Geometric Effects Analysis and Verification of V-Shaped Support Interference on Blended Wing Body Aircraft

Abstract

A special V-shaped support for blended wing body aircraft was designed and applied in high-speed wind tunnel tests. In order to reduce the support interference and explore the design criteria of the V-shaped support, interference characteristics and geometric parameter effects of V-shaped support on blended wing body aircraft were numerically studied. According to the numerical results, the corresponding dummy V-shaped supports were designed and manufactured, and verification tests was conducted in a 2.4 m × 2.4 m transonic wind tunnel. The test results were in good agreement with the numerical simulation. Results indicated that pitching moment of blended wing body aircraft is quite sensitive to the V-shaped support geometric parameters, and the influence of the inflection angle is the most serious. To minimize the pitching moment interference, the straight-section diameter and inflection angle should be increased while the straight-section length should be shortened. The results could be used to design special V-shaped support for blended wing body aircraft in wind tunnel tests, reduce support interference, and improve the accuracy of test results.

1. Introduction

Wind tunnel testing is an important method to obtain the aerodynamic characteristics of aeronautical vehicles, especially for the optimization and verification of the new layout of aircraft. During the test, a scaled model was installed in a wind tunnel through a support system, the flow over of which was quite different from an unconstrained real flight. The accuracy and reliability of test results were influenced by the support system, leading to support interference [1,2]. Since the birth of the wind tunnel, how to minimize support interference, improve accuracy, and provide high-quality data for aircraft design have been common pursuits of all wind tunnel test institutes. In order to satisfy the different experimental demands of aircraft, plenty of support forms have been developed in high-speed wind tunnels, such as tail sting support, blade support, side wall support, wing tip double support, vane cable suspension support, and vertical tail support [3,4,5,6,7,8,9,10,11,12,13,14,15]. These support forms are widely applied in test institutes, such as the National Transonic Facility (NTF), European Transonic Windtunnel (ETW), Central Aerohydrodynamic Institute of Russia (TsAGI), and China Aerodynamics Research and Development Center (CARDC) [1,16,17,18,19,20,21,22,23,24,25]. For the cylindrical fuselage of a conventional-layout test model, the selection and design of the support system is not difficult [26]. Among the above-mentioned support forms, the tail sting support, which is the most commonly applied primary support form in high-speed wind tunnels, meets the test requirements in both longitudinal and horizontal directions [27,28,29].

In recent years, the blended wing body (BWB) layout has been considered to be the best choice for aircraft of the next generation [30,31,32]. The wing and body of BWB aircraft are highly integrated so that there is no obvious fuselage. The aerodynamic efficiency of BWB aircraft is 20% higher than that of conventional-layout aircraft [33]. In order to take advantage of high aerodynamic efficiency, the test accuracy of BWB aircraft must be higher than that of conventional-layout aircraft; therefore, the support system of BWB aircraft should be designed in detail. Accurate aerodynamic characteristics of the full model cannot be obtained by wing tip double support or vertical tail support, while the requirements of the horizontal test cannot be met by blade support/sidewall support or vane cable suspension support [34,35,36,37]; therefore, all of these can only be applied as auxiliary support. The basis aerodynamic testing of BWB aircraft still needs to use tail sting support as the primary support form. However, the design of a tail sting support system is quite difficult because of the unconventional configuration of BWB aircraft. On one hand, the internal space of a BWB aircraft’s central body is rather limited and the trailing edge is quite thin, with the result that the test model distortion by traditional straight tail support cannot be avoided, making it almost impossible to correct the support interference [38]. On the other hand, the exhausts of the engine and ailerons are located near the trailing edge. As a result, complex flows, such as the coupling of internal‒external flows and the separation of the vortex, may be influenced by traditional straight tail support, which is quite different from conventional-layout aircraft [39]. Therefore, the existing design method of tail support, which is established for conventional-layout aircraft, is not suitable for BWB aircraft. The dedicated tail sting support must be redesigned to meet the test requirements of BWB aircraft, but there exists no specification and prior experience to follow in doing this.

