44th AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit 21 - 23 July 2008, Hartford, CT

AIAA 2008-4588

Inlet/TBCC/Nozzle Integration Concept Design Min Chen1, Hai-Long Tang2, Zhi-Li Zhu3, Hui Ou-Yang4 ,Jin Zhang5 School of Jet Propulsion, Bei Hang University,Beijing, China, 100191

The present paper focuses on the Inlet/TBCC/Nozzle integration concept design for the hypersonic civic aircraft. Firstly, several basic principles were put forward for the Inlet/TBCC/Nozzle integration concept design .Secondly. A set of fast, versatile, and trusted analysis tool was developed which could analyze the design constraints and the matching relationship between the various components among a wide flight range. Assisted by this tool, the design issues of Inlet/TBCC/Nozzle, the multiple-variable control law for an optimized TBCC propulsion system were analyzed .Then a solution for the inlet/TBCC/Nozzle integration concept design was presented. The result shows that, to meet the conflict requirements in the long range from take-off to Mach 5, the turbine based combine cycle propulsion system should apply a variable cycle engine concept via the modulation of 11 variable geometries. It can also be concluded that the solution for the TBCC propulsion system concept could meet the basic mission requirements for the hypersonic airplane.

Nomenclature A8 A9 A8/A8ds BPR CBP dN3 EXNZ FBP FVABI F/Fds Fs H HPCP HPTB LPTB LPT VGV Ma N2zh N1zh MSV RM RJAB RJBP RVABI SFC SMfan

= = = = = = = = = = = = = = = = = = = = = = = = = =

variable exhaust nozzle throat area variable exhaust nozzle exit area normalized A8 bypass ratio common bypass length variation of N3 Ramp exhaust nozzle fan bypass front variable area bypass injector normalized thrust specific thrust altitude high pressure compressor high pressure turbine low pressure turbine variable guide vane of low pressure turbine mach number corrected rotate speed of high pressure compressor corrected rotate speed of fan mode selector valve recirculation margin ramjet afterburner ramjet bypass rear variable area bypass injector specific fuel consumption surge margin of fan

1

Post Doctor, School of Jet Propulsion, [email protected]. Associate Professor, School of Jet Propulsion, [email protected]. 3 Professor, School of Jet Propulsion, [email protected]. 4 Phd. Candidate, School of Jet Propulsion,[email protected] 5 Professor, School of Jet Propulsion, [email protected]. 2

1 American Institute of Aeronautics and Astronautics

Copyright © 2008 by Min Chen. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.

SMc T3 T4 T5 Wfram Wfturbo Wa Wmg W/Wds Wtotal Wturbo Wram δ3 δ6 πt V9

= = = = = = = = = = = = = = = =

surge margin of compressor compressor exit total temperature high pressure turbine inlet total temperature low pressure turbine exit total temperature ramjet afterburner fuel flow turbofan combustion chamber fuel flow air mass flow rate gas mass flow rate normalized total flow engine total air mass flow rate during mode transition air mass flow rate through the turbofan mode air mass flow rate through the ramjet mode wedge angle of N3 Ramp N6 Cowl Lip Ramp angle engine total pressure ratio Nozzle Exhaust velocity

I. Introduction

H (m)

It would be possible to take only 3～4 hours from Beijing to New York once the hypersonic civic aircraft comes into reality whose cruised mach number is 5. The flight mission profile of this aircraft was presented in figure 11,2. Figure 2 also shows its flight envelope and trajectory. From the figure, it can be seen that the flight Mach range of this aircraft is 0～5 and the height range is 0～33km. To fulfill the mission stated above, turbine based combined cycle Figure 1． Flight mission profile of the hypersonic civic aircraft engine would be the most promising propulsion system due to its advantages such as low specific fuel consumption, horizontal take-off ,being very reusable, 35000 highly reliable and etc3. As shown in the figure 3, a co-axial 30000 and tandem configuration was 25000 selected4,5, for compact engine size and weight, as the turbine based combined 20000 cycle engine which consists of turbofan 15000 mode , ramjet mode and turbo/ramjet mode transition . The turbofan engine M a m in 10000 mode operates from take off to Mach 3 M a m ax and the ramjet engine mode operates 5000 (F lig h t tra je c to ry ) from Mach 3 to Mach 5, considering the 0 propulsion efficiency for these two 0 1 2 3 4 5 engine modes. The operation point for M a the turbo/ramjet mode transition was selected at Mach 3, height 21 km. In order to avoid redundant components, these three engine modes share the two Figure 2． Flying envelope and trajectory of the hypersonic civic dimensional mixed compressed inlet, the common bypass, the ramjet afterburner and the two dimensional rectangle convert-divert nozzle. However, despite the advantages of this engine concept, many challenging technology issues should be resolved for an optimized overall propulsion system. At the initial phase of the concept design, Inlet/TBCC/Nozzle integration 2 American Institute of Aeronautics and Astronautics

research would offer a good opportunity to decompose these critical technology issues through close interaction to all the components.

Figure 3． The schematic flow diagram of turbine based combined cycle propulsion system To facilitate the Inlet/TBCC/Nozzle integration concept design, it is critical to develop a set of fast, versatile, and trusted analysis tool. Computational fluid dynamics (CFD) tools can provide high fidelity results for diverse problems, but they are slow, expensive, and require expert attention. The use of appropriate preliminary design tools would reduce cost, increase design confidence, and allow considering the matching restrictions between Inlet/TBCC/Nozzle under a wide range of off-design operating conditions. The present paper focuses on the Inlet/TBCC/Nozzle integration concept design for the hypersonic civic aircraft. SectionⅡ presents the basic principles for the Inlet/TBCC/Nozzle integration concept design .Section Ⅲ describes the modeling methodology and validation of the propulsion system analysis tool. Section Ⅳ discusses the design issues of the TBCC engine, inlet and nozzle. Section Ⅴ analyzes the control law for TBCC propulsion system with multiple variables. SectionⅥ presents the solution of the inlet/TBCC/Nozzle integration concept design. Section VII draws the conclusion.

II. Basic Principles for Inlet/TBCC/Nozzle Integration Concept Design During the concept design of the hypersonic propulsion system, several basic principles should be compromised to develop an optimized solution. The principles considered in this paper are shown below:  High specific thrust as well as low specific fuel consumption was pursued in the case of meeting the engine thrust requirement at the critical operation points shown in table 11,4. (4 sets of TBCC propulsion system are needed for the hypersonic civic aircraft.)  Both the steady and transient performance of the TBCC engine was required to be fine along the flight trajectory.  Turbine inlet temperature over the safe limit as well as over rotate speed of the engine should be avoided to let the engine go smoothly.  Surge margin of the fan and the compressor should be kept above 15%.  All components of the engine were required to work with high efficiency along the flight trajectory.  The three work modes of the TBCC engine should match well not only in structure but also in aerodynamic due to the specific TBCC tandem co-axle configuration in which the inlet, the bypass, the afterburner and the nozzle are shared by the turbofan and the ramjet.  Smooth turbo/ramjet mode transition should be ensured whose basic criterion includes no surge in the fans and the compressors ,no flameout in the combustion, no flow recirculation in the mixing zone and almost constant engine thrust and total engine mass flow rate during the mode transition process .  The hypersonic mixing inlet being designed should be capable of starting in as wide Mach range as possible, low total pressure loss as well as various low inlet drags were required.  The two dimensional convert-divert nozzle should afford high thrust coefficient along the flight trajectory.  The number of the variable geometries adopted for the propulsion system should be as less as possible for engineering feasibility, in the precondition of good integration performance along the trajectory. Also, the adjustment of these variable geometries should be kept within the proper range.

