57

Conceptual design of a medium size flying wing R Martı´nez-Val
, E Pe´rez, P Alfaro, and J Pe´rez Departamento de Vehı´culos Aeroespaciales, ETSI Aerona´uticos, Universidad Polite´cnica de Madrid, Madrid, Spain The manuscript was received on 2 February 2006 and was accepted after revision for publication on 25 July 2006. DOI: 10.1243/09544100JAERO90

Abstract: Flying wings are one of the most promising concepts regarding the ever-increasing air traffic demand. Furthermore, they help improving economic efficiency and are environmentally friendly, both in terms of emissions and noise. In the first place, the paper deals about the initial design of a medium size C-type flying wing, of the 300-seat class, showing that the aircraft is operationally efficient, and can beat conventional airplanes of similar capacity. It specifically exhibits some considerable gains in field and cruise performances. Second, the paper addresses the potential of some emerging technologies, such as laminar flow control, vectored thrust, and active stability, which provide additional improvements and allow the simplification of the original configuration to a U-type arrangement. A preliminary assessment of emergency evacuation is included. Keywords: flying wing, conceptual design, performances, emergency evacuation

1

INTRODUCTION

Most air traffic forecasts predict a remarkable increase over the next two decades, in spite of the serious downturn after the year 2000 crisis and the terrorist attack of 11 September 2001. The overall revenue passenger-kilometre figure goes up at a pace slightly above 5 per cent, as in references [1] to [3], remarkably over the world economic growth. Needless to say, the predicted traffic growth varies from region to region, with the USA at the bottom and North East Asia – Pacific Rim at the top. Freight traffic is forecast to increase at even higher rates, also requiring a noticeable number of new airplanes as well as the conversion of ageing airliners. However, this tremendous demand of around 20 000 new airplanes will have to cope with the continued pressure to achieve significant reductions in both direct operating cost and environmental impact. Commercial aviation has been mainly based over the last 50 years on what is currently called the conventional layout. This is characterized by a slender fuselage mated to a high aspect ratio wing, with


Corresponding author: Departamento de Vehı´culos Aeroespaciales, ETSI Aerona´uticos, Universidad Polite´cnica de Madrid, Plaza del Cardenal Cisneros 3, Madrid 28040, Spain. email: [email protected]

JAERO90 # IMechE 2007

aft-mounted empennage and pod-mounted engines under the wing [4]. A variant with engines attached to the rear fuselage was also developed during the 1950s and it is still broadly used in business and regional jets. However, it seems that this primary configuration is approaching an asymptote around the size of A380 in its productivity and capacity characteristics [5, 6]. The ever-changing market and technology scenario lead the process of designing new airplanes. And the major questions are, as usual [7]: What does the market need? What design fits better in the long-term scenario? And, what level of technology improvement or new research is required? Within this framework, one of the most promising configurations is the flying wing in its different concepts: blended-wing body, C-wing, tail-less aircraft, etc. It may provide significant fuel savings and, hence, a lower level of pollution. Moreover, the engines are located above the wing and the aircraft does not need high-lift devices in low speed configuration, which results in a quieter airplane. This explains the great deal of activity carried out by the aircraft industry and numerous researchers throughout the world to perform conceptual design-level studies, to address the problems and challenges posed by this layout [8 –13]. Most of the papers deal with very high capacity aircraft, up to 1000 passengers, but forecasts are very promising for medium capacity airliners too. Proc. IMechE Vol. 221 Part G: J. Aerospace Engineering

R Martı´nez-Val, E Pe´rez, P Alfaro, and J Pe´rez

58

Taking into account all this information and the idea that a medium size flying wing would pose fewer or lower level problems than a gigantic 1000seat aircraft, two precedent papers [14, 15] demonstrated the technical feasibility and operational efficiency of a 300-seat flying wing in a C-layout. The results were greatly encouraging in terms of efficiency and productivity, as well as regarding airport compatibility. The present research work points towards confirming that relevant emerging technologies [laminar flow control (LFC), vectored thrust, and active stability] are very well matched to this type of aircraft, and may provide additional improvements. A preliminary assessment of emergency evacuation is also included. 2

THE C-FLYING WING CONFIGURATION

The conceptual design of a C-type flying wing is summarized in this section. The initial specifications of the aircraft correspond to a common long-range mission: 10 000 km with a full passenger load (300 passengers, or 28 500 kg) at M ¼ 0.8. This mission covers many interesting routes between Europe and the US, West US coast to Far East, etc. The selected Mach number, 0.8, is not optimized but simply a representative of the common practice in conceptual design of high subsonic airplanes [9, 16]. Straight leading and trailing edges, and a nose bullet in the apex to accommodate the cockpit with adequate visibility are the basic planform, depicted in Fig. 1. The overall layout belongs to the C-wing type, which exhibits the minimum induced drag among a large group of alternatives [9]. It goes without saying that the 80 m wing span limit of ICAO F category [17] has been respected. In a payload-driven design, the cabin arrangement (around 1 m2 per passenger in a three-class seating

and cabin ceiling higher than 1.85 m) is of maximum relevance. The cabin surface is linked to wing geometry in equation (1) Scab ¼ f (wing planform, inner arrangement, S spar location, A, l, t=c, . . . ) (1) where Scab is the cabin area, S wing gross area, A aspect ratio, l taper ratio, and t/c relative thickness of airfoil. By definition, the influence of the wingspan, b, is given in equation (2) S¼

