A

Proceedings of ASME Turbo Expo 2018

tte
Turbomachinery Technical Conference and Exposition

nd
ee
GT2018
June 11-15, 2018, Oslo, Norway

R
ea
d-
O
nl
GT2018-75110

y
C
op
y
INVESTIGATION ON A BLOCKERLESS CASCADE THRUST REVERSER SYSTEM
BASED ON RESPONSE SURFACE METHOD

Li Zhou Zhanxue Wang

Northwestern Polytechnical University Northwestern Polytechnical University
Xi’an, Shaanxi, China Xi’an, Shaanxi, China

Jingwei Shi Xiaobo Zhang

Northwestern Polytechnical University Northwestern Polytechnical University
Xi’an, Shaanxi, China Xi’an, Shaanxi, China

ABSTRACT especially for the interaction effect between FPR and X jet ,
The blockerless cascade thrust reverser is one of the interaction effect between FPR and β . Optimization design
innovative thrust reverser systems, which replaces the with objective of maximum reverse thrust was carried out to
traditionally mechanical blocker door with the aerodynamic determine the best parameter settings, and reverse thrust ratio
blocker door by high-pressure secondary injection, thus
significantly reduces the nacelle weight and the complexity of
 Trev of 60% is achieved under the constraint of the
the actuator, and especially suitable for high-bypass-ratio secondary flow ratio.
turbofan engine. In order to obtain the optimum performance of
a blockerless cascade thrust reverser system and provide the NOMENCLATURE
guidance for the design of the blockerless cascade thrust X jet : secondary injection position mm
reverser system, a blockerless cascade thrust reverser system
was studied in this paper based on the Response Surface  jet : secondary injection angle (º)
Method (RSM), focusing on the effect of different geometric and β : cascade installation angle (º)
aerodynamic parameters on the thrust reverser performance. δ : cascade solidity
Results show that the secondary injection with high pressure *
FPR: fan pressure ratio ( Pfan / Patm )
forms the blockage effect to the fan flow, then forces the fan
*
flow to deflect and discharge from the cascade window, SPR: secondary pressure ratio( Psec / Patm )
realizing the reverse thrust. The thrust reverser performance is  Trev : reverser thrust ratio
mainly affected by fan pressure ratio (FPR), secondary flow
pressure ratio (SPR), secondary injection position ( X jet ),  Frev : thrust reverser flow ratio
secondary injection angle (  jet ) and cascade installation  Jrev : secondary injection flow ratio
angle (β), and the dominated factors are FPR, SPR and X jet . Fi , fan
: ideal thrust of fan bypass
According to the obtained response equation of the thrust Frev
reverser performance, the relationship between reverse thrust : reverse thrust
efficiency and various parameters are clearly described, and W cas : flow rate discharging from the cascade window
performance of thrust reverser can be quickly evaluated.
Significant interaction effects exist between different two
W fan : flow rate of the fan bypass
factors, which must be taken into consideration in the design W jet : flow rate of the secondary injection
process of the blockerless cascade thrust reverser system,

