Method and system for determining brake wear

Abstract

Method and system for determining brake wear This method (PDU) makes it possible to determine the wear of a brake caused by a single braking operation. It comprises: - a step of obtaining (E10) state variables of said braking; and - a step of determining (E20) said wear using a wear determination device (CV) which takes into account said state variables and wear modeling parameters, said wear modeling parameters being previously determined by implementing the following steps: - for at least one sequence of single braking operations: obtaining state variables of each of said single braking operations of said sequence; obtaining cumulative wear of the brake caused by all the single braking operations of said sequence; - determining modeling parameters of said determination device from said state variables and said cumulative wear. Figure for the abstract: Fig.2

Description

Title of the invention: Method and system for determining brake wear Prior art

[0001] The invention lies in the general field of devices for determining the wear of a brake.

[0002] It finds a preferred but non-limiting application for determining the wear of an aircraft brake.

[0003] In the current state of the art, brake wear sensors do not have sufficient resolution to determine the wear caused by a single braking operation, but only the wear caused by a series of braking operations. In particular, the brakes of an aircraft deteriorate very slowly, and between two flights for example, the resolution of the wear sensors does not allow the difference to be made between before and after the flight.

[0004] A method developed consists of carrying out several tests in order to isolate the braking that can be isolated, then deduce the others. Thus, for example, if it is possible to chain together several landing tests, it is possible to measure the cumulative wear caused by the braking of all of these landings, then to deduce the brake wear caused by a single landing. It is then possible to chain together a landing and a hot taxi, to measure the cumulative wear by these two phases, and to deduce the brake wear caused by the hot taxi alone.

[0005] Such a practice is very tedious to implement and does not give satisfactory results because it does not allow for taking into account external conditions, such as temperature and pressure which have a strong impact on the braking phenomenon.

[0006] The present invention aims at a solution for determining the wear of a brake which does not have these drawbacks. Statement of the invention

[0007] More specifically, the present invention relates to a method for determining the wear of a brake caused by a single braking operation, this method comprising: - a step of obtaining state variables of said braking; and - a step of determining said wear using a wear determination device which takes into account said state variables and wear modeling parameters, said wear modeling parameters being previously determined by implementing the following steps: - for at least one single braking sequence: a / obtaining state variables of each of said unique braking operations of said sequence; b / obtaining cumulative wear of the brake caused by all the single braking operations of said sequence; - determination of said modeling parameters of said determination device from said state variables and said cumulative wear.

[0008] Correlatively, the invention relates to a system for determining the wear of a brake caused by a single braking operation comprising: - a module for obtaining state variables of said braking; and - a wear estimation device which takes into account said state variables and wear modeling parameters,

[0009] said wear modeling parameters being previously determined by implementing the following steps:

[0010] - for at least one sequence of single braking operations: a. obtaining state variables of each of said unique braking operations of said sequence; b. obtaining cumulative brake wear caused by all single braking operations in said sequence; - determination of said modeling parameters of said determination device from said state variables and said cumulative wear.

[0011] Thus, and in general, the invention proposes to model the wear caused by a single braking operation by parameters determined from the cumulative wear resulting from a braking sequence and the state variables at the time of each of the braking operations in the sequence.

[0012] In this disclosure, the state variables of each of said single braking operations may for example be comprised among a temperature, a pressure, a duration, measured during said single braking operation.

[0013] According to a first embodiment, the determination of said modeling parameters comprises: - for each single braking, a determination of the individual wear caused by this braking from the state variables of this braking using said estimation device; - a determination of said modeling parameters to minimize a difference between the sum of individual wear and said cumulative wear.

[0014] In a particular embodiment, each of the individual wears caused by a single braking is estimated by the same neural network taking as input the state variables of this single braking.

[0015] In a second embodiment, the determination of said modeling parameters comprises a resolution of a linear system which defines said cumulative wear of said at least one sequence as a sum of the individual wears caused by the single braking operations of said sequence, each individual wear caused by a single braking operation being a linear combination of said modeling parameters and functions of said state variables of this single braking operation.