Two view sketch of the C-flying wing, showing the internal arrangement

Proc. IMechE Vol. 221 Part G: J. Aerospace Engineering

(2)

To select the wing aspect ratio, taper ratio, and relative thickness, three criteria were used: proper aerodynamic performance; minimum maximum take-off weight (MTOW, Wto); and maximum area per passenger, for comfort and emergency evacuation reasons. Thus, for a given wingspan (within the 80 m limit) the cabin area increases noticeably on reducing the aspect ratio. Figure 2 shows the combined effect of aspect and taper ratios, which dramatically restrict available values for such variables. On the other hand, in a pure flying wing with straight leading and trailing edges and constant airfoil type, the wetted area, the dominant term in aerodynamic drag, is related to the internal volume and wing features as
2=3 Wetted area 1=3 t / A c (Volume)2=3

(3)

which again leads the design in favour of low A and high t/c. However, if the aspect ratio diminishes too much, or the relative thickness becomes too large, the aerodynamic performances deteriorate quickly. Slightly aft loaded, 17 per cent thick airfoil sections are used in the outer part of the wing in agreement

Fig. 2 Fig. 1

b2 A

Cabin area fraction of gross wing area, in terms of wing aspect ratio for taper ratio l ¼ 0.1 (upper line), 0.11, 0.15, 0.2, 0.25, and 0.3 JAERO90 # IMechE 2007

Conceptual design of a medium size flying wing

59

with findings reported in the literature [8, 9, 11]. However, for trimming purposes, an upward rear curvature is used in the central part [10], approximately corresponding to the passenger cabin. The transition between both types of airfoils occurs along the freighthold. The airfoil-relative thickness is kept constant at 17 per cent over all wingspan, and the spars always run at 11 and 67 per cent of the chord for structural compatibility. The airfoil thickness in the outer wing looks a bit too high, but it is required for structural reasons since the wing tip area must withstand the loads of ailerons and vertical stabilizers. The overall arrangement results in a CLmax ¼ 1:5 at low speeds. From an aerodynamic viewpoint, such uncommon wing depth is admissible as shown in equation (4) of reference [18] t ¼ (0:90  0:1CLcr )  (Mcr þ 0:02) cos0:5 L c

(4)

where t/c is the relative wing thickness, CLcr the airplane cruise lift coefficient, L the c/4 swept angle, and Mcr the cruise Mach number, assumed to be 2 cents below the drag rise Mach number. Structurally, the wing itself is arranged as a dual entity: an unconventional inner wing with pressurized torque box between the spars, for passenger cabins and holds; and an outer wing with fairly conventional architecture, including fuel tanks outboard of the cargo holds. The structural solution adopted for the inner wing is a vaulted double-skin ribbed shell layout, which is superior to a reinforced thin semi-monocoque shell, for weight saving, load diffusion and fail-safe features [19, 20]. A third spar, external to the torque box, is located behind the rear spar to create adequate spaces for landing gear, APU and other equipment, and to attach elevons that run over a part of the trailing edge. In this initial design, the nacelles were mounted over the wing attached to the second and third spars, and, also, to two main ribs. From the safety viewpoint, this arrangement provides important advantages since it impedes the impact of uncontained engine debris on essential items. As shown in Fig. 3, the passenger cabin is formed by a set of six parallel bays, separated by wing ribs. The bays, of generous narrow body transverse dimension, are connected by slanted corridors in the spanwise direction at the front and rear. Two symmetrical couples of type A doors are located on the sides of the front corridor, through the front spar and leading edge, and another symmetrical pair is located at the rear, through the second spar and trailing edge. All galleys, toilets, and wardrobes are located at the rear of the cabin for aesthetic and operational reasons. This arrangement of exits and JAERO90 # IMechE 2007

Fig. 3

Cabin arrangement in a three-class layout, showing the number of seats in each section. The outer bays are symmetrical. L and G indicate lavatory and galley, respectively. A indicates cabin assistant folding seat

services is very efficient and improves emergency evacuation. In this conceptual design the maximum foreseen capacity is 330 passengers, at 76 – 79 cm pitch, consistent with current regulations for three pairs of type A exits [21] and goes down to 237 seats in a three-class arrangement, corresponding to 0.97 m2 per passenger. First class and business travellers occupy the central bays to benefit from improved comfort levels, although recent investigations indicate that unpleasant accelerations could be counterbalanced by smoothed manoeuvres and multimedia equipment [22]. In conventional designs, the maximum wing loading and thrust over weight ratio are selected according to four common criteria [18, 23, 24]: mid-point cruise capability, take-off field length, second segment climb angle, and approach speed. In the present case, the wing area is already known since it has been determined as part of the cabin sizing process, and a first estimation of MTOW is also available. Hence, the maximum wing loading is Wto/S ¼ 250 Pa. The aforementioned criteria are used here to obtain the thrust over weight ratio and field performances. The estimation of main weights is as follows. By definition, the MTOW to perform the mission is MTOW ¼ OEW þ PL þ TF þ RF