3 American Institute of Aeronautics and Astronautics

Table 1． installed thrust requirement at main operation points of TBCC Mission Profile H (km) Ma Mode Thrust (kN) Take off

0

0

Turbofan

260

Climb

20.7

3

Turbofan

163

Climb

20.7

3

Ramjet

163

Climb

28.3

5

Ramjet

135

Hypersonic Cruise

28.3

5

Ramjet

120

However, what makes the design task so challenging is the strong coupling of all these disciplines. For instance, along the climb trajectory across a wide Mach number spectrum, the mass flow rate afforded by the inlet is required to be equal to the one needed by the TBCC engine in order to reduce inlet spillage and inlet bypass bleed drag. The mass flow rate that the inlet could afford is not only determined by the inlet performance at the design point but also dependent on the control schemes of the moveable inlet ramps at off-design conditions. Also, the mass flow rate required by the TBCC engine is not only dependent on the installed thrust requirements shown at table 1 but also determined by the engine control schemes at off-design conditions. At the same time, the throat area variation of the shared nozzle should adapt to the change of the engine mass flow rate along the trajectory. This gives for tight coupling relation between the inlet /TBCC/ nozzle designs. It can only be adequately investigated by an integral analysis tool that simulates these interrelations.

III. Analysis Tool for TBCC Propulsion System An object-oriented numerical simulation model was created for the concept design of TBCC propulsion system, under the Visual C++ 6.0 software environment. This model assembles 7 function modules on the basis of the C++ class mechanism, whose module chart was presented in Fig 4. From figure 4, it can be seen that TBCC Engine Performance Simulation Module was the hinge, where data communication between these modules could go well. The feature of the seven function modules was presented below:

Figure 4． Module chart of the numerical simulation model for TBCC propulsion system

1) TBCC Engine Performance Simulation Module. It was a 1-D variable cycle TBCC engine model controlled by multiple-variables. It affords a platform suitable for parametric cycle analysis, performance cycle analysis, control law study of the turbofan mode, the ramjet mode and turbofan/ramjet mode transition. The engine model is viewed as a set of interconnected gas components whose performance is described in terms of characteristic maps or empirical formulas. As presented in Figure 5, several major stations are defined along the engine gas path. Each gas component has at least one inlet station and one exit station .At each station, an array of gas conditions in terms of pressure, temperature, velocity, area, mass flow, fuelair ratio, enthalpy, entropy, etc is defined and used to pass information from one component to the next. Beside this, the gas components are coupled together through the balance relationships that are summarized below: • Power-balance equation for each rotor with the rotor inertia term (turbine power = compressor power + parasitic power + acceleration power). • flow compatibility between the connected gas components.

4 American Institute of Aeronautics and Astronautics

• Assurance of static pressure balance at the mixing surface boundary of the duct and core flows.

FVABI

Figure 5． The station number define for the TBCC engine As an example, figures 6 presents flow chart of the turbo/ramjet mode transition6 .It can be seen that the number of the guessed variables are equal to the number of the balance relation function .Once the user input data and the initial value of the guessed variables are given, a multi-dimensional Newton-Raphson iteration technique was used to simultaneously satisfy all the balance relation functions to achieve cycle balance at each operation point. More details about general 0-D engine performance simulation could be seen in literature 7. MSV

R JB P FVABI

CBP FBP A q u a n tity o f c o o lin g a ir A n o th e r q u a n t ity o f c o o lin g a ir

I n le t

Fan

CBT

HPCP

S ta t ic p r e s s u r e b a la n c e

HPTB

LPTB

RVABI

R JA B

F lo w c o m p a t ia b ility

EXNZ

F lo w c o m p a t ia b ility

D y n a m ic w o r k c o m p a t ia b ilit y F lo w c o m p a t ia b ility

D y n a m ic w o r k c o m p a t ia b ilit y G iv e n : e tc.

H , M a, W

m a2 II

, W f b , W fa b , A 3 I I , A 5 I I , A 8 , L P T V G ,

G u e s s e d : W m a 2 I , N l, W m a 2 . 5 I , T 4 , W m g 4 , W m g 4 . 5

Figure 6． flow chart of the turbo/ramjet mode transition   

Particularly, other important features of the engine module are shown below: ， ， ， Component characteristics of the fan, HPCP, LPTB, RJAB were proved by the rig tests 1 2 8 9; The characteristics of the inlet and nozzle are acquired through the 0-1 zoom design method which will be discussed later. Gas property difference, caused by the variation of the gas ingredients, ambient temperature and ambient humidity, was considered; 5 American Institute of Aeronautics and Astronautics



Factors are also taken into account such as the effect of altitude, mach number, extraction and return of the cooling air, power set aside for the aircraft accessories, etc;  Combine operation performance of the five variable geometries can be simulated. In order to evaluate the fidelity of this module, HYPR engine was selected as the reference engine 2,10.The simulation results of reference engine performance & gas path parameters are shown in the table2.In the table, ‘HYPR90’means the data in the column were acquired from the HYYR literature. ‘Prediction’means data in the column were predicted by the engine performance simulation module. The symbol ‘- -’in the table means the data was not found in the literature. The engine performance and gas path parameters at four major operation points are simulated and compared. The definition of the parameters was presented in the nomenclature. From the result, it can be seen that the prediction almost agrees with the HYPR90 data, which can be concluded that the fidelity of this module can meet the requirements of the TBCC engine concept design 2)1-D Hypersonic Inlet Simulation Module. This module could afford one dimensional aerodynamic analysis for the hypersonic variable mix-compressed inlet at the design point as well as at the other off-design points. What really maters is that this module could carry out the 0/1 dimensional zooming design through the data communication with the TBCC Engine Performance Simulation Module, which contributes to addressing matching relation between the inlet and the engine at the Table2. simulation results of reference engine performance & gas path parameters HYPR90 Prediction HYPR90 Prediction HYPR90 Prediction HYPR90 Prediction H(km)

0

0

10.7

10.7

18.3

18.3

20.7

20.7

Ma

0

0

0.95

0.95

2.5

2.5

3

3

BPR

0.83

0.86

0.7

0.71

0.7

0.7

0.94

0.95

πT

10.4

11.3

--

10.6

5.9

5.8

4

3.9

T3 (K)

622

640

536

556

847

854

936

933

T4 (K)

1710

1710

1340

1340

1710

1710

1873

1873

T5 (K)

--

1260

--

963

1392

1392

1410

1413

V9(m/s)

550

559

582

604

1045

1064

1095

1115

Fs(N.s/kg)

555

557

304

320

323

330

220

230

0.81

1.05

1.02

1.54

1.45

1.85

1.66

SFC(kg/h/daN) 0.86

concept design process. As to the aerodynamic analysis, it solves for the flow conditions from free stream, through the hypersonic mixed compression ramps, across the terminal normal shock zone and the subsonic diffuser to the engine face. It calculates the shock locations, the capture stream tube, the additive drag, the bleed-off drag of the inlet boundary layer, the bypass drag. As Shown in the Figure 7, the gas path of the inlet was divided into four parts: external oblique shock zone, internal oblique shock zone, terminal shock zone and the subsonic diffuser zone. In the external oblique shock zone , pressure recovery for a chosen geometry can be calculated using one dimensional compressible flow equations given in Eq. 1 ~ Eq. 3, where

