final publishable summary report -...

ACP2-GA-2013-334861-ANULOID

Project acronym: ANULOID

Project title: “Investigation of novel vertical take-off and landing (VTOL) aircraft concept, designed for operations in urban areas”

Theme: Breakthrough and emerging technologies

Starting date: 01/04/2013 Duration in months: 24

Final publishable summary report

Authors: E. Carrera POLITO

M. Petrolo POLITO

of 25

EXECUTIVE SUMMARY

The principal objective of this project is the computational and experimental investigation of a novel concept of VTOL (Vertical Take-Off and Landing) aircraft - the Anuloid – with respect to its flying qualities, structural feasibility, and aerodynamic characteristics. The Anuloid is conceived for transport missions in urban areas. The Anuloid is a VTOL aircraft with a toroidal shape and a turboshaft engine in its centre that drives a ducted rotor. The outer diameter is equal to 5 m; the maximum take-off weight should not exceed 1200 kg. Only one propulsion system is implemented for both lift and cruise; this should lead to a more favourable payload-to-empty weight ratio. The forward flight speed of the Anuloid is expected to be in the 100-200 km/h range. The typical operating scenarios are the emergency missions and civil transportation in urban areas. The Anuloid concept is based on the following three main features:

The use of a ducted fan powered by a turboshaft for the lift production to take-off and fly.

The Coanda effect that is developed through the circular internal duct and the bottom portion of the aircraft to provide further lift and control capabilities.

The adoption of a system of ducted fixed and swivelling radial and circumferential vanes for the anti-torque mechanism and the flight control.

This aircraft can be considered as an enhanced VTOL with respect to helicopters in terms of lowered noise pollution (due to the ducted engine) and wider operational scenarios. The research activity had dealt with the structural design, the aerodynamic analysis – CFD and wind tunnel -, flight mechanics, and manufacturing. The main outcomes of the project are the following:

The structural and aeroelastic design of the Anuloid is not critical. The combined adoption of stiffened thin-walled structures and composite materials makes the Anuloid structure light and robust.

The Coanda effect is in place during vertical and horizontal flight. Such an effect can be considered as one of the lift and control sources.

The Anuloid surrounding flow can be 3D with flow separation regions.

The vertical flight characteristics of the Anuloid are desirable. On the other hand, the horizontal flight characteristics of the Anuloid in its current form render the aircraft flyable, but not controllable.

A few main recommendations can be drawn for future developments,

The performance of the Anuloid, in particular, the CL/CD ratio for horizontal flight, should be improved.

A nonlinear multi-channel and multi-loop controller should be developed for Anuloid that will allow piloted control of the aircraft.

To support efficiently the Coanda effect on the lower annular cavity some active flow control techniques can be also taken into consideration. For example, the suction or the presence of synthetic jets appropriately positioned and oriented can maintain the flow attached to all the curved surface.

DESCRIPTION OF THE PROJECT

The principal objective of this project is the computational and experimental investigation of a novel concept of VTOL (Vertical Take-Off and Landing) aircraft - the Anuloid – with respect to its flying qualities, structural feasibility, and aerodynamic characteristics. The Anuloid is conceived for transport missions in urban areas. The Anuloid is a VTOL aircraft with a toroidal shape and a turboshaft engine in its centre that drives a ducted rotor, see Fig. 1. The outer diameter is equal to 5 m; the maximum take-off weight should not exceed 1200 kg. Only one propulsion system is implemented for both lift and cruise; this should lead to a more favourable payload-to-empty weight ratio. The forward

of 25

flight speed of the Anuloid is expected to be in the 100-200 km/h range. The typical operating scenarios are the emergency missions and civil transportation in urban areas.

(a) 3D view (b) Cross-sectional view

Fig. 1. The Anuloid geometry

A robust Coanda’s effect is expected through the inner duct and along the bottom side of the circumferential wing. Figure 2a shows a CFD analysis in which the Coanda effect is clearly visible. More details about this analysis are given in the next sections. Figure 2b shows the rotor configuration and the system of fixed and movable blades that is placed in the inner vane of the aircraft.

(a) Coanda’s effect

(b) Propulsion and control

Fig. 2. Anuloid main characteristics

The Anuloid concept is based on the following three main features:

The flow from the rotor goes through the fixed and the swiveling radial vanes that eliminate the rotor torque and provide yaw control by controlling the swirl of the flow.

