Method for monitoring the health status of aircraft turbomachinery

Abstract

One aspect of the invention relates to a method for monitoring the health status of an aircraft turbomachine for a flight of interest (VI), based on a setpoint vector of input parameters relating to the turbomachine of interest in steady-state operation. The method uses a database of transient input and output parameter values ​​for a plurality of flights, including the flight of interest, and steady-state input and output parameter values ​​for a plurality of flights excluding the flight of interest. The transient data are used to estimate a transient prediction model f. This transient prediction model f is then used, in conjunction with the steady-state data, to estimate a steady-state prediction model (HVI) of the turbomachine of interest for the flight of interest and to determine the steady-state behavior of the turbomachine for the flight of interest. Figure to be published with the abstract: Figure 1 [Fig. 1].

Description

Title of the invention: Method for monitoring the health status of an aircraft turbomachine. TECHNICAL FIELD OF THE INVENTION

[0001] The technical field of the invention is that of monitoring the condition of aircraft turbomachinery.

[0002] The present invention relates to a method for monitoring the health status of an aircraft turbomachine. TECHNOLOGICAL BACKGROUND OF THE INVENTION

[0003] The "Health monitoring" of an aircraft turbomachine, for example an airplane or a helicopter, makes it possible to monitor the condition of the turbomachine throughout its life and to anticipate a defect of the turbomachine from data recorded during flights carried out by the aircraft including the turbomachine.

[0004] A classic indicator of health monitoring is the calculation of margins that allow comparison, for a desired value of at least one input parameter relating to the turbomachine, of a modeled theoretical value of at least one output parameter relating to the turbomachine with an actual value of said output parameter. For example, to obtain a desired power P emitted by the turbomachine, a physically modeled temperature To in the turbomachine's combustion chamber is theoretically required. The temperature To is compared to an actual temperature Ti in the turbomachine's combustion chamber required to obtain the power P emitted by the turbomachine.The actual temperature Ti increases throughout the turbomachine's lifecycle, and the difference between the theoretical value To and the actual value Ti is a margin studied and analyzed using physical models. This allows for the anticipation of potential failures and overheating of the turbomachine. The margin calculation can be performed for several input parameter values ​​(for example, the rotational speed of one turbomachine shaft and the rotational speed of a second turbomachine shaft) and several output parameter values ​​(for example, the power output of the turbomachine and the aforementioned turbomachine temperature).

[0005] Physical models allow the study of margins using turbomachine parameter data recorded during a stabilized turbomachine regime. However, a helicopter, for example, performing surveillance or slinging operations, may rarely or never operate in a stabilized regime during a particular flight, making margin calculations complex or even impossible.

[0006] In order to overcome the lack of steady-state data for a helicopter turbomachine, there are state-of-the-art solutions for monitoring the engine condition using indicators other than margins.

[0007] An alternative to the solution proposed above, also known in the prior art, relies on estimating, from transient input parameter values ​​of the turbomachine during a given flight, steady-state output parameter values ​​of the turbomachine. This solution allows for the independent transformation of two turbomachine parameters to obtain steady-state output parameter values ​​for given input parameters. However, the proposed solution only works for two input parameters at a time, and each parameter is transformed independently of the other to obtain a steady-state output parameter. Furthermore, the margin can therefore only be calculated on the transformed parameters, and not for any value of the input parameters, which further limits a relevant margin calculation.

[0008] There is therefore a need to estimate the behavior of the turbomachine in steady state, from one or more input parameters relating to the turbomachine in transient state, without considering the input and output parameters independently of each other. Summary of the invention

[0009] The invention offers a solution to the problems mentioned above, by allowing more efficient estimation of the steady-state behavior of a turbomachine for a particular flight.