Ma f

is the upstream Mach number,

M ab is the downstream Mach number,

 is the ramp angle,  is the shock angle, k is the ratio of specific heats,  is the pressure recovery across the

oblique shock, P is the total pressure with the "f" conditions upstream of the shock and the "b" conditions downstream. On the basis of the Oswatitsch’s theory, to minimize the shock total pressure loss, Eq. 4 also presents the relationship among the serial external oblique shocks, where subscript n indicates the upstream parameter of the shock.. For a given first ramp angle, upstream Mach number and ratio of specific heats, the total pressure loss and downstream Mach number are determined. The next oblique shock angle could also be calculated according to the

6 American Institute of Aeronautics and Astronautics

eq. 4. The downstream Mach number and the second oblique shock angle then become the inputs for the second ramp conditions, and so on along all the ramps.

tg 

M a 2f sin 2   1  2  k 1   2  M a f  2  sin    1 tg     

(1)

k

  k  1 M a 2f sin 2   k 1   2   k  1 M a 2f sin 2   Pb    1 Pf k  1  k 1  2k 2 2  k  1 M a f sin   k  1 

M ab2 

M a 2f 

2 k 1

2k M a 2f sin 2   1 k 1



M a 2f cos2  k 1 2 M a f sin 2   1 2

M a1 sin 1  M a 2 sin  2  ......M a n 1 sin n 1

(2)

(3) (4)

It should be noted that above a certain ramp angle the shock wave is no longer oblique but becomes normal and detached. The corresponding maximum ramp angle can then be determined from Eq. (5)11. It is necessary to check and avoid this condition due to the high recovery losses associated with normal shock waves.

 max = (-4.68×10-5Ma f 5+7.905×10-3 Ma f 4-5.32×10-2 Ma f 3-2.633×10-2 Ma f 2+0.7725 Ma f -0.7609) ×57.3

(5)

Figure 7 Gas path schematic chart for the hypersonic mixed-compression inlet Supersonic Compression Ramps In the internal oblique shock zone, the downstream parameters of the final external ramps will be the input of the first internal ramp conditions. In the case of internal reflecting oblique shock, the same equations are again used with the ramp angle replaced by the flow turning angle necessary to bring the flow parallel to the wall. In the terminal shock zone, the downstream parameters of the terminal shock could be calculated through the Eq. 6 ~ Eq. 8, where  is the pressure recovery across the terminal shock,  is the static pressure ratio across the terminal shock,  is the ratio between the total pressure and static pressure with certain Mach number.

7 American Institute of Aeronautics and Astronautics

  k  1 M a 2f P   b  Pf  2   k  1 M a 2f

 

k

 k 1  

1

k  1  k 1  2k 2  k  1 M a f  k  1 

(6)

2 Psb 2kM a f  (k  1)  Psf  k  1

(7) k

 (k  1) M a 2  k 1 P   1   Ps  2 

(8) In the subsonic diffuser zone, the calculation of total pressure recovery coefficients across this zone is based on subsonic inlet studies given in Ref. 12. As shown in Fig. 8, typical geometric parameters are used to calculate the recovery loss coefficient, where he is the height of the engine face, hth is the throat height, and l is the length of the duct from the throat to the engine face. The geometry of the duct provides the equivalent conical diffuser angle

 d given by the following equation 9:

 h  ht   d  tan 1  e   l 

(9)

Figure 8 Schematic chart of the inlet diffuser zone As shown in Eq. 10~ Eq. 12, the K1 and K2 coefficients represent the effects of flow separations and boundary layer growth respectively. The Kn is the combination of the K1 and K2 coefficients .Given the throat Mach number and the Kn coefficient, the pressure recovery across the subsonic diffuser zone could be calculated through Eq. 12 .Then,the inlet recovery is the product of the pressure recoveries across all the shocks and the total pressure ratio of the subsonic diffuser.

K 1  0.022 d

 0.066 d  0.44 K2 

 d  10  d  10

(10)

0.1412 tan  d

 h K n  K 1 1  t  he

(11) 2  h    0.02 K 2 1   t   he 

  

Pd  1.0  0.5M ath K n Pt 8 American Institute of Aeronautics and Astronautics

2

  

(12)

(13)

The use of shocks for flow compression inevitably leads to boundary-layer interactions. These interactions can produce unsteady shock movements and flow instabilities, which will reduce propulsion performance. A common way to govern this negative effect is to bleed the boundary layer air. The flow for bleed could be estimated through the Eq. 14, Where

 bl named bleed coefficient is the ratio between bleed flow and the total inlet capture flow,

Ma

is the free stream Mach number. This equation is derived from curve fits of experimental data based on representative hypersonic mixed compression inlet13.In addition, bleed flows are accompanied with momentum losses which will produce inlet bleed drag. The bleed drag coefficient The definition of the pressure,

CBL

could be estimated through the Eq. 15.

CBL is presented in Eq.4.16, where DBL is the bleed drag, p0 is the free stream static

A01 is the inlet capture area.  BL  0.03  0.02( M a  1) CBL   BL  2.31  0.64 M a  0.056 M a 2 

(14) (15)

C BL  2 DBL /(kMa 2 p0 A01 )

(16) Besides the bleed drag, the inlet drag includes spillage drag and bypass drag due to dumping of inlet air. Similarly to Eq. 15, the bypass drag coefficient could be calculated through Eq. 4.17, where coefficient,

 BP is the dumping flow coefficient.

f BP is the correct

CBP  f BP  BP  2.31  0.64M a  0.056 M a 2 

(17) The spillage drag includes supersonic spillage drag and subsonic spillage drag. The difference of these two kinds of drag depends on the location of the normal shock. When the normal shock was pushed outside the cowl lip, the drag was called subsonic spillage drag, whereas was called supersonic spillage drag. The supersonic spillage can be determined by integrating the static pressure ratio along the capture stream tube. The subsonic spillage coefficient C spill can be interpolated through Fig.8, where data of the representative hypersonic inlet13.

 is the inlet flow coefficient. Fig.9 was based on the experimental

9 American Institute of Aeronautics and Astronautics

M a = 1 .2 M a = 1 .0 M a = 0 .8 M a = 0 .7 M a = 0 .6

1 .0

0 .8

0 .6

C sp ill 0 .4

0 .2

0 .0

0 .0

0 .2

0 .4

0 .6

0 .8



Figure 9 Subsonic spillage coefficient of the hypersonic mixed compression inlet 3)1-D Hypersonic Nozzle Simulation Module. Similar to the above module, this module could afford one dimensional aerodynamic analysis for the hypersonic two dimensional convert-divert nozzle with variable geometries at the design point as well as at the other off-design points. It could also perform the 0/1 dimensional zooming design through the data communication with the TBCC Engine Performance Simulation Module. Detailed algorithm for this module was given in reference14. 4) Multi-goals Optimization Module. It could work out the Multi-goals optimization problems encountered in the Inlet/TBCC/Nozzle integration research .Instead of searching for the unknown optimized solution directly, definite optimized value for each goal is set firstly. Then the Multi-goals optimization problem is converted to the problem with n variables and n sets of compatible equations. For example, to guarantee smooth turbofan/ramjet mode transition, optimization objects during the transition process should include: Total flow rate to be kept almost constant. It can be converted to the equation Z0=(Wtotal-Wbefore)/ Wbefore=0, where Wtotal is the total mass flow rate during the mode transition, Wbefore is the mass flow rate before the mode transition. Engine thrust to be kept almost constant. It can also be converted to the equation Z1=(Ftotal-Fbefore)/ Fbefore=0, where Ftotal is the total thrust during the mode transition, Fbefore is the total thrust before the mode transition. Recirculation margin at the front mixed injector zone to be kept positive. It can be converted to the equation Z2=(Ps2I-Ps33)/ Ps2I=3％、5％、7％、9％… ,where Ps2I is the static pressure at the ramjet bypass outlet，Ps33 is the static pressure at the fan bypass outlet. Almost no total pressure difference between the two streams at the back mixed zone. It can be converted to the equation Z3=(P5-P16)/ P5=0 ,where P5 is the total pressure at the low pressure turbine outlet，P16 is the total pressure at the common bypass outlet. Then, the optimization problem stated above is converted to 4 sets of compatible equations which are presented in Eq. 18. As shown in Eq. 19, four unknown variables are chosen to solve the compatible equations, where X0 is the normalized area of the front variable area bypass injector, X1 is the normalized area of the rear variable area bypass injector, X2 is the normalized area of exhaust nozzle throat area, X3 is the normalized ramjet fuel flow rate.