The individual swiveling of the circumferential vanes can control the tilting of the aircraft by local variations of the resistance to the airflow.

The Anuloid can head towards the tilting direction. Control forces and moments for the Anuloid are obtained through the manipulation of the ducted airflow and the Coanda effect.

The structural frame is made of fiber reinforced composites and sandwich panels. Figure 3 shows the three main structural components, namely

A grid of circumferential and radial stiffeners as main structural frame.

A set of seats with structural purposes that provides additional rigidity.

A set of radial ribs along the circumferential wing that enhance the modal characteristics of the aircraft and avoid aeroelastic phenomena.

of 25

(a) Stiffener grid

(b) Wing ribs and stiffening seats

Fig. 3. Anuloid main structural frame The Anuloid has 5 participants; Politecnico di Torino (Polito), Universite Paris Ouest Nanterre (UPO), TU Delft (TUD), Vyzkumny a zkusebni letecky ustav, a.s. (VZLU), and FESA. Polito is the coordinator. Furthermore, Polito is in charge for the structural and aeroelastic analyses of the aircraft, and the wind tunnel activities. UPO is in charge for the design of digital mock-ups, and for the anti-torque mechanism experimental activities. TUD deals with the flight dynamics and control design and the flying qualities of the aircraft. VZLU deals with CFD analyses and the manufacturing of the fuselage of the scaled aircraft model. FESA’s contributions are CFD analyses, optimization, design, and CAD modelling. The project has five work packages: Project Coordination; Design and Analysis of Anuloid Aircraft; Design and Manufacturing of Scaled Model; Measurements and Wind Tunnel Tests; Dissemination. The main objectives of the project are the following:

Setup of the initial configuration, and preliminary flyability studies.

Creation of a full-scale digital mock-up of the aircraft.

Analysis of the external and internal aerodynamics of the full-scale aircraft through CFD analysis tools.

Creation of the flight dynamics model of the full-scale aircraft to analyse its flying qualities, performance parameters, and flight envelope.

Analysis of the structural and aeroelastic behaviour of the full-scale aircraft.

The design of the reduced scale mock-up of the aircraft for the manufacturing and the following measurements and wind tunnel tests.

Manufacturing of the fuselage of the reduced scale mock-up

Measurements and wind tunnel tests. A brief description of the main activities that were carried out is given hereafter. The initial structural configuration was improved by the design of a structural frame that is composed of radial and circumferential stiffeners and ribs. A preliminary flyability analysis was conducted from M1 to M6. The flyability analysis consisted of a static flight performance analysis, a static stability analysis and a dynamic stability analysis. These analyses exploited the results from the initial configuration setup (mass properties) and the results from the computational fluid dynamics (CFD) for the horizontal and vertical flight. The CFD analyses highlighted the efficiency of the control vanes and the stability of the Coanda effect given a minor modification of the internal duct geometry. The flyability analysis showed that the vertical flight of the Anuloid has satisfactory flying qualities. The dynamic stability analysis showed that the Anuloid is dynamically unstable during horizontal flight. While not dramatically unstable, the Anuloid can have inadequate handling qualities during forward flight. Due to the observed instability, the Anuloid can require an automatic flight control system that is capable of stabilizing the aircraft, in particular during horizontal flight. The most critical issues were detected, and are related to the stability of the Coanda effect, the effectiveness of the anti-torque mechanism and the dynamic instability of the aircraft. On the basis of these outcomes, from M7 to M12,

of 25

priority was given to the experimental and CFD analyses of the Coanda and anti-torque mechanism and to further flight mechanics developments. The main workplan variations at the end of the first year were the following:

The aerodynamic experimental activity was extended to a small scale 3D printed model of the aircraft to investigate the Coanda effect.

A small scale simplified model of the rotor-vane mechanism was built to investigate the anti-torque mechanism.

Other activities - Power means, flight control means and remote (or tethered) control system; Fuselage with integrated power means, flight control means and other devices – were suspended.

The final months of the project aimed at the definition of configuration modifications for future developments of the aircraft with higher TRL.

DESCRIPTION OF THE RESULTS

Structural Design The structural activity was focused on the following main tasks:

Setup of the initial structural layout of the fuselage. This activity was carried out on the basis of the CAD model and by adding appropriate stiffening elements.

Definition of a set of loads (payload, engine, fuel).

Static and dynamic FEM analyses of the initial configuration.