[0016] In one embodiment, the linear system which defines said cumulative wear of said at least one sequence is: A = A.0 where A, 0 and A respectively represent said state variables, said modeling parameters and said at least one cumulative wear.

[0017] In a particular embodiment, the different steps of the wear determination method are determined by computer program instructions.

[0018] Consequently, the invention also relates to a computer program on an information medium, this program being capable of being implemented in a computer, this program comprising instructions adapted to the implementation of the steps of a method as described above.

[0019] This program may use any programming language, and be in the form of source code, object code, or code intermediate between source code and object code, such as in a partially compiled form, or in any other desirable form.

[0020] The invention also relates to an information medium readable by a computer, and comprising instructions of a computer program as mentioned above.

[0021] The information carrier may be any entity or device capable of storing the program. For example, the carrier may comprise a storage means, such as a ROM, for example a CD ROM or a microelectronic circuit ROM, or a magnetic recording means, for example a hard disk.

[0022] On the other hand, the information medium may be a transmissible medium such as an electrical or optical signal, which may be conveyed via an electrical or optical cable, by radio or by other means. The program according to the invention may in particular be downloaded from a network such as the Internet.

[0023] Alternatively, the information carrier may be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the method in question. Brief description of the drawings

[0024] Other characteristics and advantages of the present invention will emerge from the description given below, with reference to the appended drawings which illustrate an exemplary embodiment thereof without any limiting character. In the figures:

[0025] [Fig.l] [Fig.l] represents, in its environment, a system for determining the wear of a brake caused by a single braking operation in accordance with the invention.

[0026] [Fig.2] [Fig.2] represents, in the form of a flowchart, the main steps of a method for determining the wear of a brake caused by a single braking operation in accordance with the invention;

[0027] [Fig.3] [Fig.3] represents, in the form of a flowchart, the main steps of a method for configuring a device for estimating individual wear of a brake according to a first embodiment;

[0028] [Fig.4] [Fig.4] represents a configuration system of a device for estimating individual wear of a brake in accordance with this first embodiment;

[0029] [Fig.5] [Fig.5] represents a configuration system of a device for estimating individual wear of a brake according to a second embodiment;

[0030] [Fig.6] [Fig.6] represents a neural network of the configuration system of [Fig.5];

[0031] [Fig.7] [Fig.7] represents, in the form of a flowchart, the main steps of a method for configuring a device for estimating individual wear of a brake in accordance with this second embodiment;

[0032] [Fig.8] [Fig.8] represents the hardware architecture of a system for determining the wear of a brake caused by a single braking, in a particular embodiment of the invention; and

[0033] [Fig.9] [Fig.9] represents the hardware architecture of a system for configuring a device for estimating individual wear in a particular embodiment. Detailed description of embodiments

[0034] [Fig.l] represents, in its environment, an SDU system for determining the wear of a brake caused by a single braking operation in accordance with the invention.

[0035] In the embodiment described herein, this SDU system is used to determine the wear of a brake F in an AER aircraft.

[0036] It comprises a MOD_OBTSDu module for obtaining state variables xl of the aircraft at the time of braking / and a CV device configured to determine the wear <5 of the brake F caused by this single braking / as a function of these state variables xl and wear modeling parameters.

[0037] This CV device can be considered as a virtual brake wear sensor.

[0038] The state variables xl are for example obtained from data measured by sensors C; of the aircraft and / or from data provided by a CAL computer of the aircraft.

[0039] These state variables are for example a temperature and a pressure at the brake F at the time of braking / and a duration of braking / provided by the CAL computer.

[0040] In one embodiment, the wear § of the brake F caused by the braking f is recorded in a memory M of the wear determination system SDU.

[0041] [Fig. 2] represents, in the form of a flowchart, the main steps of a PDU method for determining the wear of a brake caused by a single braking operation in accordance with the invention.

[0042] This method comprises a step E10 of obtaining state variables xl of the aircraft at the time of braking / and a step E20 of determining the wear 5 of the brake F caused by this single braking / as a function of these state variables xl and wear modeling parameters 0 / .