10 American Institute of Aeronautics and Astronautics

Z0  0 Z1  0 Z 2    0,   3%,5%,7%,10％...

(18)

Z3  0

X 0  FVABI / FVABI ds X 1  RVABI / RVABI ds X 2  A8 / A8 ds

(19)

X 3  W fram / W framds Thus the optimized problem with 4 variables and 4 compatible equations could be solved through the multidimensional Newton-Raphson method whose detailed algorithm was given in reference 7[pp.273]. 5）1-D Size Calculation Module for TBCC Engine Module. This module aims at carrying out the gas path of the TBCC engine based on 1-D aerodynamic theory and empirical strength calculation methods15. In order to calculate the radial and axial size of the main components, two kinds of information were required. One kind of the information is the empirical parameters such as the thickness and the diffusion factor of the rotating components .The other kind is the output of TBCC Engine Performance Simulation Module, which includes performance parameter and working parameters such as total pressure, total temperature, mass flow rate, pressure ratio, of the main engine stations at critical operation points . The operation points chosen here are the ones whose aerodynamic load or mechanical load is the highest among the flight regimes. The calculated components include fan, compressor, combustion chamber, common bypass, high pressure turbine, low pressure turbine, mixer, ramjet bypass, ramjet burner, shared exhaust nozzle. The shared inlet was too complicated to be calculated in this module due to its integration with the airframe. . 6）Weight Calculation Module for TBCC Engine Module. Once the engine flow path is defined, this module could compute the various components of TBCC by using the density of the material and predicted component volumes. Several key factors were considered in components’ weight calculation, such as allowable material stress, maximal allowable temperature, stage load and maximal allowable rotator speed. Weight adders based on empirical data were applied to account for additional features such as variable geometries. More details about the algorithm could be seen in reference15 7) Airplane performance simulation module. This module could simulate the flight performance along the trajectory through iterative calling the related modules. As shown in Fig.1, flight mission is broken into 7 segments: take off, climb, hypersonic cruise, descend, subsonic cruise and land. Each segment was defined in detailed parameters such as initial and final Mach number and altitude. Take take-off segment for example, it was divided into 200 calculated steps. To make the module work smoothly, two kinds of information were required. One kind is about the aircraft, which can be set at first step, including range, take off gross weight, total fuel weight, wing area, lift-drag polar table, load factor, specific excess power, climb time and so on. The other kind is the TBCC engine installed performance data acquired from the interaction between the inlet, TBCC engine and nozzle modules at each calculated step. Then the flight performance could be calculated on the basis of the basic flight dynamic equations shown in reference 16. The output of the first step would be the input of the next calculated step and so on for all the other steps. Similarly, the flight performance of the other segments could be calculated one by one. Then the overall flight performance data such as total fuel consumption, total time-consumption could be got after the last segment is calculated.

IV. Design issues for the TBCC Propulsion System TBCC propulsion system integration study needs strong interrelation between the inlet, TBCC engine and the nozzle concept design. However, several steps must be performed in series since results of one are the input of the next step. According to the thrust requirements at critical operation points shown in table 1, it could be the first step to analyze the TBCC engine thermodynamic cycle characteristics. 1) Design issues for the TBCC Engine Two design points were selected to guarantee the TBCC engine to work efficiently along the wide range (Ma =0～5,H=0～33km).One was for the turbofan mode and the other was for the ramjet mode. The design point of the 11 American Institute of Aeronautics and Astronautics

turbofan mode was selected at height 20.9km and Mach 3 both considering the rigor thrust requirements at Mach 3 and the stability of the turbofan/ramjet mode transition. Table 3 presents the design parameters and the performance of the turbofan mode. As shown in the table 3, major cycle parameters include bypass ratio, fan pressure ratio, high pressure turbine inlet temperature and the ramjet afterburner outlet temperature. From the cycle analysis, low bypass ratio and high pressure turbine inlet temperature contribute to high specific thrust. However, low bypass ratio means large exhaust velocity which aggravates the noise pollution at take off. Thus the bypass ratio was 0.7 after the tradeoff between high specific thrust and acceptable noise pollution. 2000 Kelvin degree was selected for the high pressure turbine inlet temperature considering the current development of the turbine material and cooling technology. The fan pressure ratio is 1.3 to avoid flow recirculation at the front variable area bypass injector mixed zone(named station 3III shown in fig.5)during the turbofan/ramjet mode transition. Due to the high inlet total temperature 604k at design point, the total pressure ratio should be low to add more heat to the gas with the limit of high pressure turbine inlet temperature 2000k. So the compressor pressure ratio was 3.1. Thrust augmentation with afterburner burning contributes to reduce TBCC engine size since it can produce more specific thrust. However, the penalty in increase in specific fuel consumption is heavy. Another negative effect is that the noise pollution will be aggravated since higher exhaust velocity is required to meet the same thrust requirement at take off with the reduced engine size. Thus the ramjet afterburner outlet temperature was 1459k after the tradeoff between engine size, specific fuel consumption and noise pollution. The other parameters for the engine components were set according to the present components design level. It should be noted that the inlet pressure recovery coefficient here is an initial value to let the thermal cycle analysis go smoothly. It can be got later through the inlet 0/1D zoom design. Table 3 Design parameters and the performance of the turbofan mode (at design point)

Mach Altitude( km) Bypass Ratio

3 20.9 0.7

0.84 0.995 0.06

1.3 3.1 2000

Compressor Efficiency Burner Efficiency Burner Total Pressure Loss Coefficient High Pressure Turbine Efficiency Low Pressure Turbine Efficiency Mass Flow Rate kg(s)

Fan Pressure Ratio Compressor Pressure Ratio High Pressure Turbine Inlet Temperature (K) Ramjet Afterburner Outlet Temperature (K) Inlet Pressure Recovery Coefficient

1459

Uninstalled Thrust (KN)

179

0.8

uninstalled Specific Fuel Consumption (kg/daN/h)

1.65

Fan Efficiency

0.81

0.85 0.88 350

The design point for the ramjet mode was set at the hypersonic climb operation point, Mach 5 and altitude 28.3km, since the hypersonic aircraft spent most of its time at Mach 5 cruise segment. From the thermal cycle analysis, high ramjet afterburner outlet temperature contributes to high specific thrust and low specific fuel consumption. Thus the ramjet afterburner outlet temperature was set 2100k considering the present material and cooling technology. The inlet pressure recovery coefficient was set 0.52 based on SNECMA hypersonic inlet experiment data given in reference17. The other design parameters for the components were set according to the present component design level. Table 4 Design parameters and the performance of the Ramjet mode (at design point)

Mach Altitude(km) Bypass Ratio Fan Pressure Ratio Compressor Pressure Ratio High Pressure Turbine Inlet Temperature (K) Ramjet Afterburner Outlet Temperature (K) Inlet Pressure Recovery Coefficient Fan Efficiency