Development of new configurations to deal with potential aeroelastic issues.

Static FEM analysis with the addition of load factors arising from flight mechanical analysis.

The shape of the Anuloid appears very suitable to build in it an appropriate reinforced shell structure which is able to contrast the applied mechanical loadings. The use of composite materials makes the structures lighter than those obtainable by using metallic materials.

(a) Maximum failure index distribution (with load factor 2.5)

(b) von Mises stress at the wing radial ribs (with load factor 2.5)

Fig. 4. Static analyses of composite and metallic components

Figure 4 shows some typical finite element static analyses of the aircraft. Failure indexes and von Mises stresses maximum values are well below the critical values. Static and modal analyses were carried out on different structural layouts and results showed large safety margins. The free vibration analysis highlighted potential aeroelastic issues related to the circumferential wing. The inclusion of radial stiffening ribs proved to be a good solution to avoid potentially dangerous aeroelastic phenomena.

of 25

Computational Fluid Dynamics First, the CFD analysis of the vertical flight was carried out to investigate the Coanda effect strength. Ansys Fluent 15.0 was used. Figure 5 shows streamline distributions for different heights. From the pictures of streamlines it is clear that the Coanda effect is very strong and generates large toroidal whirls touching the ground from basic takeoff/landing position up to about 3.75 m. The shrinking of this large toroidal whirl at this height (and above) should lead to a change of rotor lift, fuselage (pressure) lift, total lift, and required power to values characteristic for hover flight in free air.

(a) Height over ground = 0.3 m

(b) Height over ground = 1 m

(c) Height over ground = 3.75 m

(d) Height over ground = 5 m

Fig. 5. Vertical flight streamlines The response surface functions for the axial moment, the lift force and the required power were derived from the results of the 56 CFD analysis cases with various combinations of input parameters. These functions provide analytical expressions of the dependency of the Anuloid vertical flight properties on the input parameters. The properties of these functions can be easily visualized and their main utilization is in the flight dynamics and the flight performance analyses. Figure 6 shows a response surface example in which it can be seen how for each value of the volume flow the required power is very weakly dependent on the vertical velocity. The achievable lift force for the Anuloid aircraft in relation to the power of the considered engine is sufficient for the vertical flight in the range of the investigated vertical velocities (-5 m/s , +5 m/s) with the rotor diameter at least equal to 1.15 m.

of 25

Fig. 6. Required power dependency vs the volume flow through the rotor and vs the vertical velocity

The horizontal flight was then considered. Figure 7 shows pressure and Mach distributions for two combinations of vane deflections and mass flows. Table 1 presents the yaw moment values and coefficients for various blade deflections and mass flows. The results suggest that, in general, the Coanda effect is in place. However, there may be unstable flow regions around the Anuloid. Furthermore, the blade deflection angles and the mass flow influence the yaw moment coefficient. Higher the mass flow or the blade angle, higher the resulting yaw moment.

(a) Pressure, undeflected, 80 Kg/s

(b) Mach, undeflected, 80 Kg/s

(c) Pressure, deflected, 122.5 Kg/s

(d) Mach, deflected, 122.5 Kg/s

Fig. 7. Horizontal flight pressure and Mach distributions

of 25

Table 1. Yaw moment and coefficient (MN and CN)

ṁ, kg/s deflection, °

MN, Nm CN

80 0 -164 -0.00270

122.5 0 73 0.00120

80 10 2064 0.03390

122.5 10 4200 0.06900

Accelerated horizontal flight conditions were considered as well. Figure 8 shows the aerodynamic forces and moments for a nose-down configuration.

(a) Lift and drag

(b) Pitching moment

(c) Yaw moment

Fig. 8. Accelerated flight aerodynamic forces and moments Wind Tunnel Experimental Activity

The experimental activity focused on the wind tunnel investigation of the aerodynamic characteristics of the aircraft and on the analysis of the Coanda effect. The former was carried out on a 1:5 scale model (see Fig. 9), whereas the latter on a 1:22 3D printed model (Fig. 10).