1 Copyright © 2018 ASME

INTRODUCTION cascade opening[8]. Marconi explored the principle of the
With the continuous improvement of the aircraft carrying secondary injection to realize the reverse thrust, and
capacity, the corresponding landing distance gradually numerically studied effects of aerodynamic parameters and
increases. In order to decrease the landing distance, as well as geometric parameters on the performance of blockerless
deal with all kinds of emergency situations during takeoff or cascade thrust reverser, and found that the one design variable
landing, safe and reliable deceleration technology becomes that clearly showed significant effects was jet orientation [9].
increasingly important, especially for safe landing in rainy / NASA Langley Research Center experimentally investigated
snowy weather and stopping rolling urgently [1, 2]. Currently, performances of six innovative thrust reverser concepts which
the thrust reverser system is commonly used as the engine were applied to high-bypass-ratio turbofan engines, and pointed
deceleration device for aircraft with large passenger capacity out the blockerless cascade thrust reverser was one of the most
and transport aircraft, and traditional thrust reverser systems are important one. Then, a detailed study was carried out for the
usually mechanical types, which include clamshell thrust blockerless cascade thrust reverser, and the effects of various
reverser, baffle plate thrust reverser, and cascade thrust factors on the performance of the blockerless cascade thrust
reverser[3]. Massive investigations had been performed on reverser were obtained under different conditions. Results
those mechanical thrust reverser systems. He and Liu[4] carried showed the thrust reverser efficiency increased with the
out detailed studies on the mechanical cascade thrust reverser increment of the flow rate of the secondary injection and the
system, discussed influences of geometric parameters and reduction of fan pressure ratio[10]. Hall performed the
areodynamic parameters on the effectiveness of the thrust experimental investigation on the thrust reverser control of a
reverser, and pointed out the flow coefficient and the thrust turbofan engine based on Coanda effect, where the thrust
reverse efficiency were increased with the increase of the reverver window was not equipped with the cascade[11]. The
solidity, when the solidity is less than 1.59. Butterfield explored core idea was to inject the secondary flow near the outer wall of
the approach of weight reduction in a mechanical cascade thrust the fan bypass and the wall of the blocker door to form the
reverser system by modifying the vane configurations [5]. But Coanda effect, then to realize the deflection of the airflow.
when the mechanical thrust reverser systems are adopted, the Thus, the mechanical cascade could be removed, and the
nacelle weight significantly increases, the extra mechanical nacelle weight could be reduced in further. It was found that the
moving parts are necessary for the actuator, the seal structure is effectiveness of the thrust reverser was more obvious when the
also need to be arranged in order to avoid leakage, and high secondary injection were attached to the wall of the blocker
maintenance cost and maintenance difficulty are inevitable for door. Chen and Shan numerically studied the impact of the
the thrust reverser systems. Meanwhile, with the increase of the secondary injection positions, injection flow rate, hole numbers
bypass ratio, the above drawbacks of the mechanical thrust and injection angle on the performance of thrust reverser system
reverser systems become more and more apparent. Thus, the and flow characteristics of bypass flow [12, 13], and showed
exploration of new thrust reverser type became the focus of secondary injection angle had a dominant effect on thrust
research, NASA, GE and other research institutions began to reverser performance, an optimum injection flow rate existed to
investigate new type of the thrust reverser system. As compared get the highest thrust reverser efficiency, the blocking effect of
with the traditional mechanical thrust reverser, the thrust the secondary flow on the fan flow changed with the secondary
reverser based on aerodynamic control has advantages of injection position. It is pointed out that the effects of jet hole
simple structure, light weight, fast response, et al., it is known position, angle and jet flow rate were connected to consider on
as the promising direction for the development of the thrust the design of the blockless thrust reverser, but how to include
reverser system in the future. the interaction effect among these three parameters in the design
The blockerless cascade thrust reverser is one of the was not involved. Zhang et al. simulated the flow field of the
innovative thrust reverser systems, which replaces the blockless thrust reverser system, denoted optimum injection
traditionally mechanical blocker door with the aerodynamic angle and injection pressure resulted in the highest thrust
blocker door by high-pressure secondary injection, thus efficiency[14]. In the previous investigations of the authors,
significantly reduces the nacelle weight and the complexity of experiments and numerical simulation on the blockerless
the actuator, and especially suitable for high-bypass-ratio cascade thrust reverser system, effects of fan pressure ratio,
turbofan engine. Then, researchers paid attention to the secondary pressure ratio, secondary injection position and
blockerless cascade thrust reverser system and carried out a secondary injection angle on the performance of the blockerless
great deal of research work. Chen and Yao discussed the idea cascade thrust reverser were analyzed in detail[15-17]. Results
using a secondary injection to form an aerodynamic blocker showed fan pressure ratio and secondary injection flow rate had
door in the fan bypass, instead of a mechanical blocker door, prominent impact on the efficiency of the thrust reverser,
and gave an early indication of whether a cascade-less reverser complicated coupling effect displayed between these two
was a viable option for a particular application [6, 7]. Gilbert parameters, the efficiency of the thrust reverser increased with
verified the concept of the blockerless cascade thrust reverser, the increment of the fan pressure ratio when the injection flow
with a small portion of secondary flow injected into the fan rate is below a certain value, but expressed a reverse trend when
bypass, 100 percent of the fan flow could be turned into the the injection flow rate is larger than this certain value. Thus, it

2 Copyright © 2018 ASME

had to consider the complicated coupling effect for the design cases, and the corresponding form of the response surface
of the blockless thrust reverser. equation is as follows [19,20]:
Attentions had been drawn on the blockerless cascade k k k 1 k

thrust reverser system, and it had been known that the Y      j x j    j x 2j    ij xi x j (1)
j 1 j 1 i 1 j i 1
performance of the blockerless cascade thrust reverser system
was related to massive aerodynamic and geometric parameters, Where Y is the response variable, X ( x1 , x2 , x3 , x4 ... xk ) are
such as fan pressure ratio (FPR), secondary pressure ratio different influence factors, k is the number of influence
(SPR), secondary injection position ( X jet ) and secondary factors,  ,  j ,  j , ij are coefficients of the polynomial.
injection angle (  jet ), cascade installation angle (β), cascade
Establishment of RSM model based on DOE
solidity (δ) et al.. Although a lot of investigations had been
performed on the blockerless cascade thrust reverser system, it The principle of approximate modeling technology is to
must be aware that the understanding of this new thrust reverser construct the approximate model of the research object through
technology is still not enough, the emphasis of most data analysis of the selected samples, then use the approximate
investigations focused on the influences of secondary injection model instead of the exact model to conduct the research. The
parameters on the effectiveness of the thrust reverser, and basic steps of approximate modeling can be simply summarized
studied the effect of a single factor on the thrust reverser as four parts: Design of Experiment (DOE), numerical
performance by the control variate method, without taking into simulation, establishment of approximate model and evaluation
account of the interactions between the various factors, couple of approximate model.
effects among the cascade parameters (installation angle, The essence of RSM approximate modeling is to obtain
cascade solidity, cascade profile), fan flow and secondary system response functions Y on design factors X based on DOE
injection were not comprehensively analyzed, the relationship by polynomial regression or neural networks training. The DOE
among the resolved multi parameters affecting the thrust theory provides the sampling points, at which to perform the
reverser efficiency cannot be accurately established. tests, in order to minimize the number of simulations. For RSM
In order to obtain the optimum performance of a objective functions with polynomials, whose order and shape
blockerless cascade thrust reverser system and provide the must be chosen beforehand. Commonly, a second order
guidance in the design process of the blockerless cascade thrust polynomial Eq. (1) is usually used. The polynomial
reverser system, a blockerless cascade thrust reverser system coefficients  ,  j ,  j , ij are determined by means of a
was studied in this paper based on RSM, focusing on the effects standard least-square regression which minimizes the sum of the
of different geometric parameters and aerodynamic parameters square of the deviations for the predicted values from the actual
on the performance of the thrust reverser. The relationship ones for a set of points. And the validity of established
between the thrust reverser efficiency and various factors were approximate model is verified through different parameter, e. g.
obtained, and the interactions between the various factors were R 2 (the ratio of the model sum squares to total sum of squares)
discussed, then the optimization design with objective of and R2 ( R 2 adjusted to the number of parameters in the model).
maximum reverse thrust was carried out, and the best parameter
settings were determined.
GEOMETRY MODEL AND NUMERICAL
RSM APPROXIMATE MODELING SIMULATION METHODOLOGY
Geometry model
Introduction of RSM
The blockerless cascade thrust reverser configuration
RSM is the production of statistical methods and
studied in this paper is shown in Figure 1, which is referred to
mathematically methods, the approximate modeling based on
1/10 scale model of CMF56-2 engine (bypass ratio is 5:1)[10].
RSM is one of the most mature and practical modeling
Curve A-C represents the fan inlet, curve A-B is on behalf of
methods. The purpose of RSM is to analyze the response value
inner wall of the fan bypass, the position of the secondary
of sample points and establish the approximate model of
injection is located on the curve A-B, as shown in Figure 1.
research object [18].
Curve C-D and curve E-F are outer walls of the fan bypass,
According to the construction method of the fitting
curve B-F is the fan outlet, and curve H-F is on behalf of outer
function, main forms of the RSM include quadratic polynomial
surface of the nacelle, the zone D-G-H-E-D stands for the
response surface, moving least squares response surface,
cascade window of the thrust reverser.
Kinging response surface, radial function response function and
neural network response surface and so on[19].
Compare to other fitting methods, RSM based on
polynomial fitting has advantage of simply modeling, excellent
fitting accuracy and high computation efficiency, the quadratic
polynomial response surface model is widely used in most