3 20.9 0.7 1.3 3.1 2000 1459 0.8 0.81

Compressor Efficiency Burner Efficiency Burner Total Pressure Loss Coefficient High Pressure Turbine Efficiency Low Pressure Turbine Efficiency Mass Flow Rate kg(s) Uninstalled Thrust (KN) uninstalled Specific Fuel Consumption (kg/daN/h)

12 American Institute of Aeronautics and Astronautics

0.84 0.995 0.06 0.85 0.88 350 179 1.65

2) Design issues for the two dimensional mixed compressing hypersonic inlet A two dimensional mixed compressing hypersonic inlet with variable ramps was chosen for the hypersonic civic aircraft since good matching with the TBCC engine is required along the wide Mach range from take off, subsonic section, transonic section, supersonic section to hypersonic section. The inlet design point was selected at Mach 5 to pursue favorable propulsion installed performance at hypersonic cruise section because the aircraft spends most of the time at this section. As shown in the figure 10, the aircraft fore body was integrated with the inlet precompression ramps so that it can provide enough space to produce more oblique shock. It can also be seen that three outer oblique shocks and three inner oblique shocks are designed before the terminal shock at the design point. Also, the outer shock was designed to be focused on the cowl lip to avoid inlet spillage drag. The inlet throat Mach number is designed around 1.4 to ensure that the terminal shock could be located stably at the diffuser zone not far away from the inlet throat. However, throat Mach number above 1.4 is no good for the inlet pressure recovery because the higher the Mach number before the terminal shock, the higher pressure loss after the terminal shock. .The total ramp angle for the outer oblique shock was designed to be around 16o considering the trade off between total pressure recovery and cowl wave drag. The inlet design process was shown in figure 11 and its design parameters were shown in table 5.

Figure 10． Inlet shock configuration at the design point

Figure 11 Flow chart for the 2-D hypersonic mixed compression inlet design process. 13 American Institute of Aeronautics and Astronautics

Table 5 Design parameters of the two dimensional hypersonic mixing compressed inlet (at design point)

Design Point

H=28.3km,Ma=5

Mass Flow Rate

352kg/s

(boundary

layer

bleed

Inlet Throat Mach

1.4

Inlet Total Length

17.5m

(diffuser included)

flow included) Max capture area

10.2m2

Oblique shock number

Inner shock number:3 Outer shock number:3

0

δ1

4.69

δ3

5.550

δ2

5.170

Pressure Recovery Coefficient

0.517

3) Design issues for the two dimensional rectangular convert-divert nozzle The design point for the two dimensional rectangular convert-divert nozzle was selected at Mach 5, altitude 28.3km for the aircraft will spends most of the time at the hypersonic section. The design parameters of the two dimensional rectangular convert-divert nozzle calculated by 1-D hypersonic nozzle simulation module. As shown in the table 6, the area of the A8 was 0.849m2 considering the mass flow matching with the TBCC engine at this design point. The area of A9 was 6m2 far away from the required full expansion area 13.6m2, considering the trade-off between the weight penalties due to full expansion and the thrust loss caused by inadequate expansion. It can be seen that thrust coefficient was still kept at 0.9616 despite the thrust loss due to nozzle inadequate expansion. Table 6 Design parameters of the two dimensional rectangular convert-divert nozzle (at design point)

Altitude

28.3km 2

Area of A9

6m

Variable Geometries

A8、A9

Mach

5

Area of A8

0.848 m2

Thrust Efficiency

0.9616

V. Control Law Study for TBCC Propulsion System with Multiple Variables When we come to the co-axial turbine based combined cycle propulsion system with the tandem layout, its shared components such as inlet, common bypass, ramjet burner and nozzle will make the configuration more compact in size and weight. However, this advantage will be weakened by the weight penalty of the variable geometries adopted in the configuration to guarantee the propulsion system work stably and efficiently in the other flight regimes as well as the design point. Thus, it is pivotal to analyze the necessary amount of the variable geometries and their function through inlet/TBCC /nozzle integration study to avoid redundant weight penalty. 1) Variable geometry related to TBCC Engine Among the issues for the TBCC propulsion system with the tandem layout, stable turbofan/ramjet mode transition is critical to make the concept feasible. The turbofan mode works from take off to Mach 3 and the ramjet mode works from Mach 3 to Mach 5. Mach 3 was the operation point for both the turbofan/ramjet mode transition and the reverse mode transition. Smooth mode transition can be converted into the resolution for the multiple objects stated below:  How to keep the total flow rate and the engine thrust almost constant.  How to keep enough surge margin for the fans and the compressors  How to keep the recirculation margin positive. The definition of the recirculation margin RM was presented in Equation 20, where P2IIs means the static pressure of the ramjet bypass intake; P3IIs means the static pressure of the fan bypass exit (shown in fig. 12).A positive recirculation margin indicates a positive flow potential across the mode selector valve and ensures no recirculation4. 14 American Institute of Aeronautics and Astronautics

RM = (P2IIs-P3IIs)/ P2IIs

(20)

Figure12 Definition of Recirculation Margin

To meet the requirements stated above, the transition process was divided into four steps which were presented in Figure 13. Aside from the turbofan combustion chamber fuel control and the ramjet burner fuel control, four variable geometries were needed in the process, whose function was presented below:  MSV is used to modulate the mass flow rate between the turbo mode and the ramjet mode. It is closed to lead all the engine inlet air to the turbofan engine when the turbo mode operates. During mode transition and the ramjet mode, MSV is open to allow the air to flow into the ramjet bypass.  FVABI is designed to control the static pressure of the fan bypass exit via modulating the FVABI area in order to avoid the flow recirculation to the ramjet bypass.  RVABI is used to optimize the fan operating line via modulating the flow static pressure of the common bypass exit.  A8 is designed to satisfy the different flow requirements in the wide range of the flying envelope

Figure 13 Turbofan/Ramjet mode transition process

Although the angle of the low pressure turbine variable guide LPT VG was kept at -50 position during the mode transition process, it also played an important role in influencing the flow distribution between the bypass and the core engine. The angle of LPT VG includes 3 kinds of position: 50 position, 00position and -50 position. It is opened at 50 position to allow more flow into the HP compressor, thereby meet the rigor thrust requirements at high altitude and speed operations. At take off, it is closed at -50 position to increase the bypass ratio and lower the exhaust velocity for noise reduction. The figure 14 presents the prediction of the take-off operation point at the compressor map by LPT VG variation, where πHPC means the compressor pressure ratio, qmazhHPC means the compressor 15 American Institute of Aeronautics and Astronautics

corrected mass flow rate. It can be seen that the corrected compressor rotor speed at take off operation point was decreased from 102% to 93.1% when the LPT VG varied from the closed position to the open position. This variation indicates it will contribute to reduce the noise pollution by decrease the exhaust velocity. .