[0010] One aspect of the invention relates to a method for monitoring the health status of an aircraft turbomachine of interest during a flight of interest VI of the aircraft, based on a setpoint matrix Xsc comprising at least one value of at least one input parameter relating to the turbomachine of interest, the method comprising the following steps: • Construction of a training database comprising: • a so-called transient set T comprising transient learning data recorded during at least one instant of flight recording of one or more flights V of a turbomachine TM operating in transient regime and of the flight of interest VI of the turbomachine of interest TMI; • a so-called stabilized set S comprising stabilized learning data recorded at each instant of flight recording of each flight V, with the exception of the flight of interest VI, of each turbomachine TM operating in stabilized regime; • Estimation of a transient prediction model f on the transient training data of the transient set T; • For each flight V of each turbomachine TM, with the exception of the flight of interest VI, estimation of a stabilized prediction model Hv from the transient prediction model f and the stabilized training data of each instant of flight recording of flight V, included in the stabilized set S; • Construction of a stabilized HVI prediction model for flight of interest VI, from at least one stabilized HC prediction model estimated for a flight C, flight C being one of the flights V(s), • Estimation of an interest matrix YSI representing the behavior of the turbomachine of interest TMI in stabilized regime for the flight of interest VI from the stabilized prediction model HVI, the transient prediction model f and the setpoint matrix Xsc.

[0011] By "a TM turbomachine" we mean at least one TM turbomachine.

[0012] By "input parameter relating to a turbomachine" is meant a parameter relating to the turbomachine for which a setpoint value is desired.

[0013] By "behavior of a turbomachine in stabilized regime for a flight V" we mean the estimation of at least one output parameter relating to the turbomachine & of each input parameter of said turbomachine.

[0014] Thanks to the invention, the steady-state behavior of a turbomachine of interest, for any flight of interest, can be estimated from transient and steady-state data relating to at least one turbomachine in transient and steady-state operation for a plurality of flights, using the HVi steady-state prediction model. Thus, unlike the prior art, the invention not only allows the prediction of output data relating to the turbomachine in steady state but also provides a model that allows the prediction of output data from any setpoint. The estimation of the HVi model therefore makes it possible to calculate margins at different points rather than being constrained to a transformation of the transient variables, and thus to monitor the state of the turbomachine for any flight of interest.

[0015] Advantageously, the invention makes it possible to take into account all the data relating to the turbomachine at once rather than in pairs.

[0016] Advantageously, the invention allows the construction of a stabilized HVi prediction model which makes it possible to calculate margins at different points rather than being constrained to a transformation of the transient variables.

[0017] Advantageously, the training data can be taken without direct relation to the flight of interest.

