Method for self-testing an angle-of-attack probe and method for checking the velocity of an airflow provided by a series of associated pitot probes and angle-of-attack probe

Abstract

The present invention relates to a method for self-testing an angle-of-attack probe comprising the steps of controlling an angular excitation of a rotary element that is rotatable about its equilibrium position according to known excitation characteristics; acquiring angular measurements relating to the rotation of the rotary element, determining a parasitic torque applied to the rotary element on the basis of the angular measurements and of the excitation characteristics; comparing at least one component of the parasitic torque with at least one predetermined threshold and detecting an operating fault in the probe when said component exceeds the predetermined threshold.

Claims

1. A method for self-testing an angle-of-attack probe, the angle-of-attack probe being able to measure the angle of attack of an airflow in a measurement region and comprising a rotary element that is rotatable about an axis of rotation under the influence of the airflow, the axis of rotation being substantially perpendicular to the direction of said airflow, the rotary element defining an equilibrium position in which its rotation about the axis of rotation is stabilized by the airflow; the method comprising the following steps: controlling an angular excitation of the rotary element about its equilibrium position according to known excitation characteristics; acquiring angular measurements relative to the rotation of the rotary element; determining a parasitic torque applied to the rotary element from angular measurements and excitation characteristics; comparing at least one component of the parasitic torque with at least one predetermined threshold; detecting an operating fault in the probe when said component exceeds the predetermined threshold.

2. The self-testing method according to claim 1, wherein the parasitic torque is determined by using a transfer function describing the rotation of the rotary element in the presence of this parasitic torque, the parasitic torque being defined by an inertial component, a damping component, a proportional component, and a friction or imbalance component of the masses.

3. The method of claim 2, wherein the proportional component is an abnormal aerodynamic torque.

4. The method of claim 2, wherein the inertial component is a moment of inertia anomaly.

5. The self-testing method according to claim 2, wherein the inertial component, the damping component and the friction and imbalance component are determined by identifying the portion of the parasitic torque for each corresponding component by subtracting a predetermined portion of a maximum normal torque in its likelihood domain.

6. The self-testing method according to claim 1, further comprising a step for acquiring at least one value of the airflow velocity modulus, the at least one value of the airflow velocity modulus being a consensus velocity, originating from measuring means that are materially separate from said probe; said measuring means being another angle-of-attack probe or a velocity measurement series comprising one or several Pitot probes or one or several lidars.

7. The self-testing method according to claim 6, wherein the proportional component and a pressure velocity of the airflow are determined during the step of determining the parasitic torque, with respect to one another by using a first mechanization and a second mechanization; wherein: according to the first mechanization, the consensus velocity is inserted into the transfer function, the proportional component being determined by using the other components of the parasitic torque; according to the second mechanization, the proportional component is determined from a predetermined value, a pressure velocity value, called extrapolated velocity, being determined by using the set of components of the parasitic torque; the consensus velocity according to the first mechanization being compared to the extrapolated velocity according to the second mechanization, the proportional component according to the first mechanization being compared to the proportional component according to the second mechanization, and a proper operation criterion being established and the proportional component and the consensus velocity are used when said comparisons fall within a predetermined likelihood domain, otherwise the first and second mechanizations are reiterated.

8. The self-testing method according to claim 1, wherein the excitation characteristics define a sinusoidal excitation of the rotary element, the frequency of which is pseudo-random.

9. The self-testing method according to claim 1, further comprising the following steps: acquiring data relative to the aerodynamic noise; modifying the excitation characteristics as a function of the acquired data relative to the aerodynamic noise.

10. An angle-of-attack probe able to measure the angle of attack of an airflow in a measurement region and comprising a rotary element that is rotatable about an axis of rotation under the influence of the airflow, the axis of rotation being substantially perpendicular to the direction of said airflow, the rotary element defining an equilibrium position in which its rotation about the axis of rotation is stabilized by the airflow; the angle-of-attack probe being characterized in that it further comprises a processing module configured to: control an angular excitation of the rotary element about its equilibrium position according to known excitation characteristics; acquire angular measurements relative to the rotation of the rotary element; determine a parasitic torque applied to the rotary element from angular measurements and excitation characteristics; compare at least one component of the parasitic torque with at least one predetermined threshold; detect an operating fault in the probe when said component exceeds the predetermined threshold.