6 .5 6 .0

Take off(LPT-VG Open)

5 .5 5 .0

HPC

4 .5

Take off(LPT-VG Closed)

4 .0 3 .5

Mach 2.5(LPT-VG Open)

3 .0

103%

2 .5

98%

94%

2 .0

89% 78%

1 .5 40

45

50

55

60

81% 65

70

75

80

85

90

95

100

105

q m azhH P C

Figure 14 The prediction of the take-off operation point at the compressor map by LPT- VG variation

2) Variable geometries related to the hypersonic inlet. To efficiently operate over a large flight envelope the shared hypersonic inlet should employ variable geometries to make a compromise between low drag and high performance. This compromise can be converted into the resolution for the multiple objects stated below:  How to keep the inlet captured mass flow rate pass through the inlet throat without air dumping and subsonic spillage over a large flight envelope. If the inlet throat area was kept constant as the one at Mach 5, it would be too small to meet the inlet captured flow requirements along the flight trajectory. Fig. 15 presents the difference between the inlet captured flow Wa01 and the mass flow capability Wath, with fixed inlet throat area along the flight trajectory. It can be seen that this difference even reaches 5 times at Mach 5, Height 17km operation point. Without variable inlet throat area, this flow mismatch will induce either serious air dumping drag or serious subsonic spillage.  How to keep the inlet throat Mach number around 1.3 at as wide flight Mach range as possible. As an important symbol to monitor the location of the terminal normal shock, the inlet throat Mach number around 1.3 means that the terminal normal shock is more liable to locate stably near the downstream of the inlet throat even with small perturbation by the TBCC engine. Also, the inlet throat Mach number around 1.3 could avoid high pressure loss behind the terminal normal shock.  How to keep the mass flow rate afforded by the inlet match the one that the engine needed in order to reduce the possibility of air dumping and subsonic spillage along the climb trajectory.

16 American Institute of Aeronautics and Astronautics

W a01 W ath

450 400 350

2 8 .3 k m

Flow(kg/s)

300

Mass flow Rate difference :5 Times!

250 200

2 5 .9 k m

150 100 50

2 0 .9 k m 17km 2 .0

2 .5

3 .0

3 .5

4 .0

4 .5

5 .0

Ma

Figure15 Difference between Wa01and Wath with fixed inlet throat area

To meet these requirements, the optional variable geometries include: wedge angle of N1 Ramp δ1, wedge angle of N2 Ramp δ2, wedge angle of N3 Ramp δ3, The length variation of N3 Ramp dN3, N6 Cowl Lip Ramp angle δ6, whose definition and function were shown below:  The wedge angle of N3 Ramp δ3. The definition of δ3 was demonstrated in figure 16. This angle could be varied through the turning of the N3 Ramp ( shown in figure 17.) The turning of N3 Ramp was designed to meet different requirements of the flow deflection and the inlet throat area along the wide flight regimes( Ma=0～5). It should be noted that the definition and the function of δ1 and δ2 are similar to δ3 (shown in figure.16).  The length variation of N3 Ramp dN3. As shown in Figure 18, the nether half of N3 Ramp could slide along the runner of the upper half. This slide motion could change its total length. The length variation could control the location of the terminal shock through changing the position of inlet throat at different Mach regimes. At lower supersonic Mach range, the position of inlet throat moved forward to reduce unnecessary inner oblique shocks; while at higher supersonic/hypersonic Mach range, it moved rearwards to organize more necessary inner oblique shocks. In this way, the flow Mach number at the inlet throat section will be kept around 1.2～ 1.4 , which will contribute to the stabilization and the strength reduction of the terminal shock. It is also an effective way to control the inlet throat area by varying the length of the N3 Ramp, which helps to control the mass flow through the inlet throat section.  N6 Cowl Lip Ramp angle δ6. Figure 19 gives the definition of δ6. The angle could be varied through the turning of the N6 Ramp ( shown in Figure 20.) .δ6 was positive when Ramp N6 turn toward N3 Ramp whereas it was negative. The turning of N6 Ramp was designed to meet different requirements of the flow deflection and the capture mass flow rate along the flight trajectory.

17 American Institute of Aeronautics and Astronautics

Figure 16 The schematic diagram of the inlet ramps and the wedge angles

Figure 17 The schematic diagram of the turning of Ramp N3

Figure 18 The schematic diagram of the length modulation of Ramp N3

Figure 19 The definition of δ6

18 American Institute of Aeronautics and Astronautics

Figure 20． The schematic diagram of the turning of Cowl Lip Ramp N6

Aside from the multiple objects stated above, the concept inlet with variable geometries must be a design with as little variability, complexity, and weight as possible. Thus firstly single variable geometry was adopted in the inlet to solve the multiple-objects optimization problem. Once satisfactory result could not be found for the optimization problem, double variable geometries were then analyzed for the optimization problem and so on for the optimization problem with more than 2 variable geometries. The inlet 0-D/1-D zoom design method was applied to search the resolution for the optimization problem. Figure 21 presents the inlet 0-D/1-D zoom design flow chart:  Firstly, the regulating law of the inlet variable geometries is the input of the 1-D hypersonic inlet module and the module begins to work;  Secondly ,the calculated inlet pressure recovery coefficient σin is the input of the 0-D TBCC engine performance simulation module and the module begins to work;  Thirdly, the multi-goals optimization module starts to work after receiving the required information from the 1-D hypersonic inlet module and the 0-D TBCC engine performance simulation module;  Then the multi-goals optimization module outputs the information to indicate whether the solution was convergent. If the solution was convergent, the design process will stop whereas the divergence information will help to regulate the inlet variable geometries and the design process goes once again. The simulation result shows that neither the optimization problem with single variable nor the one with double variables could get the satisfactory solution. It also shows that 3 variable geometries including δ3, dN3, δ6 were needed to search for the solution for the multiple objects. The figure 22 shows the control law of the inlet variable geometries along the flight trajectory, where dN3/LN3 is the normalized length variation of the N3 Ramp. The figure 28 presents the solution for the multiple objects optimization problem with 3 variables, where Wa01/Wath means the ratio between inlet captured mass flow and mass flow capability passing through the inlet throat area, Wath/Waeng means the ratio between mass flow capability passing through the inlet throat and the engine required mass flow. From the figure 23, it can be seen that From Mach 2 to Mach 5 along the climb trajectory, the Wa01/Wath and the Wath/Waeng were kept at around 1, which indicates that the flow match between the inlet and engine goes well and no bypass dumping air was needed. The figure also shows that the inlet throat mach was kept around 1.2～1.4, which indicates no subsonic spillage, due to normal shock pushed outside the inlet , will occur. However, it should also be noted that subsonic spillage will inevitably occur at the transonic flow condition ( Mach 1.0～ 1.8)because the outer oblique shock becomes normal due to the unsuitable wedge angle.

Figure 21. the inlet 0-D/1-D zoom design flow chart

19 American Institute of Aeronautics and Astronautics

0

3( )

dLN3/LN3(%) 0 6( )

32 24 16 8 0 2.0

2.5

3.0

3.5

4.0

4.5

5.0

Ma Figure 22 Control law of the inlet variable geometries along the flight trajectory

2.0

Wa01/W ath Math Wath/W aeng

1.8 1.6 1.4 1.2 1.0 0.8 0.6 2.0

2.5

3.0

3.5

4.0

4.5

5.0

Ma

Figure 23 Solution for the multiple objects optimization problem with 3 inlet variable geometries

p r e d ic te d  in

1 .0

Oner experiment curve fit

0 .8

 in

0 .6

0 .4

0 .2

2

3

Ma

4

5

Figure 24 σin prediction under the inlet 0-D/1-Dzoom design along the trajectory.