[0018] In addition to the characteristics mentioned in the preceding paragraph, the method according to one aspect of the invention may have one or more additional characteristics from among the following, considered individually or in all technically possible combinations: • The transient learning data for the assembly T are constructed, for each flight V of each turbomachine TM, from a torque • the Xyp matrix comprising at least one row and at least one column, each row corresponding to an input parameter of the turbomachine TM in transient regime, each input parameter of each row being different from the other input parameters of the other rows, each column corresponding to a recording instant of flight V, each recording instant of each column being different from the other recording instants of the other columns, • the W yp matrix comprising at least one row and at least a column, each row corresponding to an output parameter of the TM turbomachine in transient regime, each output parameter of each row being different from the other input parameters of the other rows, each column corresponding to a recording instant of flight V, each recording instant of each column being different from the other recording instants of the other rows, The stabilized learning data for the set S are constructed, for each flight V of each turbomachine TM, with the exception of the flight of interest, from the torque ( %s, *VS • the Xyg matrix comprising at least one row and at least a column, each row corresponding to an input parameter of the turbomachine TM in transient regime, each input parameter of each row being different from the other input parameters of the other rows, each column corresponding to a recording instant of flight V, each recording instant of each column being different from the other recording instants of the other columns, the Wyg matrix comprising at least one row and at least one column, each row corresponding to an output parameter of the TM turbomachine in transient regime, each output parameter of each line being different from the other input parameters of the other lines, each column corresponding to a recording instant of flight V, each recording instant of each column being different from the other recording instants of the other columns. The TM turbomachine is the turbomachine of interest. Thus, among the couples of matrices ( vy VT' of the transient set T, at least one of these The coupler represents a flight of the turbomachine of interest. This characteristic allows the inclusion of data relating to the turbomachine of interest for at least one flight, enabling the estimation of the prediction model f, the model f allowing the determination of the behavior for a flight from data retrieved on the turbomachine of interest. A flight V of the assembly T is the flight of interest VI of the turbomachine of interest and the set T is further constructed from the pair ( ). \^VIT' ^VIT This feature of including parameters relating to the turbomachine of interest for the flight of interest, allowing estimation of the prediction model f, allowing representation of the turbomachine of interest for the flight of interest. • In one embodiment, the transient training data for set T includes: • A matrix Xy resulting from a horizontal concatenation of each matrix Xyy, • A matrix Y j resulting from the horizontal concatenation of each matrix H^y^, And the transient prediction model f is defined such that lïXy) = Y -p. • for each flight V of each turbomachine TM of the stabilized assembly S, with the exception of the flight of interest VI of the turbomachine of interest TMI: • the stabilized learning data of set S include the pair (y) • The stabilized prediction model Hv is defined such that, Hv(f(XyS)' XyS)— M'yg • The stabilized prediction model HVi is constructed from at least the stabilized prediction model Hc chosen according to a condition Cl. Advantageously, this feature allows for choosing the most relevant stabilized prediction model HVi for the flight of interest VI. • The condition Cl is: • The distance between y and is minimal; • The cost function of the stabilized prediction model Hc is minimal; or • The generalization error of the stabilized prediction model Hcest is minimal; • The prediction model f and the prediction model H are each chosen from a neural network, a decision tree, a random forest, a wide-margin separator, a polynomial regression, or a linear regression. • at least one input parameter relating to a turbomachine is one of the following operating parameters of the turbomachine: • Rotation speed of the first shaft of the turbomachine; • Rotation speed of a second shaft of the turbomachine; • Temperature of a turbomachine combustion chamber; • Torque delivered by the turbomachine; • Any other relevant thermodynamic parameter or a parameter specific to the given environment in which the turbomachine is located, from among the following: • Temperature of the given environment; • Pressure from the given environment. • Pumping power • Any other parameter influencing the thermodynamic behavior of the turbomachine (e.g., valve) • at least one output parameter relating to a turbomachine is one of the following operating parameters of the turbomachine: • Rotation speed of the first shaft of the turbomachine; • Rotation speed of a second shaft of the turbomachine; • Temperature of a turbomachine combustion chamber; • Torque delivered by the turbomachine; • Any other relevant thermodynamic parameter or a parameter specific to the given environment in which the turbomachine is located, from among the following: • Temperature of the given environment; • Pressure from the given environment; • Pumping power; • Any other parameter influencing the thermodynamic behavior of the turbomachine (e.g., valve).

[0019] Another aspect of the invention relates to a computer program product comprising instructions which, when the program is executed by a computer, lead the computer to implement the process according to the invention.

[0020] Another aspect of the invention relates to a computer-readable recording medium comprising instructions which, when executed by a computer, cause the computer to implement the process according to the invention.

[0021] The invention and its various applications will be better understood by reading the following description and examining the accompanying figures. BRIEF DESCRIPTION OF THE FIGURES

[0022] The figures are presented for illustrative purposes only and are in no way limiting of the invention. • Fig. 1 shows a schematic representation of a synoptic diagram of the process according to the invention; • Fig. 2 is an example of data recovered during the following flights: (Flight 0, Flight 1, Flight 2, Flight VI) of the turbomachine of interest. DETAILED DESCRIPTION

[0023] The figures are presented for illustrative purposes only and are not in any way limiting to the invention.

[0024] A first aspect of the invention relates to a method for monitoring the health status of an aircraft turbomachine of interest TMI for a flight of interest VI, from a setpoint matrix Xsc comprising at least one input parameter relating to the turbomachine of interest TMI in stabilized regime.

[0025] The aircraft is, for example, an airplane or a helicopter.

[0026] The term "turbomachine" means a system employing a gas turbine, in which energy is transferred between a rotating part and a gas.

[0027] The following characteristics are valid for any type of turbomachine within the scope of the invention, including the turbomachine of interest.

[0028] A turbomachine can include a combustion chamber, a first shaft, a second shaft.

[0029] The term "shaft" means a mechanical element that transmits power in the form of torque and rotational motion.