11. The probe according to claim 10, further comprising an excitation motor able to cause an excitation of the rotary element about its equilibrium position according to the excitation characteristics, the excitation motor being an electric motor coupled to a flywheel.

12. The probe according to claim 10, further comprising stabilization means for stabilizing the rotary element about its equilibrium position, said stabilization means comprising an excitation motor.

13. The probe according claim 10, further comprising a sensor capable of generating angular measurements relative to the rotation of the rotary element.

14. The probe according to claim 10, wherein the rotary element is a vane exposed directly to the airflow.

15. The probe according to claim 10, wherein the rotary element is comprised in a cone exposed to the airflow and symmetrical relative to the axis of rotation.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) These features and advantages of the invention will appear more clearly upon reading the following description, provided solely as a limiting example, and done in reference to the appended drawings, in which:

(2) FIG. 1 is a schematic view of an angle-of-attack probe according to the invention;

(3) FIG. 2 is a flowchart of a self-testing method according to the invention, the self-testing method being carried out by the angle-of-attack probe of FIG. 1;

(4) FIG. 3 is a schematic view of a series of Pitot probes and a series of angle-of-attack probes, the series of angle-of-attack probes comprising a plurality of angle-of-attack probes similar to that of FIG. 1;

(5) FIG. 4 is a flowchart of a checking method for checking the series of Pitot probes of FIG. 3, according to the invention.

DETAILED DESCRIPTION OF THE INVENTION

(6) FIG. 1 shows an angle-of-attack probe 10 according to the invention.

(7) This probe 10 is for example integrated into the front part of the fuselage 12 of an aircraft and allows local measurement of an angle of attack of the aircraft.

(8) In other words, the probe 10 allows measurement of the angle of attack in a measurement region then adjacent to the integration point of the probe 10 into the fuselage 12 of the aircraft.

(9) In reference to FIG. 1, the probe 10 comprises a stationary part having a support 15 integrated immovably into the fuselage 12 of the aircraft and a movable part 16 mounted rotatably in the support 15 relative to the axis of rotation Z.

(10) The axis Z is for example substantially perpendicular to the outer surface of the fuselage 12 and therefore to the incident airflow during the flight of the aircraft. This airflow is denoted using reference F in FIG. 1.

(11) The movable part 16 comprises a rotary element 21, an excitation motor 22, stabilization means 23 and a sensor 24.

(12) The rotary element 21 is able to rotate the movable part 16 about the axis of rotation Z under the influence of the airflow F.

(13) This rotary element 21 also defines an equilibrium position in which its rotation about the axis of rotation Z is stabilized by the airflow F.

(14) In the example of FIG. 1, the rotary element 21 is directly exposed to the airflow F and therefore assumes the form of a paddle. Thus, in the example of this figure, the probe 10 is a probe of the vane type.

(15) According to another example embodiment (not illustrated), the probe 10 is a probe with differential pressure equalization between two zones. In this case, the probe further comprises a symmetrical cone exposed to the airflow F and provided with a slot for each of the zones. This cone comprises an inner partition that is rotatable as a function of the difference in pressures in the zones. In this way, this rotary partition has a rotary element 21 within the meaning of the present invention.

(16) The excitation motor 22 allows a rotation of the rotary element 21 to be induced from a control signal.

(17) In the preferred embodiment of the invention, this excitation motor 22 has an electric motor coupled to a flywheel. It is therefore an inertial motor.

(18) According to another example embodiment, this motor 22 is a torque motor then comprising a rotor in the movable part 16 and a stator in the stationary part 15.

(19) Using an inertial motor, however, makes it possible to avoid a fault mode such as a blockage or braking specific to a torque motor.

(20) The stabilization means 23 allows stabilization of the rotary element 21 about its equilibrium position.

(21) These stabilization means may be mechanical or electrical in nature.

(22) In the preferred embodiment of the invention, the stabilization means 23 are electrical in nature and are formed by the excitation motor 22. Thus, according to this example embodiment, the inertial motor 22 makes it possible not only to induce a rotation of the movable element 21, but also to stabilize it about its equilibrium position.