Although various support systems have been designed, they are not suitable for wind tunnel testing of BWB aircraft. Thus, a special V-shaped support for BWB aircraft was designed by the high-speed institute of the CARDC. An investigation into V-shaped support has not yet been observed, and the support interference characteristics and design criteria need to be studied. The transonic flow of BWB aircraft with different V-shaped supports were numerically simulated in this study. The interference characteristics of V-shaped support were analyzed, and the geometric parameter influences of V-shaped support on aerodynamic interference were obtained. According to the numerical results, the corresponding dummy V-shaped supports were designed and manufactured, and the verification test was conducted in a 2.4 m × 2.4 m transonic wind tunnel. The test model was manufactured with material F141, which is considered to be stiff enough for aeroelastic deformations of the test model to be quite tiny. Thus, the aerodynamic interference could be validated; however, the flutter characteristics of BWB aircraft are not detailed in this paper. For a test model made of other materials, the flutter characteristics of BWB aircraft could be analyzed using the method in reference [40]. This research could be used as a reference while designing special V-shaped support for BWB aircraft in high-speed wind tunnels, which could reduce support interference and improve accuracy of test results.

2. Research Object

2.1. BWB Aircraft

The research model in this paper is a typical configuration of a BWB aircraft designed by the CARDC, which is similar to the model in reference [41]. It represents the basic characteristics of BWB layout, although the appearance and parameters are different. In fact, the interference regular and mechanism of V-shaped support is basically the same with different sizes of BWB aircraft. The central fuselage of the BWB aircraft is flat and blends with the outer wing smoothly, resulting in a narrow space in the model body and a thin trailing edge (the outline of the research model is shown in Figure 1).

2.2. Support Equipment

In a high-speed wind tunnel test, tail sting support has smaller interference and better stiffness, which meets support requirements in both longitudinal and horizontal directions, and thus it has been widely used as the primary support form. Conventional tail sting support is usually straight from the end of the front part to the end of the rear part, resulting in the thin trailing edge of BWB central fuselage being distorted and the complex flow near the trailing edge being influenced. The interference of straight tail sting support on BWB aircraft is complicated, and this is difficult to correct. In order to meet the requirements of high-speed wind tunnel tests, a V-shaped support (which is bent like the letter “V” along the longitudinal symmetry plane, as shown in Figure 2) for BWB aircraft was designed by the research team of the CARDC. The innovation of this design was that the distortion of the model was moved forward to the lower surface of the central fuselage, where the support extended into the model. Not only was the distortion of the trailing edge avoided, but also support interference was concentrated on the lower surface of the fuselage. At the same time, the end of both the front part and the rear part were designed coaxially, which could be regarded as “equivalent to straight sting support”, and the advantages of conventional tail support were retained.

In Figure 2, the V-shaped support is divided into the front connecting section (connected to the balance and model), equal straight section, first expansion section, V-shaped inflection section, second expansion section, and rear connecting section (connected to the wind tunnel) from left to right. The appearance and stiffness of the support are determined by the diameter of the straight section (D), the length of the straight section (L), the length of the first section (H, the length of the straight section plus the first expansion section), and the inflection angle (P), as shown in Figure 2. They are the most important geometric parameters, which needed to be designed in detail to reduce the interference.