of 25

(a) Aircraft in the test section

(b) Model equipped for the installation

Fig. 9. Wind tunnel experimental setup

(a) Cross-sectional view

(b) Original geometry

(c) Modified geometry

Fig. 10. Coanda effect experimental setup Figure 11 shows the lift, drag and pitching moment curves. Around the null incidence, the flow configuration can change significantly determining a complete different aerodynamic load redistribution around the body surface. It is likely that around the zero incidence the flow interaction with the body changes considerably due to the completely different geometry that characterizes the upper and lower parts of the aircraft. In the normal flight condition (i.e. negative incidences), the basic shape of the aircraft has very poor lifting characteristics and high drag. The moment is always negative. The lift curves for the negative range of incidence

of 25

do not exhibit any sensitivity to the Reynolds number. A considerable Reynolds effects is instead present for the positive incidences. For the higher Reynolds number, the corresponding curve is characterized by lower lift. The drag curves behave similarly at the two Reynolds numbers. Moreover, at the higher Reynolds number, lower values of the drag are present for the whole range of incidence. The pitching curves behave as the lift curves. At negative incidence, there is no evidence of Reynolds effect, whereas it appears at positive incidences.

(a) Lift

(b) Drag

(c) Pitch

Fig. 11. Lift, drag and pitch curves

Figure 12 shows the pressure distributions on both body surfaces. Wind tunnel (right) and CFD (left) pressure distributions are shown. A good match was found. Furthermore, highly three-dimensional features of the flow field for all the incidences were found.

(a) CFD (b) Wind tunnel

Fig. 12. Pressure distributions, α = 10° The characterization of the Coanda effect was necessary to evaluate the possibility of manoeuvring the aircraft by means of the movable curved surfaces positioned at the annular

0,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

-15 -10 -5 0 5 10 15alfa (°)

CL (Re = 2,1E+06)

CL ( Re = 6,8E+05)

0,00

0,05

0,10

0,15

0,20

0,25

0,30

0,35

0,40

-15 -10 -5 0 5 10 15alfa (°)

CD (Re = 2,1E+06)

CD (Re = 6,8E+05)

-0,10

-0,08

-0,06

-0,04

-0,02

0,00

0,02

-15 -10 -5 0 5 10 15alfa (°)

PM (Re = 2,1E+06)

PM (Re = 6,8E+05)

of 25

exit of the duct fan. The manoeuvrability of the aircraft Anuloid is based on the change of the load distributions on the lower curved surface generated by the asymmetric flow behaviour supposed to be introduced by the curved movable surfaces that would allow the flow to remain always attached to the lower curved surface characterized by a change of the curvature. Some modifications of the exit geometries were also tested to evaluate the effects of different cross sections exit on the pressure distributions. Various disks having different diameters were mounted on the base of the original configuration, namely D = 52mm, D = 62mm, D = 72mm, and D = 82mm. In particular, the smallest diameter represents the original geometry. Pressure distributions for different exit velocities and diameters are shown in Fig. 13. The original configuration has almost constant values of pressure along the whole curved surface. Such a distribution is typical of separated flows. It can be stated that the absence of attached flow on the lower surface excludes the presence of the Coanda effect on that surface. The use of larger diameters postpone the separation.

(a) D = 52 mm

(b) D = 62 mm

(c) D = 72 mm

(d) D = 82 mm

Fig. 13. Pressure distributions along the lower surface of the aircraft, the blue arrow indicates the flow separation point

Figure 14 shows the separation angle for various slot width. The results were compared with those from analytical models. The separation angles of the modified base geometry follow the same trend of the separation angle of the cylinder. The lower values of the separation angle in the case of the Anuloid model are essentially due to the different geometry between the two cases. Moreover, as the ratio width slot/radius decreases (i.e. D increases), the angle of separation increases.

of 25

Fig. 14. Estimation of the flow separation and comparison with an analytical model1

The effects of the movable surfaces were modeled by shifting the disk along the two orthogonal directions. One movement in the direction of the pressure taps (referred to as up and down movements) and the other one in the direction perpendicularly (referred to as left and right movements). The results are show in Fig. 15. Some shifted configurations effect the separation to a great extent.

Fig. 15. Effect of the asymmetric disk shifting on the separation angle

It can be concluded that the Coanda effect is achieved in this simple experiment. As the CFD results also evidenced, this new concept of manoeuvrability can be exploited on a real aircraft configuration. Flight Mechanics The static performance analysis was carried out to investigate the Anuloid capabilities in terms of lift production to take-off, perform horizontal flight and land given the maximum power produced by the power plant. The static performance analysis was focused on the analysis of lift and drag curves as a function of the angle of attack and velocity. This analysis uses the immediate responses given an initial flight condition obtained from CFD for aerodynamic lift, drag and moments.