(5)

where OEW is the operating empty weight, obtained from an empirical correlation between OEW and Proc. IMechE Vol. 221 Part G: J. Aerospace Engineering

R Martı´nez-Val, E Pe´rez, P Alfaro, and J Pe´rez

60

MTOW and wing size, analogous to the procedure described in reference [23] for OEW, MTOW, and fuselage size; PL stands for payload; TF represents the fuel burnt during the flight; and RF the reserve fuel. Both this last named and the consumption in take-off, climb, descent, and landing are considered as known fractions of actual weight in each phase [24]. The fuel burnt in cruise, Wfcr, is computed using the Breguet range equation R ¼ K Ln

Wi Wi  Wfcr

(6)

Matching of wing loading and cruise conditions

where g is 1.4, and pcr and Mcr are the pressure and Mach number at cruise conditions, respectively. Equations (9) and (10) can be rearranged as

with K ¼

Fig. 4

V L cj D

(7) pcr ¼

where R is the range, V the cruise speed, cj the specific fuel consumption, L/D the average lift over drag ratio in cruise, and Wi the initial weight. An uncommon characteristic of the flying wing studied is that the cruise capability is required at subsequent steps between 41 000 and 45 000 ft, higher than conventional airliners. This fact deserves some explanation. Let us first assume a parabolic drag polar CD ¼ CD0 þ

CL2 pAw

(8)

where CD and CL are the drag and lift coefficients, respectively, CD0 the non-lift dependent term of aerodynamic drag, A stands for aspect ratio, and w is a parameter, which incorporates the effects of both the vortex and viscous induced drag. The wingtip effect of the C-layout is evaluated as a 30 per cent direct reduction of the vortex induced drag [9, 25]. The trimming drag is assumed negligible. On the other hand, the lift coefficient for maximum range must be [26] CLcr

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ bCD0 pAw

(9)

where b is a parameter related to the Mach number dependence of the specific fuel consumption. For current high bypass ratio turbofans it is about 0.6. This results in CLcr around 0.5 for conventional airliners and 0.3 for flying wings. In flight, the aircraft is always in dynamic equilibrium. Therefore, in cruise, lift must balance weight which is expressed as Wcr ¼ L ¼

g 2C 2pcr Mcr Lcr S

Proc. IMechE Vol. 221 Part G: J. Aerospace Engineering

(10)

2 Wcr =S 2 Wcr =S pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ 2 2 g CLcr Mcr g Mcr bCD0 pAw

(11)

With ordinary Mach number, lower CD0 and much lower wing loading than in conventional airliners, the flying wing must fly at a higher altitude (as indicated in Fig. 4) to benefit from its intrinsic design features. A suitable design value for thrust over weight ratio is Tto =Wto ¼ 0:25, which includes allowance for the thrust lapse from static take-off to high subsonic, high altitude cruise conditions. As indicated earlier this C-wing concept does not require high-lift devices, because of its low wing loading. The resulting main features of the flying wing, taking into account all specifications, constraints, and tradeoffs, are summarized in Table 1. The centre of gravity(g) of the empty flying wing is at 32 per cent of the mean aerodynamic chord. Most conditions fall within a 28 –34 per cent range, much shorter than that of conventional aircraft [23, 27] Table 1 Main features of the C-flying wing Variable

Value

Overall length Overall width Height above ground Wing area Wing span Aspect ratio Taper ratio c/4 sweep angle Cabin area Three-class capacity Cargo hold volume Maximum take-off weight Operating empty weight Maximum payload Maximum fuel weight Thrust to weight ratio at take-off Maximum wing loading

46 m 77 m 16 m 893 m2 75 m 6.3 0.11 308 230 m2 237 pax 72 m3 205 200 kg 108 600 kg 35 000 kg 75 600 kg 0.25 2250 Pa