3. Numerical Simulation

3.1. Computational Fluid Dynamics Method

Numerous applications have been performed to validate the high effectiveness of the computational fluid dynamics (CFD) method in exploring flow mechanisms. In the three-dimensional coordinate system, the compressible Navier‒Stokes equations were solved by the CFL3D code [42]

$$\frac{\partial \widehat{Q}}{\partial t}+\frac{\partial \left(\widehat{E}-{\widehat{E}}_V\right)}{\partial \xi }+\frac{\partial \left(\widehat{F}-{\widehat{F}}_V\right)}{\partial \eta }+\frac{\partial \left(\widehat{G}-{\widehat{G}}_V\right)}{\partial \zeta }=0$$

(1)

where Q represents the conservation variable, E/F/G represent inviscid vector flux, E_v/F_v/G_v represent viscous vector flux, and the symbol t represents time. CFL3D is a Reynolds-averaged thin-layer Navier‒Stokes flow solver for structured grids, which became open source from August 2017. More detail about the governing equation can be found in references [42,43,44,45].

Discrete solution of Equation (1) was performed by the finite volume method. The Shear Stress Transport (SST) two-equation turbulence model was applied during the solution. The lower-upper Symmetric Gauss-Seidel (LU-SGS) implicit method was applied for time advancement, the Roe format was applied for inviscid term space marching, and the central difference format was applied for viscous term space marching. The gas viscosity coefficient was calculated by the Sutherland formula. The boundary conditions include symmetry, wall, and far field. The wall of the wind tunnel and connecting device of support were not simulated, so that the rear end of the V-shaped support was just floating in mid-air.

The calculation objects were the “BWB model with V-shaped support” and “BWB model without V-shaped support”. The influence of the V-shaped support on the aerodynamic characteristics was obtained by comparing the results with and without the support. The simulation of parameter influence was realized by changing the geometric parameters (such as D, L, H, and P) of the V-shaped support.

Most BWB aircraft cruise at high subsonic speed [46,47], for instance, the cruise Mach number of NPU-150 is 0.78. Thus, the simulating Mach number was designed as 0.75 in the current research. The angle of attack was −1°~4°, the static temperature was 288K, and the Reynolds number based on average aerodynamic chord length of the wing was 5 × 10⁶.

3.2. Grid Generation

The structural mesh was generated by Pointwise. A series of meshes for BWB aircraft and the V-shaped support with different parameters were generated separately by the nested method. The surface grids of the model and support were encrypted to simulate the flow in the boundary layer, and the radial size of the first layer was set according to the rule of non-dimensional wall distance y+ equal to 1. Numerical precision was improved with the same topology of the computational grid, and calculation efficiency was improved by avoiding grid re-generation of the aircraft. A mesh of the half-model was generated to conduct longitudinal calculation. The mesh at the downstream was discretized at a higher resolution to resolve the flow at the near wake. According to previous research on grid independence (as shown in

Table 1. Results of grid independence study.

Number of Grids	2 × 106	4 × 106	11 × 106	19 × 106
ΔCL	−0.0008	−0.0009	−0.0004	−0.0003
ΔCD	0.0031	0.0016	0.0006	0.0004
ΔCm	−0.0004	−0.0005	−0.0003	−0.0002

), a small difference in aerodynamic coefficients was observed when the grid number was larger than 11 × 10⁶, thus a mesh with a grid number of 11 × 10⁶ was adopted.

The calculation grid for the V-shaped support with different parameters is shown in Figure 3. During the calculation, the original configuration of the V-shaped support was defined with D = 52 mm, L = 300 mm, H = L2 (the reference length), and P = 15°. The variable configuration of the V-shaped support with different parameters is listed in

Table 2. Different geometric parameters of the V-shaped support.

	D/mm	L/mm	H/mm	P/°
original	52	300	L2	15
variable	44, 55	200, 400	L2 − 100, L2 + 100	13, 17, 20

.

3.3. Reliability Verification

The CHN-T1 standard model [48] was selected to verify the reliability of the numerical method. The CHN-T1 standard model was a 2m-scale transonic large aircraft designed by the CARDC. The model was tested in ETW and the 2.4 m × 2.4 m transonic wind tunnel of the CARDC for comparative verification, and the reliability of the test results was quite high.