20 American Institute of Aeronautics and Astronautics

The Figure 24 presents the inlet pressure recovery coefficient σin prediction under the inlet 0-D/1-D zoom design along the flight trajectory. From the figure, it can be seen that above Mach 3, the prediction agrees well with the experiment curve fit data given in the reference17; under Mach 3, the prediction was a bit smaller than the experiment curve fit data. 3) Variable geometries related to the hypersonic nozzle. The variable geometry related to the 2-D rectangle nozzle includes the variable nozzle throat area A8 and the variable nozzle exit area A9. As is stated above, A8 is designed to satisfy the different flow requirements in the wide range of the flying envelope. A9 is designed to avoid thrust loss due to either overexpansion or incomplete expansion among a wide range of available nozzle expansion ratio from 3 to 300. However, trade-off should be considered between thrust loss and serious weight penalty due to the required nozzle throat area for full expansion at high flight Mach number. Similar to the inlet 0-D/1-D zoom design stated above, the normalized thrust at different fixed area of A9 was predicted through iteratively calling the 1-D hypersonic nozzle simulation module and 0-D TBCC engine performance simulation module. Figure 30 presents the simulation result, where F/Ffulexp means the ratio between the thrust with constant area of A9 and full expansion thrust, δF/Ffulexp means the ratio between the thrust loss with constant area of A9 and full expansion thrust. From figure 25, it can be seen that when the area of A9 was kept constant, large thrust loss was inevitable due to either incomplete nozzle expansion or nozzle over expansion; when the fixed area of A9 was 8 m2, 17% of thrust loss occurs at take off due to nozzle overexpansion: however, when the fixed area of A9 was 1.6m2, the thrust loss even reaches 39.1% at Mach 5 due to incomplete nozzle expansion. Thus it is necessary to vary the nozzle exit area along the flight trajectory to avoid unacceptable thrust loss. Figure 26.presents the control law for A9 and its corresponding nozzle thrust coefficient prediction along the flight trajectory. From the figure 26, it can be seen that along the flight trajectory, the thrust coefficient was kept around 0.97 with the shown A9 control law.

1.00 0.95

F/Ffulexp=17%

0.90 0.85 0.80

F/Ffulexp=39.1%

F/Ffulexp 0.75

A9=8m

2

A9=6m

2

0.70

A9=3m

2

0.65

A9=2m

2

0.60

A9=1.6m

2

0.55 0

1

2

3

4

5

Ma

Figure 25 Normalized thrust prediction at different fixed area of A9 along the climb trajectory 1.1

CF

1.0 0.9 0.8 0.7 0.6

A 9 /A 9m ax 0.5 0.4 0.3 0

1

2

3

4

5

Ma

Figure 26 Control law of A9 and its corresponding nozzle thrust coefficient along the flight trajectory 21 American Institute of Aeronautics and Astronautics

VI. Solution for the inlet/TBCC/Nozzle integration concept design、 As shown in the table 7, the basic size and weight of the TBCC engine was calculated after the inlet/TBCC/Nozzle integration study. Except the inlet and the nozzle, the total length of the TBCC engine is 9.19 meters and the total weight of the engine is 9391.kilogram. Two stages are needed for the fan and five stages are needed for the compressor. The turbine includes one high pressure turbine stage and one low pressure turbine stage. Table 7 The basic size and weight of the TBCC Engine

Total Length

9.19 m

Fan Stages

Two

Max Diameter

2.76 m

Compressor Stages

Five

Total Weight

9391 kg

Turbine Stages

One HP Turbine Stage

(Inlet and Nozzle Excluded)

(Inlet and Nozzle Excluded)

+ One LP Turbine Stage

To guarantee the TBCC propulsion system work stably and efficiently along the wide flight trajectory, 11 variable geometries were needed in the concept. Besides the turbofan fuel control geometry and the ramjet fuel control geometry, other 9 variable geometries are needed in the TBCC propulsion system including δN3, δN6, dLN3, MSV, FVABI , LPT-VG, RVABI, A8 and A9 shown in figure 3. Table 8 presents the regulation scope of the multiple variable geometries. Table 8 Regulation scope of the multiple variable geometries in the TBCC propulsion system Variable Regulation Unit Variable Regulation Unit Name

Scope

Name

Scope

δN3

2.5～5.71

degree

FVABI /FVABIds

0～1

――

δN6

0～10.4

degree

LPT-VG

-5/0/-5

degree

dLN3/LN3

0～39.7％

――

RVABI /RVABIds

1～4.89

――

A8

0.8～2.67

Square meter

A9

2～6

Square meter

Figure 27 gives the TBCC installed performance prediction at the climb trajectory, where Frequire means required thrust, Finstall means installed thrust, SFCinstall means installed specific fuel consumption. Inlet drag such as spillage drag, bypass bleed drag, boundary layer bleed drag and cowl lip wave drag were considered in the installed performance. For the exhaust nozzle, only the internal drag was considered in the installed performance; the external aftbody drag was assumed to be considered in the aircraft lift-drag characteristics. It can be seen that the installed thrust at critical operation points could meet the thrust requirements shown in table 1. It should also be noticed that the installed thrust at transonic regimes was comparatively low due to the high subsonic spillage drag at this regime.

22 American Institute of Aeronautics and Astronautics

H=0km Finstall (kN) Frequire (kN)

250 200 Finstall(kN)

H=20.9km H=28.3km

Transonic Regimes

150 100 50

SFCinstall(kg/dN/h)

3.0

0

1

2

3

Ma

4

5

SFCinstall(kg/dN/h)

2.5 2.0 1.5 1.0 0

1

2

3

4

5

Ma

Figure 27 TBCC installed performance prediction at the climb trajectory

Figure 28 shows the TBCC Installed performance prediction at hypersonic cruise mission segment. As indicated in the figure, the installed thrust decreases and the installed specific fuel consumption increases with the altitude increase of the hypersonic float cruise. During the whole hypersonic float cruise, the installed thrust varies from 112 to 74.9 kN and the range of the installed specific fuel consumption is from 2.32 to 2.35kg/daN/h.

23 American Institute of Aeronautics and Astronautics

120

F in stall (kN )

H = 2 8 .3 km

Finstall(kN)

110

100

90

80

70 2 9 0 00

30 0 0 0

3 1 00 0

H /m

32000

S F C in stall (kg /d N /h )

SFCinstall(kg/dN/h)

2 .3 5

2 .3 4

2 .3 3

2 .3 2

2 .3 1 2 9 0 00

30 0 0 0

3 1 00 0

H /m

32000

H /m

Figure 28 TBCC Install performance prediction at hypersonic cruise mission segment Turbofan/Ramjet mode transition simulation was shown in Figure 29.From the figure, it can be seen that:  Surge margin of the fan and compressor is no less than 30%  Positive recirculation margin RM is guaranteed during mode transition, which. indicates no flow recirculation happens at the front bypass variable bypass injector sector.  The range of the flow Mach number at the inlet of the ramjet afterburner Ma7 is from 0.29 to 0.33,which indicates no flameout would happen during mode transition  Total flow and engine thrust are maintained almost constant although slight fluctuation in thrust occurs at the moment when ramjet afterburner light  The simulation results shows stable turbofan/ramjet mode transition could be guaranteed through proper control of the variable geometries plus the turbofan and ramjet fuel control. W total

单 位 ： kg/s

0 .1 0

RM

W turbo

320 240

0 .0 5

W ram

160 80 0 .0 0

0 0

5

10

单 位 ： kN

15

T im e /s

160

20

25

0

5

10

15

20

25

T im e /s

F install

M a7

0 .3 4 0 .3 2

120

0 .3 0 0 .2 8

80

0 .2 6 40

0

5

10

15

20

25

T im e /s

0 .9

0

5

10

15

90

60

N 1zh

0 .6

25

S M com pr

0 .8 0 .7

20

T im e /s

120

N 2zh

1 .0

S M fa n

30 0

5

10

15

T im e /s

20

25

0

5

10

15

T im e /s

Figure 29 Turbofan/Ramjet mode transition simulation 24 American Institute of Aeronautics and Astronautics

20

25

Table 9 presents the definition of the aircraft requirements and figure 30 presents the assumed hypersonic aircraft lift-drag polar table. Table 10 presents the flight performance simulation at different mission segments. It can be seen that it only takes 3.4 hours to fulfill the designed flight range about 12000 kilometers. The fuel consumption along the whole range is 241764 kilogram. Therefore, it can be concluded that the concept design for TBCC propulsion system could meet the basic mission requirements for the hypersonic airplane. The typical flow path for the three modes of TBCC propulsion are also shown in figure 31 ～33.