1 Newman, B. G., “The Deflexion of Plane Jets by Adjacent Boundaries – Coanda Effect”, Boundary Layer and Flow Control,

edited by G. V. Lachmann, Vol. 1, Pergamon Press, Oxford, pp. 232–264, 1961.

of 25

Figure 16 shows the static analysis of the required power and lift for a given vertical velocity as a function of the rotor volume flow. Note that W is defined along the positive Z-axis in the body reference frame, (i.e. W < 0 is ascending flight). Clearly, the maximum output from the Honeywell T3517 (1.35MW) is sufficient to sustain a maximum mass flow of 125 m3/s at any vertical velocity between -5 m/s and +5 m/s. It can be seen that at 120 m3/s and W = -5 m/s (ascending flight) a sufficient lift is produced to support the weight of Anuloid. The CL/CD graphs were then evaluated (see Figure 17). The Anuloid has a very wide flight envelope in terms of angle of attack, this means that the graphs cannot be compared directly to similar graphs for conventional aircraft. Despite this, the graphs provide important information on the static performance during the horizontal flight.

(a) Power

(b) Lift

Fig. 16. Power and lift vs mass flow

(a) 1 m/s

(b) 20 m/s

Fig. 17. CL/CD as a function of the angle of attack for different horizontal velocities

The static stability analysis was focused on the analysis of the static moment equilibrium at different angles of attack. In particular, when the pitching moment increases as the angle of attack increases, the aircraft is considered statically unstable, and vice versa. This analysis used directly the response surfaces obtained from CFD for the aerodynamic lift, drag and moments. The static stability analysis was conducted over the angle of attack (AoA) range from 0 to 80 degrees and at horizontal velocities of 1 m/s and 20 m/s. Figure 18 shows the pitching moment for maximum positive and negative deflections of the circumferential ones. At AoA < 10 and at 1 m/s; the Anuloid is statically stable for negative vane deflections and unstable for positive vane deflections. For AoA > 10 degrees, the opposite behavior was found; the Anuloid is statically unstable for negative vane deflections and stable for positive

of 25

vane deflections. A statically unstable aircraft is, in general, not flyable without a stability augmentation system. At a horizontal velocity of 20 m/s the situation is the same. By comparing the plots for 1 m/s and 20 m/s an important observation can be made, the effect of the control surfaces reverses for AoA < 30; this means that there is a significant nonlinearity in the dynamics of Anuloid and implies that a nonlinear control system must be used to control the aircraft.

(a) 1 m/s

(b) 20 m/s Fig. 18. Pitching moment coefficient for maximum positive and negative deflections of the

pitch control vanes

The dynamic stability during the vertical flight was analyzed. For this, the aircraft was first brought into a trimmed condition in the sense that the body accelerations and the angular rates are equal to zero. At the trimmed condition, a step input was given on the radial control vane.

Fig. 19. Step input on the radial control vane response

of 25

Figure 19 shows the responses and it can be seen that the yaw rate increases linearly with the radial vane extension. This is a desirable behavior, any damping coefficients will limit the maximum yaw rate.

Fig. 20. Step input on the radial control vane response Then, a step input on the volume flow control parameter was considered and the results are shown in Fig. 20. The ascent velocity increases to more than 5 m/s, but the increase is nonlinear. The yaw rate increases due to the torque effect of the main rotor which is not compensated by an extra radial vane deflection. Again, these are desirable properties from the stability point of view. The dynamic stability during the horizontal flight was investigated by first bringing the aircraft into a semi-trimmed condition in the sense that the body accelerations and the angular rates are close to zero. At this semi-trimmed condition, a pulse input was given on the pitch control vanes.

of 25

Fig. 21. Pulse input on the pitch control vane response Figure 21 shows that the pitch rate q suddenly increases in magnitude and keeps increasing even after the vanes have been moved back to the trimmed state. These results show that Anuloid is dynamically unstable during forward flight. The time to double amplitude, which is a measure of instability, is around 2 seconds. Next, a pulse input on the mass flow control parameter was given (Fig. 22). Again, the pitch rate departs from the trimmed state and significant instability is now also present in the forward and vertical velocities. Again, the time to double amplitude is 2 seconds.