The comparison between the calculation results and test results is shown in Figure 4. The research models of CFD and experimental fluid dynamics (EFD) are almost the same. The tests were conducted in two different wind tunnels with the same tail support. It was shown that the numerical results of the CHN-T1 model was in good agreement with the experimental results in the two wind tunnels. In particular, the simulation accuracy of aerodynamic coefficients at α = 0° and in cruise conditions was quite high (the deviation of C_L was less than 0.0055 and the deviation of C_D was no more than 0.0005, as shown in

Table 3. Aerodynamic coefficients comparison of CHN-T1 (Ma = 0.78, Re = 3 × 10⁶).

	EFD-CARDC	EFD-ETW	CFD
CL (α = 0°)	0.1251	0.1210	0.1265
CD (α = 0°)	0.0223	0.0225	0.0228
CD,Cruise (α = 2°)	0.0276	0.0280	0.0278

). The results show that the numerical method of this study was reliable and met the simulation requirements for support interference.

4. Interference Characteristics of V-Shaped Support on BWB Aircraft

The interference of V-shaped support on the longitudinal aerodynamic characteristics of BWB aircraft at transonic conditions is shown in Figure 5. The symbol D52_L300_L2_P15 in the figure denotes D = 52 mm, L = 300 mm, H = L2, and P = 15°, respectively. In the figure, the existence of the V-shaped support resulted in positive lift and negative pitching moment interference, and the influence magnitude was independent of the attack angles. The drag coefficient of the BWB aircraft decreased with the existence of the V-shaped support. When the attack angle was near 0°, the interference of the V-shaped support on the BWB aircraft was ΔC_L = 0.0157, ΔC_D = −0.0009, and ΔC_m = −0.0044. Compared with conventional-layout aircraft (ΔC_D is −0.0008~−0.0030) [3,4,8], the interference of the drag coefficient in the current research was smaller but the influence of the pitching moment could not be neglected. Therefore, the geometric parameter of the V-shaped support had to be designed in detail to reduce the interference of the pitching moment.

The V-shaped support extended into the BWB’s fuselage through the cavity of the lower surface (as shown in Figure 2), so that the influence on the upper surface was limited. The change of pressure distribution near the cavity of the lower surface was the determinant factor of support interference, which is shown in Figure 6 (part of the lower surface near the cavity). The comparison showed that the existence of the cavity and V-shaped support resulted in a pressure increase in front of the cavity, while the original low-pressure area on the lower surface decreased (as shown in red rectangle). The rise of pressure on the lower surface increased the pressure difference between the upper surface and lower surface, resulting in a higher lift coefficient (as shown in Figure 5a). The pressure recovery of the BWB aircraft was improved so that the drag coefficient decreased (as shown in Figure 5b). The pressure interference caused by the V-shaped support was concentrated on the trailing edge of the lower surface, which resulted in a significant negative pitching moment (as shown in Figure 5c).

5. Influence of Geometric Parameter

5.1. Diameter of Straight Section

According to the definition in Figure 2 and Figure 3, the influence of the straight-section diameter on support interference is shown in Figure 7 and

Table 4. Interference results from the BWB aircraft with different straight-section diameters (D = 44 mm/52 mm/55 mm, Ma = 0.75, α = 0°).

D/mm	ΔCL	ΔCD	ΔCm
44	0.01592	−0.00113	−0.00736
52	0.01573	−0.00090	−0.00441
55	0.01490	−0.00090	−0.00310

. The results in the figure are the difference between the “BWB aircraft with different V-shaped support” and the “BWB aircraft without any support”. In the figure, the difference of lift and drag with different straight-section diameters were almost negligible. As the diameter of the straight section increased from 44 mm to 55 mm, the magnitude of |ΔC_m| decreased. When the attack angle was 0°, the ΔC_m achieved a maximum value of 0.0042, which was close to 60%, as shown in Table 4.