[0030] A turbomachine is for example a turbojet, a turboprop or preferably a turbomotor.

[0031] By "input parameter relating to a turbomachine" is meant a parameter relating to the turbomachine for which a setpoint value is desired.

[0032] An input parameter relating to the turbomachine can be one of the following operating parameters of the turbomachine: rotational speed of the The first shaft of the turbomachine; the rotational speed of the second shaft of the turbomachine; the combustion chamber temperature of the turbomachine of interest; and the torque and power delivered by the turbomachine. The turbomachine input parameter may also be a parameter specific to the external environment in which the aircraft containing the turbomachine is located during a given flight, such as: external temperature; external pressure. A turbomachine input parameter may also be any relevant thermodynamic or environmental parameter.

[0033] For example, two input parameters can be chosen from the parameters relating to the turbomachine previously: a setpoint value for the power delivered by the turbomachine and a setpoint value for the rotational speed of the first shaft of the turbomachine can be chosen or required.

[0034] An output parameter of the turbomachine is a parameter that is a consequence of one or more input parameters of the turbomachine and the state of the turbomachine.

[0035] At least one output parameter relating to the turbomachine of interest may be one of the following operating parameters of the turbomachine: the rotational speed of the first shaft of the turbomachine; the rotational speed of the second shaft of the turbomachine; the temperature of the combustion chamber of the turbomachine, torque delivered by the turbomachine and power delivered by the turbomachine.

[0036] For example, if the input parameters are: the rotational speed of the first shaft of the turbomachine, the rotational speed of the second shaft of the turbomachine, the temperature of the given external environment; the pressure of the given external environment, the output parameters relating to the turbomachine can be the temperature of the combustion chamber of the turbomachine, and the torque delivered by the turbomachine.

[0037] In the following text, the phrase "turbomachine-related parameters" includes the turbomachine-related input parameter(s) and the turbomachine-related output parameters.

[0038] By "steady state of a turbomachine" is meant a state during which the input and output parameters relating to the turbomachine do not change or change very little over time. A steady state can be called permanent or stationary state.

[0039] A steady-state regime of a turbomachine is opposed to a transient regime of a turbomachine, during which input parameters and output parameters relating to the turbomachine evolve over time.

[0040] The setpoint matrix Xsc is a matrix comprising at least one value of at least one input parameter relating to the turbomachine of interest, each value being able to be randomly generated or chosen. The matrix Xsc may have a number of rows greater than or equal to 1 and a number of columns greater than or equal to 1. The coefficients of the matrix Xsc are denoted (xscj)i>0jj>0.

[0041] Fig. 1 is a synoptic diagram of the process 100 according to the invention.

[0042] The process 100 may include a first step 101 of constructing a base training data.

[0043] Construction step 101 may include a first substep 1011 of data recovery D, for at least one turbomachine TM, the data D being recorded during one or more moments of flight recordings during one flight V or several flights V of said turbomachine TM.

[0044] According to one embodiment, the turbomachine TM is the turbomachine of interest TMI.

[0045] According to one embodiment, the data D comprises recorded data during one or more moments of flight of a plurality of flights (Vo, ... VN)N >0 of the same turbomachine.

[0046] According to one embodiment, the data D comprises data recorded for a plurality of turbomachines (TM0, ..., TMk)k>i during one or more flight recording moments of a plurality of flights (V0.Tm o, ..., Vn o, V0-Tm k-..., Vp. TMK)p>id of a plurality of turbomachines. For each turbomachine in the plurality of turbomachines, the number of flight moments for each flight in the plurality of flights may be different or the same from one flight to another and / or from one turbomachine to another.

[0047] According to an embodiment in which the turbomachine TM is the turbomachine of interest TMI, the flight V carried out by the turbomachine of interest is carried out prior to the flight of interest VI.

[0048] According to one embodiment, the recovered data D includes, for each turbomachine TM and for each flight V of the turbomachine TM, transient data recorded during the transient regime of the turbomachine TM during at least one instant of flight of each flight V of the turbomachine TM, the flight V being able to be the flight of interest VI, and includes stabilized data recorded during the stabilized regime of the turbomachine TM during at least one instant of flight of each flight V, with the exception of the flight of interest Vb of the turbomachine TM.