(23) In a variant, the stabilization means 23 are also electrical in nature, but form separate means from the excitation motor 22.

(24) The sensor 24 allows generation of a measurement signal comprising the angle of rotation of the movable part 16 about the axis of rotation Z.

(25) This sensor 24 may be mechanical in nature, such as an inertial gyrometer for example, or electrical or optical.

(26) According to the invention, the probe 10 further comprises a processing module 30 connected both to the motor 22 and to the sensor 24.

(27) This module 30 allows generation of a control signal intended for the motor 22 and reception of a measurement signal generated by the sensor 24.

(28) More generally, this processing module 30 allows implementation of a self-testing method 100 of the probe 10, explained in more detail hereinafter.

(29) To do this, the processing module 30 at least partially assumes the form of software and/or a programmable logic circuit of the FPGA type (Field-Programmable Gate Array).

(30) In the first case, the software is implemented by using a processor and a memory, for example an existing on-board computer. This software is therefore connected to the motor 22 and the sensor 23 by transmission means known in themselves.

(31) In the second case, the software circuit may be arranged near the stationary part 15 of the probe or remote therefrom.

(32) The self-testing method 100 of the probe 10 will now be explained in reference to FIG. 2, showing a flowchart of its steps.

(33) This method 100 is for example carried out periodically or continuously during the flight of the aircraft. According to another example embodiment, this method is carried out upon express request by a pilot and an avionics system.

(34) During step 110 of the method 100, the processing module 30 generates an angular excitation command of the rotary element 21 about its equilibrium position according to known excitation characteristics, at different instants t.sub.1, . . . , t.sub.n.

(35) The excitation characteristics for example assume the values of the current sent to the motor 22 at the corresponding instants. Thus, these characteristics for example take the form of an excitation vector Te having the following form:

(36) $Te=\left(\begin{array}{c}Te\left(t_1\right)\\ .Math.\\ Te\left(t_n\right)\end{array}\right).$

(37) Advantageously, the excitation characteristics Te define a sinusoidal excitation of the rotary element, preferably pseudo-random.

(38) During step 120, the processing module 30 acquires angular measurements relative to the rotation of the rotary element 31 at the instants t.sub.1, . . . , t.sub.n and measured by the sensor 24.

(39) Then, the processing module 30 determines first and second time derivatives of these measurements to obtain a measurement matrix M as follows:

(40) $M=\left(\begin{array}{cccc}{?}^{}\left(t_1\right)& {?}^{}\left(t_1\right)& ?\left(t_1\right)& 1\\ .Math.& .Math.& .Math.& .Math.\\ {?}^{}\left(t_n\right)& {?}^{}\left(t_n\right)& ?\left(t_n\right)& 1\end{array}\right).$

(41) Then, during step 130, the processing module 30 determines a parasitic torque Tq applied to the rotary element 31 from values obtained in steps 110 and 120.

(42) This parasitic torque for example comprises four components assuming the following form:

(43) $Tq=\left(\begin{array}{c}Tq3\\ Tq2\\ Tq1\\ Tq0\end{array}\right),$ where: Tq3 is an order 3 component called inertial component and for example corresponding to a moment of inertia anomaly; Tq2 is an order 2 component called damping component; Tq1 is an order 1 component called proportional component and for example corresponding to a lift anomaly; and Tq0 is an order 0 component called mechanical friction and imbalance component.

(44) The physical meaning of these components originates from the rotation model of the rotary element 21, assuming the following form:

(45) $\left(J+\frac{Tq3}{\frac{d^2?}{d{t}^2}}\right)\frac{d^2?}{d{t}^2}+\left(b+\frac{Tq2}{\frac{d?}{dt}}\right)\frac{d?}{dt}+\left(K+\frac{Tq1}{?}\right)?+\mathrm{noise}\left(t\right)+Tq0+Te\left(t\right)=0,$ where: J, b are coefficients evolving relatively little during the lifetime of the probe; K is a coefficient proportional to the velocity of the airflow F squared; and noise(t) is an aerodynamic noise.