The pressure distribution near the cavity of the BWB aircraft with different straight-section diameters is shown in Figure 8. The typical point A/B/C and profile F-F was chosen in the sensitive area of the cavity (as shown in Figure 8a). The pressure characteristics of the sensitive area were reflected by the variation of the pressure coefficients of profile F-F and point A/B/C. The pressure coefficient of point A/B/C with different straight-section diameters is shown in

Table 5. Typical pressure coefficients of the BWB aircraft with different straight-section diameters (D = 44 mm/52 mm/55 mm, Ma = 0.75, α = 0°. The position of point A/B/C is shown in Figure 8a).

D/mm	Cp(A)	Cp(B)	Cp(C)
44	0.163	−0.067	0.183
52	0.115	−0.140	0.180
55	0.108	−0.159	0.179

. As the straight-section diameter increases from 44 mm to 55 mm, the pressure coefficients of point A/B/C and profile F-F (as shown in Figure 8b) decreased. The sensitive area is close to the trailing edge of the fuselage; therefore, the BWB aircraft was influenced by the positive pitching moment and the magnitude of negative interference in Figure 7c decreased.

As the straight-section diameter increased, the pitching moment interference of the V-shaped support on the BWB aircraft was reduced. The interference influence of lift and drag were negligible with different straight-section diameters.

5.2. Length of Straight Section

The influence of the straight-section length on the support interference is shown in Figure 9 and

Table 6. Interference results of the BWB aircraft with different straight-section lengths (L = 200 mm/300 mm/400 mm, Ma = 0.75, α = 0°).

L/mm	ΔCL	ΔCD	ΔCm
200	0.01450	−0.00060	−0.00200
300	0.01573	−0.00090	−0.00441
400	0.01170	−0.00090	−0.00080

. As the length of the straight section increased from 200 to 400 mm, ΔC_L and |ΔC_m| of the BWB aircraft increased, followed by a decrease. When the angle of attack was 0°, the maximum difference of ΔC_L and |ΔC_m| was 0.004 and 0.0036 (as shown in Table 6). The optimization effect of ΔC_m was quite obvious. The difference of the drag coefficient between different parameters of L was less than 0.0003, which was so small that it could be ignored.

The pressure distribution near the cavity with different parameters of L is shown in Figure 10, and the pressure coefficient of typical point A/B/C is shown in

Table 7. Typical pressure coefficient of the BWB aircraft with different straight-section lengths (L = 200 mm/300 mm/400 mm, Ma = 0.75, α = 0°. The position of point A/B/C is shown in Figure 10a).

L/mm	Cp(A)	Cp(B)	Cp(C)
200	−0.091	−0.278	0.026
300	−0.089	−0.269	0.061
400	−0.104	−0.291	0.049

. With straight-section length increases from 200 to 300 mm, the pressure coefficients of point A/B/C and profile F-F (in Figure 10b) increased, resulting in a negative pitching moment and a higher negative interference of C_m (in Figure 9). The difference of pressure between the upper surface and lower surface increased, resulting in a higher interference of the lift coefficient. On the contrary, the pressure of the sensitive area decreased when the straight-section length increased from 300 to 400 mm (as shown in Figure 10b and Table 7). The ΔC_L and |ΔC_m| decreased in the current situation.

Compared with the original configuration (L = 300 mm), the interference of pitching moment was reduced by increasing or decreasing the straight-section length. In order to increase support stiffness and reduce support interference, it is recommended to apply the design of L = 200 mm. The interference difference of drag coefficient was negligible.

5.3. Length of First Section

The interference results with different first-section length is shown in Figure 11 and

Table 8. Interference results of the BWB aircraft with different first-section lengths (H = L2 − 100 mm/L2/L2 + 100 mm, Ma = 0.75, α = 0°).

H/mm	ΔCL	ΔCD	ΔCm
L2 − 100	0.01480	−0.00100	−0.00410
L2	0.01573	−0.00090	−0.00441
L2 + 100	0.01490	-0.00080	−0.00400

. In Figure 11, the interference difference of the aerodynamic characteristics with different parameters of H was quite small. The difference of ΔC_D with different parameters of H was less than 0.0002. Compared with the influence of parameters D and L, the influence of the first-section length on support interference was almost negligible.