JAERO90 # IMechE 2007

Conceptual design of a medium size flying wing

consistent with the location of the aerodynamic centre, estimated to be at 32 per cent in cruise [12]. Vertical and horizontal stabilizers have been incorporated in the design to form a C-type layout, but with moderately low sizes. The presence of a horizontal tail was considered appropriate regarding the trimming of the aircraft and to improve its stability and control characteristics. Therefore, half-span horizontal stabilizers are fitted at the extreme of each vertical tail, pointing towards the plane of symmetry (Fig. 1). This solution is rather heavy, since the concentrated loads from the tailplanes have to be carried through vertical fin and wing tip, which are relatively small. The engine, sized following the aforementioned requirements, is a high bypass ratio turbofan like the PW4000, RR Trent, or GE90, rubberized to Tto ¼ 256 KN. Climb and cruise performances have been calculated as a function of weight, Mach number, and altitude. Just after take-off, the maximum vertical speed is 19 m/s (3700 ft/min). The service ceiling at 0.95 Wto is above 45 000 ft at M ¼ 0.8. The aircraft takes a bit more than 30 min to climb up to an initial cruise altitude of 41 000 ft, travelling some 300 km, and burning fuel equivalent to 0.025 Wto. Since flying, wing aerodynamics also benefits from the very high Reynolds number and the relatively low wetted area, cruise lift over drag ratio is 23.4, in good agreement with the values claimed in other studies [8, 11, 12]. Field performances are estimated with energybased methods [24]. The take-off field length is as short as 1860 m without requiring high-lift devices, whereas the landing field length is 1320 m. A three-step cruise (common of this type of studies [16]) at 41 000, 43 000, and 45 000 ft satisfies the initial range specification of 10 000 km with 300 passengers (i.e. 28 500 kg). The fuel efficiency for this route is 19.8 g/pax.km; the same value reported by other authors for larger blended-wing-body aircraft [8, 12]. Regarding the flight mechanics of this novel configuration, the stick fixed positive static margin in cruise is between 4 and 10 per cent of the mean aerodynamic chord, which is assumed adequate, perhaps a bit too high [18, 23]. The short period, phugoid, and Dutch roll modes have been investigated in cruise conditions at 0.85 MTOW, M ¼ 0.8, and h ¼ 45000 ft. For the short period mode, the undamped natural frequency is vs ¼ 1.01 rad/s and damping ratio zs ¼ 0.46, which are acceptable values [23, 28]. For the phugoid mode, the damping ratio is zp ¼ 0.056, which is again acceptable [18]. Finally, for the Dutch roll, the frequency is vd ¼ 0.55 rad/s with zd ¼ 0.065. According to military standards, for class III aircraft and category B JAERO90 # IMechE 2007

61

flying conditions (cruise), this corresponds to level 2 of reference [29], which means minor deficiencies. Consequently, the dynamic response of the aircraft would benefit from a stability augmentation system. A comparison of performances and operation has been carried out against two modern twins of relatively similar capacity: A330-200 and B777-200. These two airplanes are similar in size, but the flying wing is much smaller both in length and height, although wider in span. No major differences are found in airport terminal operations, provided that the rear doors of the flying wing are used for cabin cleaning, and galley and toilet servicing. In this situation passenger services, cargo/baggage handling, and airplane servicing can be done simultaneously with the usual overlap of activities. Interestingly, the loading and unloading of passengers in airport piers requires fingers positioned at about 5 m above the ground for the two wide bodies, but only with a narrow body height of around 3 m for the flying wing. On the other hand, the doors of cargo compartments are at a similar height, around 2.5 –3 m, in the three cases. It is in field and cruise performance, where the flying wing better exhibits its great potential. With unmatched take-off (1860 m) and landing (1320 m) field lengths, the C-wing requires only narrow body-length runways compared with larger, although moderate, values for the A330 and B777; typically of the order of 2300 and 1600 m, respectively. Fuel efficiency, expressed in terms of total fuel burnt per passenger-kilometre is 19.8 g/pax.km for the flying wing, and 21.5 and 23.5 g/pax.km for the A330 and B777, respectively.

3

EMERGING TECHNOLOGIES

The C-type flying wing may efficiently benefit from some emerging technologies, which can further improve its outstanding performances. Specifically, the technologies considered in the present section are LFC, vectored thrust, and active stability. As indicated earlier, the flying wing has a fairly low wing loading of the order of 2000 Pa in cruise. This means low-lift coefficients, with typical section values of Cl ¼ 0.3. This implies a moderate acceleration over the upper surface and, thence, a mild development of the boundary layer. Although the wing chord is rather long, the adverse pressure gradient is very weak, so LFC is easily achievable by means of boundary layer suction [30 – 32]. The selected structural arrangement of a vaulted double-skin shell is well suited to the LFC. The outer skin does not need to be as strong as in conventional aircraft and, on the other hand, the space between the pressurized inner shell and the outer skin can Proc. IMechE Vol. 221 Part G: J. Aerospace Engineering

R Martı´nez-Val, E Pe´rez, P Alfaro, and J Pe´rez

62

accommodate the required equipment for managing the boundary layer. This occurs along the pressurized part of the torque box as well as in the leading edge and up to the third spar; that is almost 85 per cent of the chord over around 35 per cent of the wing area. The parasitic drag coefficient diminishes in accordance to the decrease in friction coefficient due to laminarization and the relative wetted area under its influence, as expressed in equation (12) cfw ¼ 1:328Re 0:5