Table 9 The definition of the aircraft requirement Seats Wing Span (m) Aircraft Length(m) Max take-off gross weight(kg)

300 46 134 440000

Cruise Mach Number Wing Area(m2) Flight Range(km) Max take-off thrust(kN)

5 1690 12000 260*4

Ma=0.25 Ma=0.65 Ma=3 Ma=5

1.2

1.0

CL

0.8

0.6

0.4

0.2

0.0 0.0

0.1

CD

Figure 30 Hypersonic aircraft lift-drag polar table

25 American Institute of Aeronautics and Astronautics

0.2

Table 10 Flight performance simulation at different mission segments Mission Segment Take Off Climb Hypersonic Cruise Descend to Mach 0.9 Subsonic Cruise Descend and Land Total

Fuel Consumption (kg) 1221.1 66774 144041 1287.3 18839 9600.6 241764

Time Spending (h) 0.033 0.82 1.69 0.15 0.45 0.28 3.4

Flight Distance (km) 1.99 1307 9200 557 520 478 12064

Figure 31 Flow path at turbofan mode

Figure32 Flow path at turbofan/ramjet mode transition

Figure 33 Flow path at ramjet mode

26 American Institute of Aeronautics and Astronautics

VII. Conclusion The conceptual design of Inlet/TBCC engine /Nozzle for hypersonic civic aircraft is complex, involved many engineering design disciplines. The integration of the most important disciplines into one tool has the benefit that much more criteria can be considered simultaneously than with the conventional approach of iterating between specialists from different departments. To meet the above goal, an object-oriented numerical simulation model was created for the concept design of TBCC propulsion system, under the visual C++ 6.0 software environment. This model assembles 7 function modules on the basis of the C++ class mechanism, which include: TBCC engine performance simulation module,1-D hypersonic inlet simulation module, 1-D hypersonic nozzle simulation module, multi-goals optimization module,1-D size calculation module for TBCC engine module, weight calculation module for TBCC engine module and airplane performance simulation module. What really matter is that this model could afford a platform to analyze the matching relations between the inlet, TBCC engine and nozzle along a wide flight mach range. Aided by the tools stated above, several important results were got:  a co-axial and tandem configuration was selected, for compact engine size and weight, as the turbine based combined cycle engine which consists of turbofan mode , ramjet mode and turbo/ramjet mode transition.  To meet the various compromised design principles, totally 11 variables were required to be adopted in the variable cycle propulsion system after trade-off between high installed performance and low weight penalty. .  Stable turbofan/ramjet mode transition could be guaranteed through proper control of the 4 variable geometries plus the turbofan and ramjet fuel control.  3 variable geometries including δ3, dN3, δ6 is needed for the two dimensional hypersonic mixed compression inlet to guarantee inlet start, low inlet total pressure loss, satisfactory flow match between the inlet and TBCC engine along a wide flight mach range (Mach 2～Mach 5 )  Variable exhaust nozzle area A9 is needed to avoid thrust loss due to either overexpansion or incomplete expansion among a wide range of available nozzle expansion ratio from 3 to 300.  Based on the presented solution for the Inlet/TBCC Engine /Nozzle integration concept, the flight performance simulation shows that it only takes 3.4 hours to fulfill the designed flight range about 12000 kilometers, where the fuel consumption along the whole range is 241764 kilogram.

Acknowledgements Thanks will go to Pro. Zhu, Zhi-li, Pro. Zhang, Jin and Associated Pro. Tang Hai-long for their valuable guidance. I will also thank my wife Miss Zhou Xuan for her help in the page arrangement.

Subscripts 1 2I 2II 22 2.5I 2.5II 3I 3II 3III

=Inlet =Fan inlet =Ramjet inlet =Fan outlet =Compressor outlet =Fan bypass inlet =Compressor outlet =Fan bypass outlet =Ramjet bypass inlet

4 4.5 5I 5II 6 7 8 T S

=Burner outlet =High pressure turbine outlet =Low pressure turbine outlet =Common bypass outlet =Ramjet afterburner inlet =Ramjet afterburner outlet =Exhaust nozzle throat =Total =Static

References 1

Itahara, H. , Nakata ,Y. and Kimura, T. , “Research and Development of HYPR90-T Variable Cycle Turbo Engine for HST,”ISABE 97-7013, 1997. 2 Watanabe, Y., Miyagi, H., Sekido, T., “Conceptual Design Study on Combined Cycle Engine for Hypersonic Transport”, ISABE 93-7018, 1997. 3 Marshall, A.W. , Ashwani Gupta, K. , Mark J. Lewis, Tom Lavelle, “Critical Issues in TBCC Modeling, ”AIAA 2004-3827,2004. 4 Miyagi, H., Miyagawa, T., Monji, “Combined Cycle Engine Research in Japanese HYPR Project, ”AIAA-952751, 1995. 27 American Institute of Aeronautics and Astronautics

5

Miyagi, H., Kimura, H., Kishi, K., “Combined Cycle Engine Research in Japanese HYPR Program[R]. AIAA-983278, 1998 6 Chen, Min, Zhu, Zhi-Li, Zhu, Da-Ming,Zhang, Jin,Tang, Hai-Long, “Performance analysis tool for turbine based combined cycle engine concept,”Yuhang Xuebao/Journal of Astronautics, vol. 27, no.5, 2006, pp. 854-859. 7 “Performance Prediction and Simulation of Gas Turbine Engine Operation for Aircraft, Marine, Vehicular, and Power Generation ”, RTO ,R-AVT-036. 8 Masayuki Suzuki, Naoki Kuno, Akihiro Tobita, “Current Status of Fan Component Research in HYPR Program,” AIAA 98-3280,1998. 9 Kunihiro Shimizu, Ryouma Nogami. “Current Status of Low Pressure Turbine Component Research in HYPR Program,”AIAA 98-3282,1998. 10 Testsuji Fujimura, Kiyoshi Ishii, Shunji Takagi, “HYPR90-T Turbo Engine Research for HST Combined Cycle Engine,”SAE 951991, 1995. 11 Chen, Da-Guang, Zhang,Jin, “Aircraft/Engine Performance Match and Optimization,” Beijing University of Aeronautics and Astronautics Press, Beijing, 1990. 12 T. J. Benson, “An Interactive Design and Educational Tool for Supersonic External-Compression Inlets,”AIAA 94-2707,1994. 13 Wang,Hai-Yuan, Li,Zhi-Guang, “Flat Supersonic Inlet,”Chinese Aviation Information Center, Beijing, 1996. 14 Liu Da-Xiang, Ye Pei_Liang. “Aero Engine Design Manual No.7 -Intake and Exhaust Devices,”Chinese Aviation Industry Press, Beijing, 2000. 15 Zhang Qing-Hang. “Mission Analysis and Optimization for Aircraft / Propulsion System,”Master Dissertation, School of Jet Propulsion, Beijing University of Aeronautics and Astronautics, Beijing, 1997. 16 L.R. Jenkinson, P. Simpkin, D.,Rhodes, “Civil Jet Aircraft Design”AIAA Education Series, AIAA, New York, 1999. 17 Michio Hamano, Hisamisu Sono, Kanji Murashima, “Overview of The Japanese National Project Super/HYPRSonic Transport Propulsion System,”ISABE 97-7004, 1997.

28 American Institute of Aeronautics and Astronautics