[0049] According to one embodiment, the recovered data D comprises, for each turbomachine TM and for each flight V of the turbomachine TM, transient data recorded during the transient regime of the turbomachine TM during at least one flight instant of each flight V of the turbomachine TM and during at least one flight instant of the flight of interest VI, and comprises stabilized data recorded during the stabilized regime of the turbomachine TM during at least one instant of flight of each flight V, with the exception of the flight of interest Vb of the turbomachine TM.

[0050] According to one embodiment, for each turbomachine TM and for each given flight V of the turbomachine TM the transient data include a pair of matrices ( \ ^VT' Y VIT

[0051] According to the preceding embodiment, each matrix XyT comprises at least one row and at least one column, each row corresponding to an input parameter of the turbomachine TM, each input parameter of each row being different from the other input parameters of the other rows, and each column corresponding to a recording instant of flight V during a transient phase, each recording instant of each column being different from the other recording instants of the other columns.

[0052] Let xTVjj be a coefficient of the matrix xTv_ÿ being equal to the value of the input parameter i for the recording time j during so-called transient phases of the given flight V. The coefficients i and j are respectively natural numbers greater than or equal to 0. The index V represents the given flight V and the index T represents the transient regime.

[0053] For example, for a flight V of a turbomachine TM in transient regime, for which the values ​​of p input parameters relating to the turbomachine TM are measured, p being an integer greater than 0, the values ​​being measured for 1 flight instant recordings, 1 being greater than 0, the matrix XyT is as follows:

[0054] XTy_oo xTV_01' XVT~ = '■ = ,XTV^ ■■■ XTV_pl,

[0055] According to the preceding embodiment, each matrix yp comprises at least one row and at least one column, each row corresponding to the different output parameters of the turbomachine TM in transient regime, each column corresponding to the different times of recording of the flight v during a transient phase.

[0056] Let yTVjj be a coefficient of the matrix Wyr yTVjj the value of the output parameter i for the recording time j during so-called transient phases of the given flight V. The coefficients i and j are respectively natural numbers greater than or equal to 0. The index V represents the given flight V and the index T represents the transient regime.

[0057] For example, for a flight V of a turbomachine TM in steady state, for which the values ​​of q are input parameters relating to the turbomachine TM next: ^VT~ are measured, q being an integer greater than 0, the values ​​being measured for 1 flight time recordings, 1 being greater than 0, the matrix y^ is the VT ... VT V_00 V 01 yTV_qO "■ yTV_ql,

[0058] For each flight V of the turbomachine TM, the flight recording times of XyT correspond to the recording times of Wy?.

[0059] According to one embodiment, the transient data comprise, for each flight V of each turbomachine TM, the matrices (Xy^ and (Wy-jX, (Xyj / being the transpose of the matrix Xyj- and (Wyp)T being the transpose of the matrix Wyy-

[0060] According to an embodiment complementary to the preceding embodiment, for each turbomachine TM and for each given flight V of the turbomachine TM, with the exception of the flight of interest VI of the turbomachine of interest, the stabilized data comprise a pair of matrices (i Xys' ^vsr

[0061] Each Xyg matrix comprises at least one row and at least one column, each row corresponding to an input parameter of the TM turbomachine, each input parameter of each row being different from the other input parameters of the other rows, and each column corresponding to a recording instant of the V flight in stabilized phase, each recording instant of each column being different from the other recording instants of the other columns.

[0062] Let xsv_ij be a coefficient of the matrix XyS> xsv_ÿ being equal to the value of the input parameter i for the recording time j during so-called stabilized phases of the given flight V. The coefficients i and j are respectively natural numbers greater than or equal to 0. The index V represents the given flight V and the index T represents the transient regime.

[0063] For example, for a flight V of a turbomachine TM in steady state, for which the values ​​of p input parameters relating to the turbomachine TM are measured, p being an integer greater than 0, the values ​​being measured for r flight time records, r being greater than 0, the matrix XyS is as follows:

[0064] '^VOO ■■■ XSy0r' tX^y■■■ X^y_pr

[0065] Each matrix W ys, each comprising at least one column and at least one line, each line corresponding to the different output parameters of the turbomachine TM in transient regime, each column corresponding to the different times of recording of flight V.

