hybrid model is used to simulate several flow rate conditions near the stall initial stage. The findings indicate that the increasing adverse pressure gradient and the high-pressure zone move along the suction blade towards the impeller inlet as the flow rate is reduced; however, the relative velocity is constantly decreasing. When the fluid can’t provide enough kinetic energy to maintain its continuous flow along the suction surface, flow separation occurs. The stall vortex, which results from flow separation, blocks the passage impeller. The increasing adverse pressure from the impeller outlet to the inlet is the main cause of flow separation; and the adverse pressure gradient is a major manifestation of the stall vortex.

To understand the jet flow characteristics of turbofan separate exhaust system, a parametric design method based on the initial Class Shape Transformation function was developed. HBPR and UHBPR turbofan separate exhaust systems were designed. Furthermore, the jet flow characteristics of the HBPR turbofan exhaust system under take-off condition with zero angle of attack were studied based on numerical simulation. The jet flow characteristics of the HBPR and UHBPR turbofan exhaust system under take-off condition with high angle of attack were also simulated. The effects of angle of attack and bypass ratio on the jet flow characteristics were investigated and the related flow mechanisms were analyzed. Results show that the axisymmetric plumes of the HBPR turbofan exhaust system are distributed around the engine axis under take-off condition with zero angle of attack. With the plug wake as the center, the core flow, the fan/core shear layer, the fan flow, the fan/free stream shear layer and the free stream are wrapped around the plug wake from inside out. Vortexes appear in the lee area at the back of the cowl and jet flow under take-off condition with high angle of attack. These vortexes cause cross sectional secondary flow and expose the high-velocity core flow to the low-velocity free stream. The contact area and velocity gradient in the mixing region among the free stream, fan flow and core flow increase. Therefore, the mixture among jet flow and free stream strengthens. So the high-velocity region, the high-vorticity region, and the high turbulence kinetic energy region shorten by 55.1%, 47.7% and 50.9% respectively. The vorticity values and turbulence kinetic energy level peak on the upper side of the exhaust plumes increase by about 30% and 87% respectively. Relative to these parameters from the HBPR turbofan exhaust system, the jet velocity peak value of UHBPR turbofan decreases by 5.5% under take-off condition with high angle of attack. The vorticity values and turbulence kinetic energy level reduce due to decreased velocity gradient in shear layers downstream of the nozzle exit plane. The turbulence kinetic energy level peak on the upper side of the exhaust plumes decreases by 29.3%. The reasons are that the contact area between high-velocity core flow and the free stream decreases due to thicker fan flow and the velocity gradient in the core flow and free stream mixing region decreases because of the lower core flow velocity.