3 Copyright © 2018 ASME

 crossing the mesh face that lapped with the wall boundary. The
flow field convergence is judged by monitoring the local
velocity distribution of the point at cascade window inlet.
To evaluate the effect of grid sizes on the numerical
results, four grid sizes are selected to solve the flow field, which
are 40,000 60,000, 80,000 and 100,000 respectively. Static
pressures (P/Pb) on the pressure surface of the cascade are
compared in Figure 3, where FPR=1.4, SPR=4, X jet =65mm,
FIGURE1. GEOMETRY MODEL OF THE BLOCKERLESS  jet =50°, β=60°. The result of grid independence check shows
CASCADE THRUST REVERSER
that the deviation of the pressure decreases with the increment
Numerical simulation methodology of the grid number, and the pressure with grid number of 80,000
Numerical simulations were performed for the blockerless coincides with that it has for grid of 100,000, the maximum
cascade thrust reverser model using FLUENT software. error in the local position is 0.21%, and therefore grid number
Equations solved were steady, compressible, Reynolds- of 80,000 is used for the flow field calculation.
averaged Navier-Stokes equations. The implicit density-based
algorithm was used to solve the equation. The equations were
discretized in finite volume form on each hexahedral control
volume. Second order upwind scheme was used in the spatial
discretization. Standard κ-ε turbulence model was selected as
turbulence model.
Details of the computational domain and boundary
condition settings are presented in Figure 2. Relative to the fan
inlet, 6 times of the fan inlet length is extended upstream of the
ambient region surrounding the thrust reverser (boundary A), 10
times of the fan outlet length downstream of the fan outlet
(boundary C), and 3 times of cascade window width along the
direction which is normal to the cascade window (boundary B).
FIGURE3. GRID INDEPENDENCE CHECK
A structured grid of the model was created, grid around
secondary injection ports and near walls were refined to meet Performance parameters
the Y+ requirement of the standard κ-ε turbulence.
Reverse thrust ratio Trev , reverse thrust flow ratio
 Frev and the reference parameter of secondary injection flow
ratio  Jrev are used to measure the performance of blockerless
cascade thrust reverser in this paper.
Reverse thrust ratio Trev is defined as follows:
Trev  Frev / Fi , fan (2)
where Fi , fan is the ideal thrust of fan bypass, Frev is X
momentum of the flow discharging from the cascade window in
Figure 5, which is the axial reverse thrust at outlet of the
cascade window. Trev represents the percentage of the
A, B, C-pressure far field; D, F-pressure inlet; E-interior
FIGURE2. GRID AND BOUNDARY CONDITIONS reverse thrust in the thrust of fan bypass, is a form that can
reflect the reverse thrust efficiency.
Pressure inlet conditions are applied to the fan inlet
Fi , fan and Frev can be expressed as:
(boundary D) and secondary injection inlet (Boundary F), of
which total pressure, total temperature and flow angle are Frev   V y • Vx dA (3)
prescribed. The ambient regions surrounding the thrust reverser
 1
(boundary A, B and C) are defined as the pressure far field 2 1 
Fi , fan  W p , fan R jTt , fan [1  ( ) ] (4)
condition, with static pressure, static temperature and Mach  1 FPR
number specified. Impermeable, non-slipped and adiabatic wall
boundaries are applied on the fan bypass wall and cascade to
ensure zero normal flux of mass, momentum and energy

4 Copyright © 2018 ASME

where ρ is the density, V y and Vx denote the velocity secondary injection flow ratio  Jrev must be smaller than 0.1
component of the flow at outlet of the cascade window, W p , fan in order to trade off the engine thrust.