[0066] Let ys v_ÿ be a coefficient of the matrix yg ysvij the value of the output parameter i for the recording time j during so-called transient phases of the given flight V. The coefficients i and j are respectively natural numbers greater than or equal to 0. The index V represents the given flight V and the index T represents the transient regime.

[0067] For example, for a flight V of a turbomachine TM in steady-state operation, for which the values ​​of q input parameters relating to the turbomachine TM are measured, q being an integer greater than 0, the values ​​being measured for r flight time records, r being greater than 0, the matrix Wvs is the next: yS ... y S V_qO Vçr,

[0068] For each flight V of each turbomachine TM, the flight recording times of XyS correspond to the recording times of W y g.

[0069] According to the embodiment in which the data D include transient data recorded for the turbomachine of interest TM! during at least one instant of flight of the flight of interest VI, said transient data include matrices Zy / Tet W y JT-

[0070] According to an embodiment complementary to the previous embodiment, for each turbomachine TM, and for each given flight V of the turbomachine TM, the distance between the matrices Xyj y el X yy is less than a threshold SL

[0071] According to an embodiment compatible (but not exclusive) with the preceding embodiment, the distance between the matrices Wyjy^ VT is less than one threshold S2.

[0072] The distance between the matrices can be any relevant distance, for example a Manhattan distance, a Euclidean distance, a Minkowski distance or a Chebyshev distance.

[0073] Fig. 2 is an example of data recovered during the following flights: (Flight 0, Flight 1, Flight 2, Flight VI) of the turbomachine of interest.

[0074] The first step 101 of the method according to the invention includes a second substep 1012 of distributing the data D into two sets: a so-called transient set T and a so-called stabilized set S. The transient data, for each turbomachine TM and for each flight V of the turbomachine TM, are used to construct transient training data included in set T, and the stabilized data for each turbomachine TM and for each flight V of The TM turbomachines are used to build stabilized training data, and are distributed throughout the set S.

[0075] According to one embodiment, the transient learning data of the set T are constructed, for each flight V of each turbomachine TM, the flight V being the flight of interest VI, from the pair (XyT, \ and the data Transient learning curves of the set S are constructed, for each flight V of each turbomachine TM, with the exception of the flight of interest VI, from the couple ^VS' w ï VS /

[0076] According to an embodiment complementary to the previous embodiment, the stabilized learning data of the set S comprises, for each turbomachine TM, and for each flight V, with the exception of the flight of interest VI of the turbomachine of interest, the pair of matrices (XyS, W vs^

[0077] According to one embodiment, the transient learning data of the set T includes a pair of matrices y constructed from each pair (Xyj-, W for each flight V of each turbomachine TM.

[0078] According to the preceding embodiment, the matrix Xy results from the horizontal concatenation of each matrix XyT. Thus, when the data D includes transient data recorded for at least one flight instant recording from a plurality of flights (Vo,...,Vk)k>0, the matrix XT is of the following form:

[0079] XT=[Z0T XkT]

[0080] In order to simplify the notations, each flight Vk has been denoted k in the matrix X^.

[0081] According to the preceding embodiment, the matrix Y resulting from the concatenation horizontal of each matrix W y

[0082] Thus, when the data D includes transient data recorded for at least one flight instant recording of a plurality of flights (Vo,...,Vk)k>0, the matrix YT is of the following form: [00831 yr = [V0T ...

[0084] To simplify the notation, each flight Vk has been denoted k in the matrix Y

[0085] Each column of X^ and Y? represents the same flight V.

[0086] The method further includes a second estimation step 102 of a transient prediction model f on the training data of the transient set T.

[0087] According to the embodiment, wherein the transient training data of set T comprise the matrix pair (XT Y^ estimation step 102 of the transient prediction model f models the relationship between the matrix Xj and the matrix YT such that f(_y ) — y , f is said to be global transient prediction model.