Sl S  Sl þ 0:5( log Re)2:58 S S

(12)

where cfw is the average friction coefficient over the exposed area, Re is the Reynolds number based on the mean chord, Sl the wing area with laminar boundary layer, and S the wing surface. In the present case, with about 35 per cent of the wing area laminarized, the friction coefficient diminishes by 31 per cent. The weights of the suction equipment and that of the structural reinforcement required by drills and slots are estimated to be about 0.015 Wto. In the baseline design (Fig. 1) the jet engines were mounted in pods over the wing near the trailing edge, with large separation between pod and upper surface, as in the DC-10 or MD-11 airplanes. This location poses some problems when trimming the aircraft, for the engine thrust produces a noticeable nose-down pitching moment. Also, it makes difficult inspection and maintenance. Hence, in an attempt to compensate for these drawbacks and to ameliorate the engine –wing integration, a semi-submerged solution shown in Fig. 5 was adopted. The upper surface is channelled and faired to guide air into the nacelle. This arrangement produces a decrease in drag coefficient and an increase in total pressure recovery [33, 34]. Needless to say, the modifications of upper wing geometry only occur between the corresponding main ribs, about 3.7 m in width in each side of the airplane. With this layout, the line of thrust is very little off-set with respect to the cg height, thus cancelling out the aforementioned trimming problems in the C-type aircraft. However, regarding the engines, the main advantage considered here is vectored thrust capability (VTC). This technology is already available for military engines [35]. Should this technology be introduced in civil airplanes with current technology

Fig. 5

Sketch of wing section and engine relative position

Proc. IMechE Vol. 221 Part G: J. Aerospace Engineering

Fig. 6

Planform view sketch of the U-flying wing. The dotted line indicates the mean aerodynamic chord

level of military engines, the thrust vector could aim anywhere within a cone of 288 half angle around the ordinary thrust line. During take-off, the VTC generates the same pitching moment than a deflection of about 108 of flaperons along the trailing edge. Being close to the plane of symmetry, the engines are less useful for roll manoeuvres. Modern flight control systems provide active stability whenever they operate without failure. In order to take full advantage of the concept, the flight control architecture must be designed with an adequate backup system [36]. Moreover, in the case of a longhaul aircraft, the cg may travel substantially if no provision is taken to counteract such effect. In the flying wing under consideration, the main fuel tanks are located about mid-wingspan, just outer the cargo holds, with an additional small tank below the fore part of the cabin, near the nose landing gear. Since the total volume is more than enough to carry out the mission, the cg can always be set at the appropriate location. Consequently, the trimming and control provided by fuel tank policy, the vectored thrust, and the active stability system are more than enough to eliminate the need of a horizontal tailplane. This results in a lighter and more efficient aircraft. Figure 6 depicts the planform view of this new U-type layout, which may be compared to the upper view of Fig. 1.

4

THE U-FLYING WING CONFIGURATION

This simplified and more integrated U arrangement, equipped with the LFC, vectored thrust, and enhanced active stability, exhibits various advantages: (a) better performance, because of its lower drag; (b) efficient longitudinal control; (c) decreased maintenance cost, for more accessible engines and suppression of complex; horizontal tails; (d) considerable take-off weight savings. JAERO90 # IMechE 2007

Conceptual design of a medium size flying wing

Coming back to the first item, the improvement in specific range is remarkable: around 30 per cent at the same altitude. Moreover, since the drag decreases, the aircraft can be flown higher. Instead of a 41 000, 43 000, and 45 000 ft three-step cruise of the C-type, a 45 000 and 47 000 ft two segments can be done, thus contributing to an additional increment in range or decrease in MTOW. Otherwise, flying higher diminishes the gust loads and dirt deposition, but at the expense of a 0.002 Wto weight penalty for the higher pressurization loads due to increasing the ceiling from 47 000 – 50 000 ft. To define the payload range diagram some assumptions have been made. Firstly, the cruise phase consists of two steps at 45 000 and 47 000 ft. Second, to be conservative, it was assumed that the LFC equipment fails during the last 3 h of flight, analogous to what is required in ETOPS conditions [37, 38], thus forcing to add some extra fuel allowances for the aerodynamic deterioration. Third, the preand post-cruise phases are accounted for as fractions of the current weight [24]. Fourth, the airplane weights are downsized to agree with the original mission specification: that is 10 000 km with PL ¼ 28500 kg. The final values for MTOW, OEW, and trip fuel for the C-wing with conventional technology and the U-wing with LFC is summarized in Table 2, showing 15 per cent reduction in gross weight, 10 per cent in empty weight, and more than 30 per cent in the amount of fuel burnt. Figure 7 shows the payload-range diagrams corresponding to three cases: (a) the original C-type aircraft; (b) the U-layout with LFC and the same mission specification; and (c) the U-wing with the same MTOW as the original C-wing. The resulting fuel efficiency is 14.6 g/pax.km, or 1.82 l/ pax.100 km, fairly lower than the figure claimed as astounding for the A380 [6] and comparable to the consumption of efficient cars at much a lower speed. Some attention has also been paid to field performance, mainly for the possibility of using either vectored thrust or elevons for pitch control. Landing is carried out in conventional way, except for the absence of high-lift devices, resulting in SLFL ¼ 1350 m. Take-off requires further comments. The rotation of jet transports takes place typically