Fig. 22. Pulse input on the mass flow control parameter response

of 25

The horizontal pitch and roll control authority was then investigated. Figure 23 shows the tracking performance of the 5 deg/s pitch rate reference for values of the control moment scaling parameters (KM) ranging from 1 to 5. Clearly, if KM=1 (the nominal aerodynamic model) the reference cannot be followed while the control vanes are completely saturated at their maximum positive deflection and the aircraft quickly destabilizes and becomes uncontrollable.

(a) Reference tracking

(b) Vane deflection Fig. 23. Pitch rate tracking performance of the aircraft with 5 different values for KM given a +5 deg/s pitch rate reference at 30m/s. This figure shows that KM > 3 to prevent pitch control vane saturation. Note that for KM < 3 the aircraft is not capable at all of tracking

the reference.

Figure 24 shows a more careful selection of values for KM. From this figure it is clear that a positive 5 deg/s pitch rate can be tracked without saturating the control vanes beyond the initial transient when KM>=3.5. Interestingly, the effective tracking of negative pitch rates requires a lower KM than for positive pitch rates, as can be seen in Fig. 25. This can be explained by the tendency of the aircraft to pitch forward, i.e. the pitching moment is asymmetric with respect to pitch angle. Note that even for higher values of KM we still see brief saturation of the pitch vanes during the transient phase of the maneuver. Preventing this saturation is possible, but at the cost of a longer rise time (lower pitch acceleration).


(b) Vane deflection Fig. 24. Pitch rate tracking performance of the aircraft with 5 different values for KM given

a +5 deg/s pitch rate reference at 30m/s. If KM >= 4 the pitch control vanes do not saturate after the initial transient for a positive pitch rate reference.

of 25


(b) Vane deflection Fig. 25. Pitch rate tracking performance of the aircraft with 5 different values for KM given

a -5 deg/s (Negative) pitch rate reference. If KM >= 3 the pitch control vanes do not saturate for a negative pitch rate reference beyond the initial transient.

To determine KL, the rolling moment scaling factor, a similar approach as that taken to determine KM was taken. The analysis used a roll rate reference of 5 deg/s at a velocity of 30 m/s. In Fig. 26, the results of the analysis are shown. In this case KM was fixed to a value 4 such that unstable pitch dynamics would not influence roll dynamics. In this case, we found that for a value KL>=3.25 accurate roll rate tracking could be achieved without saturating the roll control vanes beyond the initial transient. To determine KN, the approach above was repeated again. From earlier analysis it is concluded that the yaw dynamics of the aircraft are both statically and dynamically stable. The yaw rate response for a range of values for KN between 1 and 3 was analyzed for a yaw rate reference of 5 deg/s and a velocity of 30m/s. The results of the analysis are shown in Fig. 27, showing that KN =1 leads to desirable yaw dynamics responses. As a result, no scaling factor needs to be imposed on the control yawing moment model. Finally, the thrust control force scaling KZ determined using a different approach in the sense that the analysis was performed at 10 m/s as this gave the most pessimistic results. For this analysis, a climb rate controller was used with as reference a 1 m/s positive climb rate. It is found that if KZ >= 10 a climb rate of 1 m/s can be sustained at a 10 m/s forward velocity, see Fig 28. The same analysis performed at a velocity of 30 m/s showed that for KZ >= 10 the 1 m/s climb rate could easily be sustained.


(b) Vane deflection Fig. 26. Roll rate tracking performance of the aircraft with 5 different values for KL given

a 5 deg/s pitch rate reference at 30m/s. If KL >= 3.75 the pitch control vanes do not saturate for a positive pitch rate reference.

of 25


(b) Vane deflection Fig. 27. Yaw rate tracking performance of the aircraft with 3 different values for KN given

a 5 deg/s yaw rate reference at 10m/s. If KN = 1 the yaw control vanes do not saturate beyond the initial transients.


(b) Vane deflection

Fig. 28. Climb rate tracking performance of the aircraft with 5 different values for KZ given a 1 m/s climb rate reference at 30 m/s. If KZ >= 10 the volume flow does not

saturate beyond the transient.

An analysis of the scaled aerodynamic forces was then carried out to determine whether the scaled aerodynamic moments can still be considered physically valid. In Fig. 29a, the nominal pitching moment for a maximum positive vane deflection is shown, clearly demonstrating that the total pitching moment is always negative, hence leading to an uncontrollable aircraft. In Fig. 29b, the pitching moment is again plotted but this time with the scaled moment coefficient factor KM=4.The figure shows that the total pitching moment never exceeds 3000 Nm which is within the range of the nominal moment magnitudes, and can thus be considered physically valid.