means the flow rate of the fan inlet, Tt , fan is total temperature Validation of Numerical Method
of the fan outlet, FPR is on behalf of fan pressure ratio, R j is In order to validate the applicability of CFD methodology
in this paper, the flow field of the blockerless cascade thrust
gas constant, and  is ratio of specific heat.
reverser was simulated in condition that SPR was 8, with the
secondary injection located at the fifth injection position. The
predicted reverse thrust flow ratio  Frev is compared with the
experimental data[10], as shown in Figure 5. Under different
FPR, the predicted  Frev is slightly higher than the
experimental result, agreement is achieved in the distribution
trends with the maximum relative error of 6.28%, the
approximate treatment of the calculation model and the leakage
in the experiment may be the main reason to cause the error.
Generally, the calculated results agrees with the experimental
FIGURE 4. SKETCH OF THE BLOCKERLESS CASCADE data, especially for the case under large FPR, so it shows that
THRUST REVERSER
the predicted flow field is reasonable, the present calculation
Reverse thrust flow ratio  Frev is the ratio of flow rate method is able to be used to simulate the flow field in the
blockerless cascade thrust reverser in following sections.
through the cascade window to the total flow rate including fan
bypass flow and secondary injection flow rate.  Frev reflects
the share of the deflected fan flow, can also represent the
reverse thrust efficiency. The definition of  Frev is:
 Frev  Wcas /(W fan  W jet ) (5)
where W cas is the flow rate discharging from the cascade
window, W fan is the flow rate of the fan bypass, and W jet is
the flow rate of the secondary injection.
Secondary injection flow ratio  Jrev denotes the
proportion of the secondary injection flow to the flow rate of
fan bypass, which is defined as:
 Jrev  W jet / W fan (6)
For the blockerless cascade thrust reverser, secondary
injection flow ratio  Jrev is an important parameter because it FIGURE 5. COMPARISON OF  Frev BETWEEN
CALCULATION AND EXPERIMENT UNDER DIFFERENT
indicates the amount of air bleeding from the core engine,
FPR
which is used to deflect the fan flow.The air bleeding from the
core engine is the product of  Jrev and bypass ratio. If  Jrev FLOW CHARACTERISTIC OF BLOCKERLESS
is too high, that means the engine has to supply large amount of CASCADE THRUST REVERSER
air bleeding, which definitely induces the work capacity of the Figure 6 shows the typical flow field of the blockerless
turbine to drop dramatically, then the thrust produced by the fan cascade thrust reverser system, the principle of the blockerless
flow also strongly reduced, and the corresponding reverse thrust cascade thrust reverser is as follows: the blocker door is
would be decreased obviously even though all the fan flow is replaced by the secondary injection, and the secondary flow
deflected from the cascade window of the thrust reserver. with high pressure is bleed from the core engine and injected
Therefore,  Jrev is restricted by the bypass ratio, and it should into the fan bypass. The interaction between the secondary
injection and the fan flow forms the recirculation zone in the fan
consider the constraint of  Jrev in pursuit of high reverse bypass due to the inconsistent flow direction between the
thrust efficiency. And for the engine with bypass ratio of 5:1, secondary injection and the fan flow, and generates a blockage
effect to the fan flow, then forces the fan flow to deflect and

5 Copyright © 2018 ASME

discharge from the cascade window, realizing the reverse thrust. window under SPR of 5, the increments of Trev and  Frev
The size of the recirculation zone behind the secondary
are only induced by the increase of the secondary injection flow
injection can reflect the deflection degree of the fan flow, the
larger the recirculation zone is, the more deflection of the fan rate. It can also be seen that  Jrev has already exceeded 0.1
flow is. The maximum advantage of the blockerless cascade when SPR is larger than 5, which can not satisfy the air bleeding
thrust reverser system is that the complex actuation system and limitation. Thus, there exists an allowable maximum SPR
the heavy blocker door are omitted, with the satisfactory reverse although performance of the thrust reverser increases with the
thrust effectiveness obtained at the same time. Therefore, the increment of SPR, with the constraint of  Jrev considered.
engine weight with the blockerless cascade thrust reverser
system is much lower than that of the mechanical thrust reverser
system, and the performance of the engine can also be improved
by eliminating the leakage in the components of the blocker
door.

FIGURE 7. PERFORMANCE OF THRUST REVERSER UNDER

FIGURE 6. TYPICAL FLOW FIELD OF BLOCKERLESS DIFFERENT SPR
CASCADE THRUST REVERSER
Figure 8 shows reverse thrust ratio Trev distribution
Performance of the blockerless cascade thrust reverser under different cascade installation angle β, with the SPR of 5
under different secondary pressure ratio SPR is given in Figure adopted, and FPR varies from 1.3 to 1.6. Under the smaller
7, FPR=1.5, X jet =50mm,  jet =50°and β=60° are set for the FPR of 1.3, an optimum cascade installation angle β existed to
calculation. The diameter of the secondary jet in this paper is all get the highest thrust reverser performance. Due to the
chosen as 0.68mm. It can be found that both reverse thrust ratio relatively small momentum of the fan flow under smaller FPR
of 1.3, the fan low can get much stronger deflection under the
Trev and reverse thrust flow ratio  Frev increases with the same blockage effect of the secondary injection, and the fan low
increment of SPR. When SPR is lower than 4, as the momentum enters in the cascade window with large flow angle, thus, the
of the secondary injection is not large enough, the injection optimum Trev is obtained under the much larger β. Under
depth of the secondary flow into the fan bypass is relatively
small, the blockage effect generated by the secondary injection larger FPR, the corresponding larger momentum of the fan flow
induces the deflection angle of the fan low entering the cascade
is limited, thus, Trev and  Frev increase slowly under lower
window to be much smaller, as compared with the case under
SPR. When SPR increases from 4 to 5,  Frev abruptly FPR of 1.3, the satisfactory reverse thrust effectiveness is
increases from 0.345 to 0.946, and Trev changes from 0.223 achieved under smaller β. And Trev decreases with the

to 0.592, it shows that almost all the fan flow discharges from increase of β under FPR of 1.6, the maximum Trev is
the cascade window under SPR of 5. In this case, large achieved under the smallest β resolved in this research due to
recirculation zone forms behind the secondary injection position the smallest flow deflection angle of the fan flow. When FPR
owing to the large momentum of the secondary injection, the increases from the value of 1.3, Trev drops with the
blockage effect generated by the secondary injection has the
sufficient capacity to deflect almost all the fan flow out the increment of β, especially with an obvious descending gradient
cascade window, then obtains the excellent reverse thrust under larger β.
performance. When SPR continues to increase,  Jrev increases
obviously. Although Trev and  Frev keep on rising, as
almost all the fan flow has been deflected to the cascade