(46) This last relationship may also be written in a matrix form in view of the preceding notions and as follows:
C=(M.sup.TM).sup.?1M.sup.TTe, where

(47) $C=\left(\begin{array}{c}J+Tq3\\ b+Tq2\\ K+Tq1\\ \mathrm{noise}\left(t\right)+Tq0\end{array}\right).$

(48) To determine the components Tq3 and Tq2, it is assumed that the coefficients J and b are known or may be determined by calibration and do not change significantly during the operation of the probe 10.

(49) Thus, these components Tq3 and Tq2 may be determined directly by using the values obtained during the preceding steps.

(50) The component Tq0 is determined by minimizing the aerodynamic noise.

(51) To do this, during step 140, the processing module 30 acquires data relative to the aerodynamic noise and for example originating from external sensors, such as an inertial unit of the aircraft.

(52) Then, during step 145, the processing module 30 modifies the excitation characteristics Te allowing reduction of this noise and therefore carries out steps 110 and 120 again.

(53) To determine the component Tq1, the processing module 30 implements a first mechanization or a second mechanization.

(54) According to the first mechanization, the component Tq1 is determined by using at least one consensus velocity value.

(55) In such a case, the processing module 30 implements step 150, during which it acquires at least one value of the airflow F velocity modulus, called consensus velocity, originating from measuring means that are materially separate from the probe 10.

(56) Preferably, said measuring means is another angle-of-attack probe or a series of Pitot probes preferably including several Pitot probes, the measurements of which are consistent amongst themselves and several static probes. Thus, in this case, the consensus velocity is determined as the pressure velocity determined by the series of Pitot probes.

(57) The consensus velocity is preferably described by its mean, its variance, the number and the expected precision of sources if there are multiple sources.

(58) Then, with the consensus velocity, the processing module 30 determines the coefficient K and, from this coefficient, the component Tq1.

(59) According to the second mechanization, the processing module 30 sets the value Tq1 and determines a value of the modulus of the airflow F velocity, called extrapolated velocity, from this value Tq1.

(60) Advantageously, the first and the second mechanizations are reiterated several times.

(61) In this case, at the end of step 150, the processing module 30 compares the consensus velocity according to the first mechanization with the extrapolated velocity according to the second mechanization, and the value of the component Tq1 obtained according to the first mechanization with the value of the component Tq1 used for the second mechanization.

(62) If these comparisons fall within a predetermined likelihood domain, the processing module 30 establishes a proper operating criterion and therefore uses the values Tq1 and the consensus velocity. Otherwise, the processing module 30 reiterates the first and the second mechanizations.

(63) During step 160, the processing module 30 compares at least one component of the parasitic torque Tq with at least one predetermined threshold.

(64) Preferably, during this step, the processing module 30 compares each of the components of the parasitic torque Tq with at least one predetermined threshold.

(65) Then, during step 170, the processing module 30 diagnoses a fault of the probe when, during step 160, at least one of the components of the parasitic torque Tq exceeds the corresponding threshold.

(66) When a fault is diagnosed, the processing module 30 for example emits an alarm intended for the pilot or for any other avionics system.

(67) Advantageously, according to the invention, the fault nature may be determined by identifying the component of the parasitic torque Tq having exceeded the corresponding threshold.

(68) In particular, when it involves the component Tq0, the failure may be due to friction or to an imbalance of the masses of the corresponding probe.

(69) When it involves the component Tq1, the failure may correspond to an aerodynamic lift anomaly of the paddle: either the paddle is broken or deformed, or it is covered in ice.

(70) The cases of ice and of breakage are also detectable by the inertial component TO.

(71) For a given probe, the case of deformation may be diagnosed easily by using data originating from other probes, since in general, two probes cannot deform in the same way.

(72) Lastly, exceeding of the corresponding threshold by the component Tq2 would mean abnormal damping of the paddle.

(73) One can then see that the self-testing method according to the invention allows a self-testing function of the probe 10 to be carried out during the flight of the aircraft, without using data originating from other angle-of-attack probes.

(74) This allows a diagnosis without using a majority vote, which offers the desired level of safety while retaining an architecture with two probes. Owing to the location of the fault that it allows, which greatly strengthens the location by majority vote, the device also strengthens the integrity and availability on architectures with more than two probes.