(a) ) KM=1

(b) KM=4

of 25

Fig. 29. Pitching moment (factor KM=1) components for maximum positive vane deflection (Pitch Up)

The lift force before scaling was compared to the lift force after scaling with KZ = 10. As discussed in previous sections, this scaling is necessary to allow a climb rate of 1m/s during forward flight. In Fig. 30a, the nominal lift force components are plotted. From this figure, it is clear that the magnitude of the volume flow control effector is at least an order of magnitude smaller than the uncontrolled lift component, indicating a very low effectiveness of this control input. In Fig. 30b, the lift force components are shown after scaling with KZ=10. This time the control component of the lift force is of the same magnitude as the uncontrolled component. The total lift force is significantly higher than that of the unscaled total lift force leading to the conclusion that this scaling may not be physically valid.

(a) ) Kz=1

(b) Kz=10

Fig. 30. Lift force components for a volume flow of 120m3/s

The results from the dynamic force and moment analysis were then considered. These forces and moments are indicative for the forces and moments expected during normal flight. The analysis consisted of two high amplitude pitch rate maneuvers flown at 30 m/s. The analysis uses the modified aircraft aerodynamic database (i.e. KL=4, KM=4, KZ=10). The dynamic damping coefficients were given values such that the order of magnitude of the damping moments was equal to the order of magnitude of the gyroscopic moments. Both maneuvers were flown in closed loop by the automatic control system. The climb rate controller for the engine volume flow was used to keep altitude constant. The first maneuver was a sine shaped reference on the pitch rate. A decomposition of the forces and moments acting on the aircraft during the sine maneuver are shown in Figs. 31 and 32. The control moments and aerodynamic moments are dominant factors in the decomposition. The dynamic damping peaks at 87 Nm (pitch damping), while gyroscopic moment (rolling moment) peaks at 80 Nm. Note also the relatively low magnitude of the resultant moment which peaks at 128 Nm. The uncontrolled lift forces steadily decreases during this maneuver which is due to the decreasing airspeed. This is compensated for with the climb rate controller which commands a steady increase in thrust.

of 25

Fig. 31. Aircraft states as function of time for a sine shaped reference on the pitch rate

Fig. 32. Decomposition of forces and moments during sine shaped reference on the pitch rate

of 25

The same analysis was repeated for a step input on the pitch rate, see Figs. 33 and 34. The control moments and aerodynamic moments again dominate. This time, however, the resultant pitching moment is very significant at 1330 Nm. The dynamic damping peaks at 108 Nm (pitch damping), while gyroscopic moment (rolling moment) peaks at 98 Nm. From the figure it can be seen that there is a sharp peak in the uncontrolled lift force which is due to the sudden increase in pitch rate. This sharp peak is compensated for by the climb rate controller which aims to keep the climb rate constant. This analysis shows that while not negligible, dynamic damping and gyroscopic moments are of limited magnitude. Despite this, especially the contribution of the gyroscopic moment cannot be neglected because of its tendency to couple rotations along a given axis to an axis perpendicular to it. For example, a strong pitch up maneuver will lead to a significant moment along the roll axis and vice versa. This coupling must be taken into account during control system design.

Fig. 33. Aircraft states as function of time for a pitch rate step input

of 25

Fig. 34. Forces and moments acting on the aircraft during a pitch rate tracking task, where a pitch rate of 5 deg/s is tracked after 3 seconds. Notice the sharp peak in the resultant

pitching moment during onset of the pitch maneuver

Dissemination This WP was regularly carried out. Significant result:

The results that were obtained in M0-M6 were published in a paper: Petrolo M., Carrera E., D’Ottavio M., de Visser C., Patek Z. and Janda Z., On the Development of the Anuloid, a Disk-Shaped VTOL Aircraft for Urban Areas, Advances in Aircraft and Spacecraft Science, 1(3), pp. 353-378, 2014.

The following paper conference was presented by Dr. Janda (FESA): ANULOID Novel VTOL aircraft for urban areas. Concept, methodology and project‘s interim results, at AVIA-INVEST international conference, Riga 10 –11 April, 2014

An oral presentation to a wider public: was carried out by UPO: ANULOID Overview at the 50th anniversary of the Université Paris Ouest Nanterre La Défense.

Another journal paper will be submitted in the next weeks in which the final results of the project will be described.