[0043] In Figure 1, a real sensor CR is shown configured to measure the wear of the brake F. This real sensor is not precise enough to measure the wear 5 of the brake F caused by a single braking f, but only the wear of the brake caused by a series of brakings.

[0044] We will now describe a method and a system for configuring the CV wear determination device in two embodiments. First embodiment

[0045] [Fig. 3] represents, in the form of an organizational chart, the main stages of a PCD1 method for configuring a device for determining wear individual of a brake F according to a first embodiment.

[0046] We note Ôÿ the wear caused by the ith braking of a braking sequence Sj.

[0047] It is recalled that the wear Ôa caused by braking f. ■ depends on state variables xl J IJ and we note . the values ​​of these state variables at the time of this braking.

[0048] The PCD1 method comprises a step F101 of obtaining, for at least this brake F, and for at least one sequence Sj of several braking operations: - the values ​​of the state variables x\j at the time of each braking of the sequence Sj; and - the cumulative wear Aj of this brake caused by all the braking operations in sequence Sj.

[0049] The cumulative wear Ay can be measured by the real sensor CR, for example in measuring the thickness of the brake F before the braking sequence, and the thickness of the brake after the braking sequence, the cumulative wear A, being obtained by the difference between these two thicknesses.

[0050] The cumulative wear Ay- resulting from a series of braking operations can also be calculated by sensors external to the aircraft AER, for example when the braking operations are carried out during a series of tests for example. It can also be determined by a simulation system.

[0051]

[0052]

[0053]

[0054]

[0055]

[0056]

[0057]

[0058]

[0059]

[0060]

[0061] As indicated above, the actual sensor CR may not have the capacity to measure the wear Ôjj caused by a single braking of the sequence Sj but only the cumulative wear Ay resulting from all the brakings of the sequence Sj, in other words the difference between the thickness of the brake F before the first braking of the sequence and the thickness of the brake after the last braking of the sequence. In this first embodiment, it is assumed that for any single braking f^, the individual wear ôjj caused by a single braking is a linear combination of modeling parameters 0 / and functions of the state variables of this single braking With these notations: : equation (1) It is assumed that the %i functions are known. In a very simple particular case to implement, individual wear is a linear combination of the modeling parameters and the state variables ^.. AJ= : equation (2) This equation (2) can be written in matrix form in the form A = A0 where A, 0 and A represent respectively the said state variables ^., the modeling parameters 0 / and the cumulative wear Ay. The PCD1 method includes a step F201 for determining the modeling parameters 0 / from the state variables x[^ of the cumulative wear Ay and possibly the functions 8 / in the general case of equation (1). This step F201 is a step for solving a linear system SL. It is known to those skilled in the art. It makes it possible to determine the modeling parameters 0 / which can be used to configure the wear determination device CV of [Fig.l]. [Fig.4] represents a SYSL1N system for configuring a device for determining individual wear of a brake F according to a first embodiment. In the embodiment described here, this SYSlin system uses a database BD which includes, for at least this brake F, and for at least one sequence Sj of several braking operations: - the cumulative wear Ay of this brake caused by all the braking operations of the sequence Sj; and - the values ​​of the state variables Xy at the time of each of these braking operations.

[0062] The SYSL1N system includes a MOD_OBTCDe module configured to obtain the cumulative wear Aj and the values ​​of the state variables from the base of BD data.

[0063] In the embodiment described here, the SYSlin system comprises a MOD_SL module configured to solve the aforementioned linear system SL to determine the modeling parameters 0 / from the state variables x\p of the cumulative wear Ay and possibly the functions ^z in the general case of equation (1). Second embodiment

[0064] [Fig.5] represents a SYSNN system for configuring a device of determination of individual wear of a brake F in accordance with a second method of realization.

[0065] We recall that the wear Ôa caused by braking f.. depends on several state variables xl and we note Xÿ the values ​​of these state variables 9 at the time of this braking.

[0066] We can note: f ( X- ; ) = &• J 0 \ / 'J

[0067] This second embodiment can in particular be implemented when the individual wears 0,; caused by the single brakings f.. cannot be modeled in the form of a linear combination of modeling parameters &i and known functions ^z of the state variables x\j of this single braking

[0068] The SYSNN system aims to determine the wear caused by each of the braking events fj of the sequence Sj, in other words to determine 5jj regardless of i.