(75) The device allows checking of the information originating from other probes used in the aircraft, and in particular Pitot probes allowing measurement of the pressure velocity of the airflow F.

(76) In fact, generally, an aircraft comprises a measuring series made up of a plurality of angle-of-attack probes and a measuring series made up of Pitot probes.

(77) FIG. 3 shows a series of angle-of-attack probes VI including a plurality of angle-of-attack probes A.sub.1, . . . , A.sub.N and a series of Pitot probes VP including a plurality of Pitot probes P.sub.1, . . . , P.sub.M, each of the Pitot probes being able to measure a pressure called total pressure. The series of Pitot probes VP further comprises a plurality of static probes (not illustrated) able to measure a pressure called static pressure.

(78) The series of Pitot probes thus allows measurement of the pressure velocity of the airflow F by using total pressure and static pressure values.

(79) Each of the angle-of-attack probes A.sub.1, . . . , A.sub.N is similar to the angle-of-attack probe 10 previously described insofar as it allows implementation of a self-testing method as previously described.

(80) FIG. 3 further shows a checking module 190 at least partially assuming the form of software and/or a programmable logic circuit of the FPGA type (Field-Programmable Gate Array).

(81) This checking module 190 allows checking of the integrity of the data supplied by the two series VI and VP, and to do so, implements a checking method 200 that will now be explained in reference to FIG. 4, showing a flowchart of its steps.

(82) This checking method is implemented when no operating fault of the angle-of-attack probes A.sub.1, . . . , A.sub.N has been detected subsequent to implementation of the self-testing method by the processing module 30 of each of these probes.

(83) Furthermore, it is initially considered that the assembly of the anemometric installation, that is to say, the Pitot probes P.sub.1, . . . , P.sub.M associated with the static probes and the pressure measuring means, associated or not with a lidar, indicates a velocity value of the pressure velocity type of the airflow F, which will be referred to hereinafter as first velocity value.

(84) In the present document, the term velocity of the airflow or velocity of the air refers to the pressure velocity, calculated from only the values supplied by the Pitot probes and the static probes.

(85) As is known per se, the pressure velocity may also be called conventional velocity or Conventional Air Speed or CAS. In particular, this pressure velocity must not be confused with the true velocity or True Air Speed or TAS, which refers to the mean velocity of the air molecules.

(86) During the initial step 210, the checking module 190 acquires all of the first velocity values indicated via the Pitot probes P.sub.1, . . . , P.sub.M and/or via the lidar(s).

(87) During the same step, this checking module 190 further acquires velocity values of the airflow F determined by the processing modules 30 of the set of angle-of-attack probes A.sub.1, . . . , A.sub.N, during step 130.

(88) Each of these velocity values will be called second velocity value hereinafter.

(89) During the following step 220, the checking module 190 compares the second velocity values with the first velocity values and, after this comparison, uses velocity values to indicate the velocity of the airflow F.

(90) In this comparison, several cases are possible.

(91) In particular, when all of the first and second velocity values are consistent with one another, the checking module 190 keeps the set of first velocity values and second velocity values at nominal precision.

(92) When one of the first velocity values is inconsistent with respect to the other first velocity values and the second velocity values, which are consistent amongst themselves and with respect to the other first velocity values, the checking module 190 keeps said other first velocity values and all of the second velocity values at nominal precision.

(93) When the first velocity values are inconsistent amongst themselves and none of these values is consistent with the second velocity values, which are consistent amongst themselves, the checking module 190 keeps the first velocity values are used at degraded precision and the second velocity values at nominal precision.

(94) Whenall of the first values are consistent amongst themselves but inconsistent with respect to the second velocity values, which are consistent with one another, the checking module 190 keeps the set of first velocity values and of second velocity values at nominal precision with an indication of doubt regarding the supplied values.

(95) In all other cases, no velocity value is used by the checking module 190.

(96) During the following step 230, the checking module 190 communicates the kept velocity values, optionally with their precision, to the pilot or to any other avionics system.

(97) One can thus see that the checking method 200 according to the invention allows checking of all of the Pitot probes. This presents a particular advantage of the invention, since among all of the components of an anemometric system generally equipping an aircraft, the Pitot probes are the components whose integrity is most difficult to establish in flight, given their environment.