Table 2 MTOW, OEW, and fuel efficiency for C- and U-type flying wings, and conventional competitors

C-flying wing U-flying wing with LFC A330-200 B777-200

JAERO90 # IMechE 2007

MTOW (kg)

OEW (kg)

Fuel efficiency (g/pax.km)

205 200 175 000 230 000 247 200

108 000 97 200 119 600 139 000

19.8 14.6 21.5 23.5

Fig. 7

63

Payload-range diagrams corresponding to the original C-type aircraft, the U-layout with LFC (thick line), and U-layout with LFC and the original MTOW (dashed line)

at around 1.15 Vsto (Vsto being the stall speed in take-off configuration) to allow reaching a 1.2 Vsto at the end of the manoeuvre. However, in the U wing this would imply a too high angle of attack, uncomfortable attitude, and long landing gear legs. Increasing the rotation speed to about 1.25 Vsto and safety speed up to 1.3 Vsto only shifts the take-off field length to 1930 m, which is still a reasonable value, while decreasing the fuselage angle by almost 28.

5

PRELIMINARY ASSESSMENT OF EMERGENCY EVACUATION

A key point addressed in this research work is emergency evacuation. In this respect, it must be recalled that any aircraft have to fulfill appropriate requirements; i.e. American FAR rules [21] or its European equivalent [39]. In practical terms the crucial aspects are: the size and location of exits, the average and maximum distance from seat to exit in distinct scenarios, and the homogeneity of passenger flow through the various exits [40]. According to the cabin area, the maximum number of passengers is close to 330, which requires 3 type A exits on each side of the airplane. In the arrangement depicted in Fig. 3, in each side of the fuselage there are two main doors through the leading edge in the front part of two of the bays and one exit at the rear in the inner bay. The leading edge doors will require specific equipment and structural reinforcement; but behave, otherwise, similar to common fuselage doors. Figure 8 shows the suitable evacuation routes and the connections among various areas. The innermost bay has no front door since the leading edge there is distorted by the nose bullet cockpit. According to current rules, only half of the exits can be used in the 90 s trials. Several scenarios have been analysed. The worst results correspond to the case where the usable exits are the two front doors on one side of Proc. IMechE Vol. 221 Part G: J. Aerospace Engineering

R Martı´nez-Val, E Pe´rez, P Alfaro, and J Pe´rez

64

symmetry, the situation would improve, but this arrangement has not been addressed in depth.

6

Fig. 8

Evacuation paths and exits in the flying wing in the all tourist layout

the plane of symmetry, and the rear one on the other side. Table 3 summarizes the results of the evacuation analysis in this condition and shows a fairly unbalanced situation. In this case the average distances are acceptable, but the maximum distance appears rather long in the rear exit. Moreover, the outermost front door seems to be empty in comparison to others, but in a real trial some of the passengers approaching the inner front door would escape through the nearby empty one for there would be no queue most of the time. These results closely resemble those of airplanes with slightly higher capacity, like A340-300, DC10-30, or L1011-200. Therefore, the flying wing configuration exhibits certain penalty for the extra wide cabin layout, but without representing any noticeable problem in terms of passenger flowrate. If the inner front door, in the mid-bay, could be moved closer to the plane of

Table 3 Evacuation results in the worst condition: two front doors on one side and rear exit in opposite side

Outer front door Inner front door Rear exit

Number of evacuees

Average distance (m)

Maximum distance (m)

54 132 138

5.3 6.3 10.7

8.9 10.3 18.5

Proc. IMechE Vol. 221 Part G: J. Aerospace Engineering

CONCLUSIONS

The flying wing is one of the most promising and efficient configurations to cope with the increasing air traffic demand and related environmental issues. Not only very large but even medium size flying wings, similar in passenger capacity to common wide bodies, exhibit a remarkable improvement with respect to conventional airplanes in field and cruise performance, as well as in emissions and noise. Moreover, the flying wing configuration may better exploit emerging technologies such as LFC, vectored thrust, or active stability, increasing even more the aforementioned advantages. Some drawbacks on the passenger acceptance side may be counterbalanced with smoothed manoeuvres and imaginative interiors and systems. Emergency evacuation issues are of the same level of difficulty than that of currently flying wide bodies. Apart from the analysed technologies, other potential gains could come from an intensive use of composites, aeroelastic tailoring in primary structure, adaptive wings, or ultra-high bypass ratio engines. The present paper shows that C-and U-flying wing layouts exhibit important advantages over conventional competitors. The U arrangement is lighter (for the absence of half a T tailplane at the wing tip), but requires the VTC which is not yet available in civil aviation. On the other hand, the inner architecture of the flying wing can be modified to become the basis of a family [11, 12, 41]. Thus, varying the number of bays and chord length of the cabin, plus some leading and trailing edge alterations would allow to easily accommodate up to 500 seats in one floor. Taking into account the highly positive results shown in this research, this new, potential paradigm of commercial aviation could enter into service within the next decade. Future availability of the LFC and VTC technologies would result in a highly improved second generation, much more efficient aircraft.