[0069] In the embodiment described here, the SYSNN system uses a database BD identical to that of the SYSL1N system described with reference to [Fig.4]. It comprises, for at least the brake F, and for at least one sequence Sj of several braking operations: - the cumulative wear Ay of this brake caused by all the braking operations of the sequence Sj;

[0070] And - the values ​​of the state variables x\j at the time of each of these braking operations. In the embodiment described here, the SYSNN system comprises a MOD_OBTCde module identical to that of the SYSlin system, this module being configured to obtain the cumulative wear Ay and the values ​​of the state variables x\j from the base of BD data.

[0071] In this embodiment, the configuration system SYSNN uses a neural network NN represented in [Fig.6]. In the example of [Fig.6], a series Sj of three brakings (i = 1 to 3) and three state variables (1 = 1 to 3) are assumed.

[0072] This neural network is parameterized by weights which constitute modeling parameters # within the meaning of the invention.

[0073] Given that the wear ôÿ caused by braking is necessarily positive, the neural network can advantageously present in its last layer a positivity constraint, namely an activation function FA which forces this positivity, without adding bias after the function.

[0074] For example, the activation functions ReLU (Rectified Linear Unit) and LeakyReLU (ReLU function with leakage) can be used for this purpose.

[0075] We recall that these functions are mathematically defined by: ReLU ( x ) = max(0, x) and LeakyReLU ( x ) = {x if x > 0 or ax if x < 0}, where a is a positive constant called the leak rate.

[0076] It is noted that in this neural network NN, each individual wear ôy is determined by the same sub-network of neurons nn taking as input the state variables gL of the single braking having caused this individual wear.

[0077] The neural network NN comprises an adder ADD configured to sum the individual wears ôy of the brakings of the series Sj.

[0078] The parameters 0 are determined so as to minimize the difference between the sum of the individual wears dfj calculated by the adder ADD and the cumulative wear A / resulting from all the braking of the sequence Sj.

[0079] The loss function to be optimized according to the parameters S can be written: 100801 i(0)=

[0081] Parameters 6 can be used to configure the CV wear determination device of [Fig.l].

[0082] [Fig.7] represents, in the form of a flowchart, the main steps of a PCD2 method for configuring a device for determining individual wear of a brake F in accordance with this second embodiment.

[0083] The PCD2 method comprises a step F102 of obtaining, for at least this brake F, and for at least one sequence Sj of several braking operations: - the values ​​of the state variables at the time of each braking of the sequence Sj; and - the cumulative wear Ay- of this brake caused by all the braking operations in sequence Sj.

[0084] This step is similar to step F101 already described in the PCD1 method.

[0085] The method PCD2 determines, during a step F102, the parameters 9 of the neural network NN of figure 6 which minimize the difference between the sum of the individual wears Ôjj calculated by the adder ADD and the cumulative wear Ay

[0086] Parameters 8 may be used to configure the CV wear determination device of [Fig.l]. Sequence normalization

[0087] In one embodiment, the sequences Sj do not all have the same number Jj of braking. To take this into account, equations (1) and (2) can be normalized by the inverse of the number of brakings.

[0088] This gives us standardized cumulative braking, comparable regardless of the number of braking operations in the sequence. I0089! A'y = J-Ay = (1))

[0090] = iM, (equation (2)) ' / j JJJ l- t

[0091] Similarly, the values ​​of the state variables x\j can be normalized by -L at the input of the NN network. •h

[0092] In one embodiment, and as shown in [Fig.8], the SDU system for determining brake wear has the hardware architecture of a computer. It comprises a processor 10A, a random access memory 10B, a read only memory 10C, a non-volatile flash memory 10D, input / output means 10E as well as communication means 10F.

[0093] These communication means 10F allow in particular the SDU system to obtain the state variables xl of the aircraft at the time of braking f.

[0094] The read-only memory 10C constitutes a recording medium in accordance with the invention, readable by the processor 10A and on which is recorded a computer program PGSDu in accordance with the invention, comprising instructions for executing the steps of a method for determining the wear of a brake as described in [Fig.2].