The Anuloid project will be presented at the next AERODAYS meeting in London.

of 25

IMPACT

The Anuloid could represent an effective mean to alleviate the increasing problems with emergency and urgent transport in urban areas. Traffic congestion and disrupting ground transport will occur during next decades more frequently and more extensively, due to an increasing urban population and traffic, as the Focus group of European Commission for future transport has recently alerted. Today, helicopters are the standard solution for emergency and urgent transport in the open environment, but their operations are severely limited by weather and they cannot take-off and land near buildings. It should be especially noted that 20% - 30% of helicopter missions are lost annually in Europe due to weather conditions, and more than half of those are directly associated with icing. The Anuloid is a ducted fan VTOL vehicle. It has the capability of performing rescue missions in areas that are totally inaccessible to helicopters. It can work in very gusty winds (relying on fly-by-wire control system), and is capable of operating in icing conditions. In addition, the Anuloid will have a much lower noise level than helicopters, achieved By using new low noise blades and by having rotor contained inside double-walled ducts, with optional additional passive or active acoustic treatment placed on the duct walls. The Anuloid project outlined a number of main technical issues to be deal with in the case of future developments. The primary Anuloid issues are the forward flight instability and insufficient control authority. The secondary Anuloid potential issue, indicated by the CFD analysis, seems to be very high drag, but the magnitude of this drag must validated in future in wind tunnel with model with fully functioning rotor and with functioning wall jet (Coanda effect), because drag calculation by CFD are very often for complex geometries (such as Anuloid) not precise due very high sensitivity of drag on computational grid resolution, type and parameters of turbulence model and on near-wall treatment of flow. Moreover, as far as the anti-torque system is concerned, it can be stated that the proposed configuration is qualitatively able to produce anti-torque. Further, quantitative analyses are required to enhance the reliability of the anti-torques system. As a general guideline, the reconfiguration of the Anuloid should move from a “flying-disk-shaped” to an “aircraft-shaped” Anuloid. This means that, for instance, a number of primary components should be added, such as

Lateral wing-like surfaces with movable surfaces.

Small rotors (both for control and anti-torque capabilities). In the event of an Anuloid follow-up, the following high-priority activities should be carried out:

1. Additional wind tunnel experiments in which rotor thrust is present must be conducted to validate the CFD dataset, in particular during forward flight. The cause of the very strong nose-down moment predicted by the horizontal CFD response surfaces should be investigated.

2. Flight testing must be performed with a scale model to further validate the accuracy of the current aerodynamic dataset. In particular, dynamic effects (gyroscopic coupling, dynamic damping, Coriolis forces), which cannot be obtained in a wind tunnel, muat ve evaluated. The true dynamic damping coefficients can be estimated via flight testing.

3. The performance of Anuloid, in particular, the CL/CD ratio for horizontal flight, should be improved; the current model cannot sustain altitude and forward velocity for flight speeds above 15 m/s. This problem is alleviated by increasing the effectiveness of the rotor volume flow as a control effector by a factor of at least 10.0.

4. Anuloid is statically and dynamically unstable, meaning that un-augmented control is infeasible. Therefore a nonlinear multi-channel and multi-loop controller should be developed for Anuloid that will allow piloted control of the aircraft.

5. To support efficiently the Coanda effect on the lower annular cavity some active flow control techniques can be also taken into consideration. For example, the suction or

of 25

the presence of synthetic jets appropriately positioned and oriented can maintain the flow attached to all the curved surface.

The results that were obtained in M0-M6 were published in a paper:

Petrolo M., Carrera E., D’Ottavio M., de Visser C., Patek Z. and Janda Z., On the Development of the Anuloid, a Disk-Shaped VTOL Aircraft for Urban Areas, Advances in Aircraft and Spacecraft Science, 1(3), pp. 353-378, 2014.

The following paper conference was presented by Dr. Janda (FESA):

ANULOID Novel VTOL aircraft for urban areas. Concept, methodology and project‘s interim results, at AVIA-INVEST international conference, Riga 10 –11 April, 2014

An oral presentation to a wider public: was carried out by UPO:

ANULOID Overview at the 50th anniversary of the Université Paris Ouest Nanterre La Défense.

Another journal paper will be submitted in the next weeks in which the final results of the project will be described. The Anuloid project will be shown at the AERODAYS 2015, London, UK 20-23 October 2015.

final publishable summary report -...

Documents