6 Copyright © 2018 ASME

 (a)  jet =15°

FIGURE 8. Trev DISTRIBUTION UNDER DIFFERENT

CASCADE INSTALLATION ANGLE β

Figures 9 displays Ma number distribution under different
secondary injection angle  jet , where FPR=1.4, β =60°,
SPR=3.0 and X jet =63.78mm. When  jet is 15°, the
secondary injection direction is opposite to the fan flow, with a (b)  jet =45°
very large momentum component against the fan flow.
Momentum component of the secondary injection used to
deflect the fan flow is very small, the secondary injection is
suppressed by the fan low and attached to the wall, the
corresponding injection depth into the fan bypass is relatively
small, and the blockage effect of the secondary injection
becomes smaller, only a small part of the fan flow is deflected
out the cascade window, as shown in Figure 9a. Similarly, when
 jet is 90°, the secondary injection has a certain component
along the fan flow direction, part of the secondary injection
flows downward directly, the injection depth into the fan bypass
is naturally limited, the proportion of the fan flow deflecting out (c)  jet =90°
the cascade window becomes smaller due to the smaller FIGURE 9. MACH NUMBER DISTRIBUTIONS UNDER
blockage effect of the secondary injection in this case, as shown DIFFERENT SECONDARY INJECTION ANGLE  jet
in Figure 8c. It can be seen from Figure 9b that the secondary Figure 10 displays the performance of thrust reverser
injection direction is essentially consistent with the curvature under different cascade installation angles β, where FPR is 1.6,
connecting the fan bypass and the cascade window when  jet SPR is 5, and X jet is 80mm. Reverse thrust flow ratio  Frev
is 45°, the fan flow is easier to enter into the cascade window gradually increases with the increment of β, and the increase
under this restraint of the secondary injection with  jet of amplitude is relatively small under larger β. For reverse thrust
45°, the secondary injection can get deeper injection depth into ratio Trev , it nearly remains constant, and then gradually
the fan bypass, the blockage effect of the secondary injection on decreases with the increase of β. To further analyze the impact
the fan flow is strengthened correspondingly. Therefore, the of cascade installation angle β to the reverse thrust
injection depth into the fan bypass is large enough at  jet of effectiveness, Figure11 shows Mach number and streamline
45°, which allows more fan flow to be deflected through the distributions of blockerless cascade thrust reverser under
cascade window and achieves excellent reverse thrust different β. It can be found the fan flow can be deflected by the
effectiveness. blockage of the secondary injection, and discharges from the

7 Copyright © 2018 ASME

cascade window against the incoming flow direction. The
deflection angle of the fan flow is relatively small as SPR is
large in this case, the fan flow enters into the cascade window
with a certain incidence, reverse flow occurs in the front
channel, and the fan flow mainly discharges from the rear
channels of the cascade window depending on the inertia effect.
When β is 30°, the fan flow need a large deflection in the
cascade channel, back flow appears for each cascade channel,
and the corresponding back flow area is large, thus, a big
blockage and constraint effect are imposed on the flow under β (a)β=30°
of 30°. With the increase of the cascade installation angle of β,
the fan flow entering into the cascade window tends to
discharge from the rear cascade channel and the front cascade
channel is completely blocked. Due to the increase of the
effective flow area of the cascade channel, the blocking effect
on the air flow is reduced, and the back flow area in the rear
cascade channel is basically disappeared. Therefore, it is easier
for the deflected fan flow to discharge from the cascade window
and the reverse thrust flow rate increases gradually. In this case,
the FPR is as large as 1.6, the energy of the secondary injection
(b)β=40°
can only deflect part of the fan flow to discharge from the
cascade window, thus, only part of the cascade channel is
needed, and the reverse thrust flow ratio  Frev increases with
the increment of the installation angle β. But for the reverse
thrust ratio Trev , with the increase of β, the angle between the
reverse thrust flow and the engine axis is increased, and the
reverse thrust flow rate only increases slightly, then, the
corresponding component of the reverse thrust is decreased,
resulting in the reduction of the reverse thrust ratio Trev . (c)β=50°

(d)β=60°

FIGURE 10. PERFORMANCE OF THRUST REVERSER

UNDER DIFFERENT CASCADE INSTALLATION ANGLE β

(f)β=70°
FiGURE 11. MACH NUMBER AND STREAMLINE
DISTRIBUTION OF THRUST REVERSER UNDER
DIFFERENT CASCADE INSTALLATION ANGLE β