The support interference of the BWB aircraft were insensitive to the changing of the first-section length.

5.4. Inflection Angle

The interference results of the BWB aircraft with different inflection angles are shown in Figure 12 and

Table 9. Interference results of the BWB aircraft with different inflection angles (P = 13°/15°/17°/20°, Ma = 0.75, α = 0°).

P/°	ΔCL	ΔCD	ΔCm
13	0.01530	−0.00080	−0.00650
15	0.01573	−0.00090	−0.00441
17	0.01520	−0.00090	−0.00320
20	0.01700	−0.00080	−0.00160

. The difference of lift interference was less than 0.001 when the P ≤ 17°, while the amount increased to 0.0016~0.003 at P = 20°. The difference of drag interference was less than 0.0001, indicating that drag was insensitive to the changing of the inflection angle. The pitching moment interference decreased as the inflection angle increased, and a significant difference was observed. In the calculation range of the current research, the influence of the inflection angle on the pitching moment interference was −78%~77%. When the attack angle was 0°, the pitching moment interference was −0.0065 at P = 13°, which is four times −0.0016 at P = 20°. The interference of pitching moment was quite sensitive to the changing of the inflection angle.

The pressure distribution near the cavity of the BWB aircraft with different inflection angles is shown in Figure 13 and

Table 10. Typical pressure coefficient of the BWB aircraft with different inflection angles (P = 13°/15°/17°/20°, Ma = 0.75, α = 0°. The position of point A/B/C is shown in Figure 13a).

P/°	Cp(A)	Cp(B)	Cp(C)
13	0.189	0.261	0.080
15	0.169	0.231	0.031
17	0.156	0.150	−0.040
20	0.127	−0.030	−0.122

. The pressure coefficient of the sensitive area decreased as the inflection angle increased, as shown in Figure 13b and Table 10, so that the BWB aircraft was influenced by a positive pitching moment and the negative interference of C_m was reduced (as shown in Figure 12c).

The pitching moment interference of the V-shaped support on the BWB aircraft was reduced effectively by increasing the inflection angle.

5.5. Multiple Design Parameters

The interference results listed in Table 4, Table 5, Table 6, Table 7, Table 8, Table 9 and Table 10 were obtained with the variation of single parameters. A simulation of multi-parameter variation was completed in order to investigate the coupling relationship between the geometric parameters of the V-shaped support. In this study, the parameter combination of L = 300 mm/H = L2 was changed into L = 400 mm/H = L2 + 100 mm, and the parameters D and P remain unchanged. The mesh of this study is shown in Figure 3 and the typical results are listed in

Table 11. Coupling relationship between parameters H and L (Ma = 0.75, α = 0°).

	ΔCL	ΔCD	ΔCm
ΔCx(L)	−0.00398	0.00002	0.00358
ΔCx(H)	−0.00079	0.00012	0.00036
ΔCx(L/H)	−0.00380	0.00006	0.00351
ΔCx(L) + ΔCx(H) − ΔCx(L/H)	−0.00097	0.00008	0.00043

. The table, “ΔCx (L)”, represents the difference caused by the changing of parameter L, while other parameters remained the same. The value of “ΔCx (L) + ΔCx (H) − ΔCx (L/H)” reflected the coupling relationship between parameters L and H. The closer this value was to 0, the weaker the coupling relationship was between them.

In Table 11, the influence of multi-parameter variation is almost equal to the sum of the single parameter influence, indicating that the coupling relationship between parameters L and H was quite weak. The conclusion could be used to predict the interference caused by multi-parameter variation.

5.6. Sensitivity Comparison of Geometric Parameter

The maximum difference of support interference with different parameters in the current research is shown in

Table 12. Maximum interference difference of the V-shaped support geometric parameters (Ma = 0.75, α = 0°).