ACKNOWLEDGEMENTS A preliminary version of this paper was presented in the 24th ICAS Congress, held in Yokohama, Japan, in August 2004 [15]. The authors acknowledge the permission of ICAS organization to prepare a journal version. The present paper was finished while one of the authors (R. M-V) was on sabbatical leave at Toulouse, France, hosted at SupAero, with financial JAERO90 # IMechE 2007

Conceptual design of a medium size flying wing

support of the Spanish Ministry of Education and Universidad Polite´cnica de Madrid. 16

REFERENCES

17

1 Long-term forecast of air traffic 2004 – 2025. Eurocontrol, Brussels (B), 2005. 2 Global market forecast 2004 – 2023. Airbus, Blagnac (F), 2005. 3 Current market outlook 2004. Boeing Commercial Airplanes, Seattle (USA), 2004. 4 Anderson, J. D. The airplane: a history of its technology, 2002 (AIAA, Reston, VA, USA). ˜ oz, T., and Cuerno, C. 5 Martı´nez-Val, R., Pe´rez, E., Mun Design constraints in the payload-range diagram of ultrahigh capacity transport airplanes. J. Aircr., 1994, 31(6), 1268 –1272. 6 Vigneron, Y. Commercial aircraft for the 21st Century – A380 and beyond. AIAA/ICAS International Air and Space Symposium and Exposition, Dayton, OH, USA, 2003, AIAA paper 2003 – 2886. 7 Strohmayer, A. and Schmitt, D. Scenario based aircraft design evaluation. In Proceedings of 22nd ICAS Congress, Harrogate, UK, September 2000, CD-ROM, paper 5102. 8 Denisov, V. E., Bolsunovsky, A. L., Buzoverya, N. P., Gurevich, B. I., and Shkadov, L. M. Conceptual design for passenger airplane of very large passenger capacity in the flying wing layout. In Proceedings of 20th ICAS Congress, Sorrento, Italy, September 1996, vol. II, pp. 1305 – 1311. 9 McMasters, J. H. and Kroo, I. M. Advanced configurations for very large transport airplanes. Aircraft Des., 1998, 1(4), 217 – 242. 10 Mialon, B., Fol, T., and Bonnaud, C. Aerodynamic optimization of subsonic flying wing configurations. 20th AIAA Applied Aerodynamics Conference, St Louis, MO, USA, 2002, AIAA paper 2002 – 2931. 11 Liebeck, R. H. Design of the blended-wing-body subsonic transport. J. Aircr., 2004, 41, 10 – 25. 12 Bolsunovski, A. L., Buzoverya, N. P., Gurevich, B. I., Denisov, V. E., and Sonin, O. V. Flying wing – problems and decisions. In Lecture Series Course on Innovative configurations and advanced concepts for future civil aircraft (Eds E. Torenbeek and H. Deconinck), 2005, paper no. 9 (Von Karman Institute for Fluid Dynamics, Rhode-Saint-Genese (B)). 13 Martinez-Val, R. and Perez, E. Medium size flying wings. In Lecture Series Course on Innovative configurations and advanced concepts for future civil aircraft (Eds E. Torenbeek and H. Deconinck), 2005, paper no. 8 (Von Karman Institute for Fluid Dynamics, RhodeSaint-Genese (B)). 14 Martı´nez-Val, R. and Schoep, E. Flying wing versus conventional transport airplane: the 300 seat case. In Proceedings of 22nd ICAS Congress, Harrogate, UK, September 2000, CD-ROM, paper 113. 15 Martinez-Val, R., Martinez Cabeza, J. A., and Perez, E. Flying wing and emerging technologies: an efficient matching. In Proceedings of 24th ICAS Congress,