8 Copyright © 2018 ASME

DOE OF THE BLOCKERLESS CASCADE THRUST TABLE 1. DESIGN SPACE OF RSM
REVERSER Factor Variable range units
From the flow feature and configuration characteristic of FPR 1.3-1.5
the blockerless cascade thrust reverser, it shows that the SPR 3-5
parameters of the secondary flow injection, fan flow and X jet 50-90 mm
cascade window all contribute on reverse thrust flow ratio
 jet 20-60 （°）
 Frev and reverse thrust ratio Trev , the parameters which
β 40-70 （°）
impact the thrust reverser performance include: secondary
δ 5-7
pressure ratio SPR, secondary injection location X jet ,
secondary injection angle  jet , fan pressure ratio FPR, RSM MODEL OF BLOCKERLESS CASCADE
THRUST REVERSER
cascade installation angle β and cascade solidity δ. Based on CFD, all the 54 points are simulated to get
According to the approximate modeling principle, the
detailed procedures of the approximate modeling of the reverse thrust ratio Trev and reverse thrust flow ratio  Frev ,
blockerless cascade thrust reverser are as follows: then quadratic analysis is conducted for Trev and  Frev , and
1) Determine the sample space of parameters study for the
the quadratic regression equations are established. The equation
blockerless cascade thrust reverser, including independent
includes constant term, first term and second term (represents
variables and the range of the parameters. Six variables are
the interaction terms). Stepwise regression method is used to
chosen: FPR, SPR, X jet ,  jet , β and δ, as shown in Table 1. perform the significance analysis of each factor, and the
2) Select DOE method and perform the related experiment significance analysis of each factor after removing the items
design. Here Box-Behnken Design (BBD) with rotatory is which have unimportant influence on Trev is shown in Table 2.
chosen for DOE, whose characteristic is that the distance The value of “Prob > F” can reflect its contribution on
between the other design points and the center point is equal,
except for the center point.
Trev . When the value of “Prob > F” is smaller than 0.05, it
3) Determine the concerned response variables. The denotes the corresponding factor has prominent influence on
response variables in this paper are  Frev and Trev . The Trev . If “Prob > F” is larger than 0.05 and smaller than 0.1, it
corresponding response variables are obtained by performing means the corresponding factor has impact on Trev . It can be
simulation for the samples. seen from Table 2 that the following factors have relatively
4) Ascertain the order of RSM objective functions and
construct the fitting function. In this paper, quadratic significant effect on Trev , first terms include FPR, SPR, X jet ,
polynomial model is chosen as the fitting function.  jet and . The interaction terms include FPR* X jet ,
5) Complete the regress of polynomial coefficient by
standard least-square, and perform significance analysis of each FPR*  jet , SPR* X jet , SPR*  jet , X jet *  jet and X jet *β.
parameter by stepwise regression method, then remove Square items include FPR*FPR, SPR*SPR and X jet * X jet .
unimportant factors and carry out polynomial coefficient TABLE 2. SIGNIFICANCE ANALYSIS OF EACH
regression again, finally execute the fitting examination of the
model, ie R2 examination, which represents the ability of the
FACTOR FOR Trev
response surface function to approximate the real value. When p-value
R2 is larger than 0.8, it denotes the fitting accuracy is acceptable Source Prob > F
and the approximate modeling is applicable, Otherwise, the Model < 0.0001
design space of the selected variables should be adjusted, and A-FPR < 0.0001
the above process from 2) to 5) need be repeated. B-SPR < 0.0001
Combined with the research object of this paper, the design
C-Xjet < 0.0001
space (range of each variable) is determined in Table 1, where
D-αjet < 0.0001
cascade solidity is represented by cascade number. 6 factors
involved and the corresponding variable range are showed, E-  0.0867
based on the principles of the BBD design of experiment, a AC < 0.0001
sample space with 54 points is generated which includes 49 AD 0.0987
factorial experiments and other 5 repeated experiments at the BC 0.0659
center point. And detailed samples with 54 points are not listed BD 0.0544
here due to the too large information of these samples. CD 0.0111
CE 0.0488

9 Copyright © 2018 ASME

 A2 0.0119 of Xjet. Due to the large kinetic energy of the fan flow, it
becomes more difficult for the secondary injection flow to
B2 0.0658
deflect the fan flow. For the first secondary flow injection
C2 0.0945 position, which is in front of the cascade window, the fan flow
R examinations show R2 of
2
Trev and  Frev are 0.9 has been accelerated in the bypass duct, the large momentum of
the fan flow induces the small blockage of the secondary flow
and 0.91 respectively, which denote approximate models can
satisfy the precision demand. Therfore, the response surface injection on the fan flow, and the corresponding Trev is
equation of Trev is as follows: relatively small. When Xjet moves backward, the deceleration of
the fan flow results in the reduction of the momentum of the fan
Trev =7.649－8.446×FPR+0.0439×SPR－0.027× X jet － flow, making the deflection effect of the secondary flow
4.58E-03× jet ＋9.829E-03× ＋0.031×FPR×X jet －6.769E- injection on the fan flow to be increased, that is the
increased Trev can be achieved.
04×FPR× jet － 1.94E-03×SPR× X jet － 2.04E-03×SPR× jet
＋1.37E-04× X jet × jet －1.395E-04× X jet ×  ＋1.969×FPR2
2
＋0.032×SPR2－7.197E-05×X jet (7)
The response surface equation of  Frev is expressed as:
 Frev =6.908－8.127×FPR－0.372×SPR－0.039×X jet －
9.610E-03×  jet ＋ 0.084×  ＋ 0.046×FPR× X jet － 4.493E-
03×FPR×  jet － 0.029×FPR×  － 4.148E-04×SPR× X jet －
1.259E-03×SPR× jet ＋ 4.241E-04× X jet ×  ＋ 1.99×FPR2 ＋
2 FIGURE 12. INTERACTION EFFECTS BETWEEN
0.066×SPR 2－1.215E-04×X jet (8)
FPR AND Xjet
From the obtained response surface equation of Trev and
The interaction effects of other major factors are given
 Frev , the relationship between the thrust reverser performance from Figure 13 to Figure 16. Those figures fully express the
and various parameters are clearly described. It shows that the significant interaction effects among different major factors.
reverse thrust ratio Trev and thrust reverser flow ratio Thus, these significant interaction effects must be taken into
consideration in the design process of the blockerless cascade
 jet  . Among
 Frev are affected by FPR, SPR, Xjet, and thrust reverser system. From the coefficient in Eq. (7) , Eq. (8)
them, FPR, SPR and Xjet are the dominated factors. The cascade and the interaction effect distributions between different factors,
it shows interaction effect of FPR×Xjet and FPR×β should be
solidity δ has unimportant impact on Trev and  Frev , which incorporated into the design.
can be ignored in the design of the blockerless cascade thrust
reverser system.
To further investigate the relationship between Trev ,
 Frev and various factors, the first order of interaction terms
are considered in Eq. (7) and Eq. (8). Figure 12 displays the
interaction effect between FPR and Xjet. It can be seen that an
obvious interaction effect between the FPR and Xjet apperars
for Trev . For example, FPR=1.3, as the position of the
secondary injection Xjet moves backward, the flow area of the
fan flow enlarges due to the opening of the cascade window,
making the fan flow to decelerate, and the corresponding static FIGURE 13. INTERACTION EFFECTS BETWEEN
pressure of the fan keeps on increasing. Then the pressure at the
secondary injection outlet increases, which makes the FPR AND αjet