[0095] In one embodiment, and as shown in [Fig.9], the SYSNN, SYSL1N system for configuring a device for determining individual wear of a brake has the hardware architecture of a computer. It comprises a processor 20A, a random access memory 20B, a read-only memory 20C, a non-volatile flash memory 20D, input / output means 20E as well as communication means 20F.

[0096] These communication means 20F allow in particular the configuration system to obtain for at least one braking sequence Sj: - the cumulative wear Aj of this brake caused by all the braking operations in the sequence Sj;et - the values ​​of the state variables x\j at the time of each of these braking operations.

[0097] The read-only memory 20C of the drive device constitutes a recording medium readable by the processor 20A and on which is recorded a computer program PROGPCd, comprising instructions for executing the steps of a method for configuring a device for determining individual wear of a brake as described in two embodiments with reference to FIGS. 3 and 7.

Claims

Claims

1. Method (PDU) for determining the wear (5) of a brake caused by a single braking operation (y) comprising: - a step of obtaining (E10) state variables (xl) of said braking; and - a step of determining (E20) said wear (5) using a wear determination device (CV) which takes into account said state variables (x1) and wear modeling parameters (dp 9), said wear modeling parameters (^ 9) being previously determined by implementing the following steps: - for at least one sequence (Sj) of single braking operations (f \'- Vu) • a. obtaining (F101, F102) state variables of each of said single braking operations j of said sequence (Sj); • b.obtaining (F101, F102) a cumulative wear (Ay) of the brake caused by all the single braking operations of said sequence (Sj); - determining (F201, F202) said modeling parameters of said determination device (CV) from said state variables and said cumulative wear (Ay).

2. Method (PDU) for determining wear according to claim 1 in which the determination (F202) of said modeling parameters comprises: - for each single braking f Y a determination of V i J / the individual wear caused by this braking from the state variables (x^ of this braking using said determination device (CV); - a determination of said modeling parameters (y) to minimize a difference between the sum of the individual wear) individual wear and said cumulative wear (Ay).

3. Method (PDU) for determining wear according to claim 2 in which each of said individual wears caused by a single braking is determined by the same neural network taking as input the state variables of this single braking.

4. Method (PDU) for determining wear according to claim 1 wherein the determination (F201) of said modeling parameters comprises a resolution of a linear system which defines said cumulative wear (Ay) of said at least one sequence (Sj) as a sum of the individual wears caused by the single brakings / ?) of said sequence, each individual wear} caused by a single braking being a linear combination of said modeling parameters and functions j of said state variables ^(3 of this single braking fy \

5. Method (PDU) for determining wear according to claim 4 in which the linear system which defines said cumulative wear (Ay) of said at least one sequence (Sj) is: A = A.0 where A, 0 and A respectively represent said state variables (X / y) ' said modeling parameters and said at least one cumulative wear.

6. Method (PDU) for determining wear according to any one of claims 1 to 5 in which said state variables of each of said single braking operations j are included among a temperature, a pressure, a duration, measured during said single braking operation.

7. System (SDU) for determining the wear (5) of a brake caused by a single braking (y) comprising: - a module (MOD_OBTSDu) for obtaining state variables (X1) of said braking; and - a wear determination device (CV) which takes into account said state variables (x^ and wear modeling parameters (¾), said wear modeling parameters ( ^ 0) being previously determined by implementing the following steps: - for at least one sequence (Sj) of single braking / 1: V97 • c. obtaining (F 101, F102) state variables of each of said single braking / ? \ of said r 97 sequence (Sj); • d. obtaining (F 101, F102) a cumulative wear (Ay) of the brake caused by all the single braking operations of said sequence (Sj); - determining (F201, F202) said modeling parameters of said determination device (CV) from said state variables) and said cumulative wear (Ay).

8. Computer program (PGPCd) comprising instructions for executing the steps of a wear determination method according to any one of claims 1 to 6 when said program is executed by a computer.

9. A medium (10C) comprising a computer program according to claim 8.