	ΔCL	ΔCD	ΔCm
D/mm	0.00102	0.00023	0.00426
L/mm	0.00403	0.00030	0.00361
H/mm	0.00093	0.00020	0.00041
P/°	0.00180	0.00010	0.00490

. The ΔC_m of the BWB aircraft was influenced obviously by the changing of parameters D, L, and P. The difference of ΔC_D with different parameters was almost negligible.

According to the calculation results, the sensitivity of the V-shaped support geometric parameters on the pitching moment is shown in Figure 14. In comparison, the sensitivity of the inflection angle was the highest. P and D are the most important parameters to be optimized in the future, in order to reduce the interference of the pitching moment.

6. Verification Test of Parameter Influence

As shown in Figure 14 and Table 12, the influence of the inflection angle on support interference was significant. In addition, the inflection angle was the most important geometric parameter of the V-shaped support (it would become the conventional straight sting support when the inflection angle is 0°). With the consideration of the stiffness limit, the dummy V-shaped supports with different inflection angles (P = 13°/15°/17°, as shown in Figure 15) were designed and manufactured to conduct verification tests of support interference.

A verification test of support interference was conducted in the 2.4 m × 2.4 m transonic wind tunnel of the CARDC. The test model of the BWB aircraft was consistent with the numerical simulation, which was installed in the wind tunnel by blade support (Figure 15b). The dummy V-shaped support was installed on the rear strut and extended into the model cavity (test model is not shown in Figure 15b, and the relative relationship can be found in Figure 2). As the movement of the whole test system was synchronous, the collision or transmission of force between the dummy support and test model was avoided. The interference of real V-shaped support on the BWB aircraft was simulated by this method in a high-speed wind tunnel. The verification results of support interference with different inflection angles is shown in Figure 16. Finally, the straight-section diameter of the dummy V-shaped support was cut from 52 mm to 44 mm (as shown in Figure 15c,d), so that the verification test of parameter D could be conducted. The results are shown in Figure 17.

In Figure 16, the pitching moment interference of the BWB aircraft decreased as the inflection angle increased, and the difference of drag interference was quite small, which was in good agreement with the numerical results in Figure 12. Similarly, the changing tendency of ΔC_m and ΔC_D with different parameters of D was basically the same in Figure 17 and Figure 7. The real test equipment, such as blade and strut, were not simulated in the numerical calculation so that the tiny difference between the CFD and EFD was acceptable. The comparison indicated that interference results of the V-shaped support with different geometric parameters in the current research are effective and reliable.

7. Conclusions and Future Work

The influence of V-shaped support geometric parameters on BWB aircraft support interference was obtained by numerical simulation and validated by wind tunnel tests. The main conclusions are as follows:

(1)

The influence of V-shaped support geometric parameters on the pitching moment interference of BWB aircraft is significant, while the changing of geometric parameters has a little influence on the drag coefficient.

(2)

The sensitivity of the inflection angle is the highest of the four geometric parameters, and the pitching moment interference of BWB aircraft can be reduced significantly by increasing the inflection angle. It is recommended that decreasing the straight-section length and increasing the straight-section diameter or inflection angle is an effective way of reducing the support interference of the pitching moment.

(3)

The pressure distribution near the cavity of the lower surface is influenced by geometric parameters of the V-shaped support, resulting in different aerodynamic characteristics of the BWB aircraft.

The research could be used to design special V-shaped support for BWB aircraft in wind tunnel tests, reduce support interference, and improve test accuracy. Furthermore, the test model was designed to be stiff enough to focus our investigations on the aerodynamic interference of V-shaped support. The aeroelastic effects were not detailed in the current research. In future, the blended wing body concept needs to go further to exhibit aeroelastic effects, especially the flutter, and effects of structural displacements on the cruise aerodynamic efficiency need to be evaluated.