recirculation zone to be smaller. Thus, Trev quickly decrease

with the increase of Xjet when the FPR is small. While for the
larger FPR of 1.6, Trev slowly increases with the increment

10 Copyright © 2018 ASME

 carried out to seek the best combination of parameters,
achieving the maximum reverser thrust.
For the engine studied in this paper, the optimization of the
blockerless cascade thrust reverser is to determine the optimum
parameter settings under the constraint of  Jrev , with the
optimization target of the maximum Trev . Table 3 shows the
specific requirement of each parameter.
TABLE 3. PARAMETER REQUIREMENT FOR THE
OPTIMIZATION
Lower Upper
FIGURE 14. INTERACTION EFFECTS BETWEEN
Parameter Requirement
limit limit
SPR AND αjet FPR within the limit 1.3 1.6
SPR within the limit 3 5
X jet within the limit 50 90
 jet within the limit 20 60
β within the limit 40 70
δ Equal to 6
 Jrev Equal to 0.1
 Frev within the limit 0 1

FIGURE 15. INTERACTION EFFECTS BETWEEN

Trev maximum
Xjet AND αjet Table 4 displays optimization results and parameter
settings achieving maximum Trev , it can be seen the
maximum Trev can reach 60% of the ideal bypass thrust
under  Jrev limitation of 0.1, the secondary injection flow
ratio  Jrev is larger than 67%, and good effectiveness of the
thrust reverser is achieved. Many parameter settings can be
combined to get the maximum Trev , SPR keeps constant of 5
among these cases, that means the fan flow is easier to be
deflected under the highest allowable secondary pressure ratio
SPR, then large amount of the fan flow goes through the thrust
FIGURE 16. INTERACTION EFFECTS BETWEEN reverser window under the suppression of the secondary
Xjet AND β injection with high momentum, achieving best effectiveness of
the thrust reverser as compared with the other smaller SPR.
TABLE 4. OPTIMIZATION RESULTS WITH DIFFERENT
OPTIMIZATION OF THE BLOCKERLESS CASCADE PARAMETER SETTINGS
THRUST REVERSER N Trev *  Frev
According to the response surface equations of reverse o FPR SPR X jet  jet β
*100
thrust ratio Trev and reverse thrust flow ratio  Frev , the . 100
1 1.48 5.00 65.29 20.00 40.93 58.09 67.12
relationship between the performance of the thrust reverser and
all major influence parameters can be clearly displayed, and the 2 1.48 5.00 66.24 20.00 41.78 57.29 68.52
performance of the thrust reverser also can be evaluated quickly 3 1.49 5.00 63.49 20.00 41.60 57.86 67.95
for one given parameter setting. Furthermore, based on the
obtained response surface equation of Trev and  Frev , 4 1.60 5.00 59.81 20.00 68.31 58.29 95.24
optimization of the blockerless cascade thrust reverser can be 5 1.60 5.00 59.36 21.36 69.00 58.78 94.31
6 1.60 5.00 59.40 21.16 69.38 58.22 97.83