18

JAERO90 # IMechE 2007

19

20

21

22

23

24

25

26

27 28

29 30 31 32 33 34

35

65

Yokohama, Japan, August – September 2004, CD-ROM, paper 1.3.2. Isikveren, A. T. Identifying economically optimal flight techniques of transport aircraft. J. Aircr., 2002, 39(45), 528 – 544. Annex 14. Part 1 Aerodromes, 2003, International Civil Aviation Organisation, Montreal (CND). Howe, D. Aircraft conceptual design synthesis, 2000, pp. 118 – 127, 315, and 317. (Professional Engineering Publishing, London, UK). Mukhopadhyay, V. Structural concept study of noncircular fuselage configuration. AIAA/SAE World Aviation Congress, Los Angeles, USA, October 1996, paper WAC-67. Rodrı´guez, C., Martı´nez-Val, R., and Gutie´rrez, L. Structural weight estimation of a flying wing with a pressurized torque box. In Proceedings of 4th International Seminar on Recent research and design progress in aeronautical engineering and its influence on education, Warsaw, Poland, Part 1, November 2000, pp. 177 – 182. Part 25. Airworthiness Standards: transport category airplanes. Code of federal regulations. Title 14. Aeronautics and space. Office of the General Register, Washington, DC, USA, 1995. Wittmann, R. Passenger acceptance of BWB configurations. In Proceedings of 24th ICAS Congress, Yokohama, Japan, CD-ROM, August 2004, paper 1.3.3. Torenbeek, E. Synthesis of subsonic airplane design, 1982, pp. 145 – 148, 309, and 318 – 320 (Delft University Press, Delft). Roskam, J. Airplane design. Part I. Preliminary sizing of airplanes, 1985, pp. 98 (Roskam Aviation, Ottawa, KA, USA). McCormick, B. W. Aerodynamics, aeronautics and flight mechanics, 2nd edition, 1995 (John Wiley & Sons, New York). Martı´nez-Val, R. and Pe´rez, E. Optimum cruise lift coefficient in initial design of jet aircraft. J. Aircr., 1992, 29, 712 – 714. Roskam, J. Airplane design. Part 5. Component weight estimation, 1985 (Roskam Aviation, Ottawa, KA, USA). Roskam, J. Airplane Design. Part 7. Determination of stability, control and performance characteristics: FAR and military requirements, 1988, p. 293 (Roskam Aviation, Ottawa, KA, USA). Pamadi, B. N. Performance, stability, dynamics and control of airplanes, 1998 (AIAA, Reston, VA, USA). Joslin, R. D. Aircraft laminar flow control. Annu. Rev. Fluid Mech., 1998, 30, 1 – 29. Bewley, T. R. Flow control: new challenges for a new renaissance. Prog. Aerosp. Sci., 2001, 37, 21 – 58. Gad-el-Hak, M. Flow control: the future. J. Aircr., 2001, 38, 402– 418. Seddon, J. and Goldsmith, E. L. Intake aerodynamics, 2nd Edition, 1999 (Blackwell Science, Oxford, UK). Taskinoglu, E. S. and Knight, D. Numerical analysis of submerged inlets. 20th AIAA Applied Aerodynamics Conference, St Louis, MO, USA, 2002, AIAA paper 2002 – 3147. Martı´nez Cabeza, J. A. La tobera vectorial de ITP. Revista de Aerona´utica y Astrona´utica, 1998, 676, 697–701.

Proc. IMechE Vol. 221 Part G: J. Aerospace Engineering

66

R Martı´nez-Val, E Pe´rez, P Alfaro, and J Pe´rez

36 Sauvinet, F. Longitudinal active stability: key issues for future large transport aircraft. In Proceedings of 22nd ICAS Congress, Harrogate, UK, September 2000, CD-ROM, paper 4101. 37 JAR-OPS 1. Commercial air transportation (aeroplanes). Amendment 9. European Aviation Safety Agency, 2005. 38 Martı´nez-Val, R. and Pe´rez, E. Extended range operations of two and three turbofan engined airplanes. J. Aircr., 1993, 30, 382 – 386. 39 Certification specifications for large aeroplanes CS-25. Amendment 1. European Aviation Safety Agency, 2005. 40 Martı´nez-Val, R. and Hedo, J. M. Analysis of evacuation strategies for design and certification of transport airplanes. J. Aircr., 2000, 37(3), 440 – 447. 41 Willcox, K. and Wakayama, S. Simultaneous optimization of a multiple-aircraft family. J. Aircr., 2003, 40(4), 616– 622.

APPENDIX Notation A b cj cfw CD CD0 CL

wing aspect ratio wingspan specific fuel consumption in cruise average friction coefficient over exposed area drag coefficient non-lift dependent term in the drag polar lift coefficient

Proc. IMechE Vol. 221 Part G: J. Aerospace Engineering

CLcr D K L M Mcr MTOW OEW pcr PL R Re RF S Scab Sl t/c TF Tto V Wcr Wfcr Wi Wto

b g l L w

cruise lift coefficient aerodynamic drag range parameter, defined in equation (7) aerodynamic lift Mach number cruise Mach number maximum take-off weight operating empty weight pressure at cruise altitude in standard atmosphere payload range Reynolds number reserve fuel wing area cabin area wing area with laminar boundary layer relative thickness of wing section trip fuel maximum static thrust at take-off cruise speed cruise weight fuel used during the cruise phase cruise initial weight maximum take-off weight parameter appearing in equation (9) specific heats ratio, equal to 1.4 in air wing taper ratio quarter chord swept angle parameter in the lift dependent term of the drag polar [equation (8)]

JAERO90 # IMechE 2007