11 Copyright © 2018 ASME

 7 1.58 4.96 57.70 20.02 57.48 58.18 83.15 4) Using the obtained RSM model on the thrust reverser
performance, the design of the blockerless cascade thrust
8 1.55 5.00 62.58 20.93 56.65 58.31 83.36 reverser system can be guided, optimization of the blockerless
9 1.60 5.00 58.33 20.01 62.81 59.56 89.41 cascade thrust reverser can be carried out to seek the best
In order to validate the accuracy of the RSM model of combination of parameters, achieving the maximum reverser
thrust. And performance evaluation can also be quickly
Trev and  Frev , parameters setting 1 and 6 are numerically performed. Under the limitation of the injection secondary flow
simulated, then the predicted Trev and  Frev are compared ratio, about 60% reverse thrust ratio can be obtained by RSM
optimum design.
with the optimization results of RSM, as shown in Table 5. It
can be seen that the relative error of Trev with parameters ACKNOWLEDGMENTS
setting 1and 6 are 1.5% and 1.7% respectively, which indicates The authors would like to express their gratitude for the
that the RSM model of Trev is reliable to be used for financial support of the National Natural Science Foundation of
China (No. 51576163 and No. 51236006).
optimization design and quick performance evaluation of the
blockerless cascade thrust reverser.
TABLE 5. VALIDATION RESULT OF RSM MODEL REFERENCES
[1] Nobel T P. Experimental Thrust Reverser Design with
parameter setting Trev Δ
Trev (%) Computational Analysis. AIAA 94-0021, 1994.
[2] Ferreira S B. Study of the Influence of Aircraft
1 0.5726 1.5%
Geometry on the Computed Flow field During Thrust Reversers
6 0.5675 1.7% Operation. AIAA 2006-3673, 2006.
[3] Hegen G H. Investigation of Aircraft Performance with
CONCLUSIONS Deployed Thrust Reversers in DNW. AIAA 2005-3702,2005.
A blockerless engine reverse system was investigated using [4]He Yan, Liu Youhong. Thrust reverser cascade
RSM focusing on the effect of different geometric and geometrical in take angle, pressure ratio impact on the
aerodynamic parameters on the performance of the thrust performance of the thrust reversers. Science Technology and
reverser system, conclusions are as follows: Engineering, 2010, 12(2).
1) The secondary flow with high pressure is injected into [5]Butterfield J. Integration of Aerodynamic, Structural,
the fan bypass for the blockerless cascade thrust reverser Cost and Manufacturing Considerations During the Conceptual
system, the interaction between the secondary injection and the Design of a Thrust Reverser Cascade. AIAA 2004-282,2004.
fan flow forms the recirculation zone in the bypass duct due to [6] Chen M W. An Experiment and Computational Study of
the inconsistent flow direction between the secondary injection the Blockerless cascade thrust reverser Concept. Master’s
and the fan flow, and generates a blockage effect to the fan flow, Degree Thesis of George Washington University,1996
then forces the fan flow to deflect and discharge from the [7] Yao H, Tweedie J, Hatrick M. Observations on the
cascade window, realizing the reverse thrust. Good performance Design of Cascade-less Thrust Reversers. AIAA 2007-7823.
of the thruster reverser can be achieved under the allowable [8] Gilbert B, Marconi F. Innovative Concept for Cascade
maximum SPR, also with optimum  jet and β existed. Thrust Reverser Without Blocker Doors. AIAA 97-823,1997.
[9]Marconi F, Gilbert B. Computational Fluid Dynamics
2) Fan pressure ratio FPR, secondary flow pressure ratio Support of the Development of A Blockerless Engine Thrust
SPR, secondary injection position X jet and secondary injection Reverser Concept. AIAA 97-3151,1997.
angle  jet , cascade installation angle β play important roles on [10] Asbury C S, Yetter J A. Static Performance of Six
Innovative Thrust Reverser Concepts for Subsonic Transport
the performance of the blockerless cascade thrust reverser Applications. NASA TM-2000-210300, 2000.
system. Among them, FPR, SPR and Xjet are the dominated [11] Hall S, Cooper R, Raghunathan S. Fluidic Flow
factors. The cascade solidity δ has unimportant impact which Control in a Natural Blockage Thrust Reverser. AIAA 2006-
can be ignored in the design of the blockerless cascade thrust 3513.
reverser system. [12] Chen zhu, Shan Yong, Sheng Xigang. Numerical
3) For various factors affecting the performance of the Study on Thrust Reverser Controlled with Secondary Flow.
blockerless cascade thrust reverser system, significant Journal of Propulsion Technology. 2014, 35(9): 1181-1187
interaction effects exist between different two factors. So the [13] Chen zhu, Shan Yong, Zhang jingzhou, et al.
interaction effects between different facotrs should be Numerical Research on a Blockerless Cascade Thrust Reverser.
incorporated into the design of the blockerless cascade thrust Aeronautical Computing Technique. 2012, 42(6): 50-53
reverser system, especially for interaction effects of FPR×Xjet [14] Zhang Yunhao, E Riqitai, Li Xiaoxing. Numerical
and FPR×β. Study for the Effects of Injection Parameters on Flow Fields of

12 Copyright © 2018 ASME

the Blockerless Cascade Thrust Reverser. Theoretical and Blockless Thrust Reverser. Journal of Engineering
Mathematical Foundation of Computer Science, 2011, 164: 1-8. Thermophysics, 2014, 35(8):1538-1541
[15] Wang Zhanxue, Fu pengzhe, Liu Chunyang. [18] Frederic C S, Crossley W A. Investigating Response
Investigation on Flowfield and Impact Parameters of New Surface Approaches for Drag Optimization of a Subsonic
Blockerless Cascade Thrust Reverser. Aeroengine. 2010, 36(2): Trubofan Nacelle. AIAA2000-4797, 2000.
4-8 [19] Ren Luquan. Optimum Design and Analysis of
[16] Shi Jingwei, Wang Zhanxue, Zhang Xiaobo, et al. Experiment (second edition). Bei Jing: Higher Education Press.
Experimental Study on Performance of Blockerless Cascade 2003: 10-58, 246-268, 400-151.
Thrust Reverser. ACTA Aerodynamics Sinica, 2013, 31(6): 763- [20] Wang Yongfei, Wang Chengguo. The Application of
768. Response Surface Methodology. Journal of the CUN (Natural
[17] Zhou Li, Wang Zhanxue, Liu Zengwen. The Influence Sciences Edition). 2005, 14(3): 236-240.
of the Injection Angle of Secondary Flow on Performance of

13 Copyright © 2018 ASME