[0088] According to the preceding embodiment, the transient prediction model f can be a model chosen from among all classical regression models, for example a neural network, a regression tree, a random forest, a wide-margin separator, or linear regression. The linear regression can be polynomial regression, for example.

[0089] According to the preceding embodiment, the estimation of the transient prediction model f is carried out by minimizing a cost function corresponding to the error between the output data ^X^) provided by the transient prediction model f and the true output data Y ^desired.

[0090] The cost function is, for example, the root mean square deviation between f(X^) and Y^. As a reminder, the matrix XT can be equal to the columns )v>i On denote (f(xTVjj))i>o,j>ov>oles coefficients of the matrix f(Xr) - Thus, the cost function can be equal to:

[0091] The method 100 further includes, for each flight V of each turbomachine TM, with the exception of the flight of interest VI, an estimation step 103 of a stabilized prediction model Hv from the transient prediction model f and the stabilized training data of each instant of flight recording V, included in the stabilized set S.

[0092]

[0093]

[0094]

[0095] According to a first embodiment, each stabilized prediction sub-model Hv for each flight V of the stabilized ensemble S is estimated independently of the other stabilized prediction models. Each Hv stabilized prediction model can be chosen from among all classical regression models, for example, a neural network, a regression tree, a random forest, a large-margin separator, or linear regression. Linear regression can be polynomial regression, for example. According to the first embodiment, the estimation of each stabilized prediction model Hv is performed by minimizing a cost function corresponding to the error between the output data Hv 1 ) provided by the prediction model stabilized Hv and the true output data V^ygdesired. According to an embodiment compatible with the first preceding embodiment, for the estimation of each stabilized prediction model Hv, the cost function is, for example, the mean squared deviation between Hv and v. ^■vsr % vs and Wys—As a reminder, the coefficients of the matrix are denoted (xTv_ij)i0jj>o and the coefficients of the matrix V^y^s are denoted (yTv_ij)i>o,j>o. We will denote (f(xsv_ij))i>o,j>o the coefficients of the matrix fv(Xy$). Thus, the cost function can be equal to:

[0096] The minimization of the cost function can be achieved using the gradient descent algorithm, the least squares algorithm, or any state-of-the-art optimization method that allows the estimation of each sub-model Hv

[0097] According to a second embodiment, each stabilized prediction model Hv for each flight V of the transient set S is estimated depending on the other stabilized prediction models using a multi-task learning method.

[0098] Multitasking is a subfield of machine learning that allows for the simultaneous resolution of several different tasks while taking into account the dependencies between the tasks. A multitasking learning model includes a part common to each task and a part specific to each task. Multitasking makes it possible to improve the learning of a particular model by using the features included in all the tasks. Thus, according to the second embodiment, for each flight V of each turbomachine TM, each prediction submodel Hv is equal to V o hv, the function being common to each model Hvet and the function fv being specific to each submodel fv.

[0099] The multitask learning model can be a model chosen from a neural network, a regression tree, a random forest, a wide-margin separator or linear regression.

[0100] For example, Each prediction model Hv can be associated with a set of training parameters, part of which is common to all other prediction sub-models, and part specific to said model.

[0101] This second embodiment is advantageous because it allows a reduction of the cost function allowing the estimation of each stabilized prediction model Hv in the case where the stabilized data of the set S are noisy or few in number.

[0102] The method 100 further comprises a step of constructing a stabilized prediction model Hvi for the flight of interest VI, from at least one stabilized prediction model Hv>

[0103] According to one embodiment, the stabilized prediction model HVi is constructed from at least one stabilized prediction model Hc chosen, the prediction model Hc being estimated in step 103 for a flight C, flight C being chosen from a plurality of flights V when there are several flights V, and being the flight V when there is only one flight V. The stabilized prediction model Hc is chosen according to a condition Cl.

[0104] The stabilized prediction model Hvi can be constructed from a chosen prediction model Hc such that HVi can be proportional to or equal to the stabilized prediction model Hc

[0105] According to an embodiment in which the transient training data included in the set T are constructed from a plurality of matrix pairs (X yT,W y? for each flight V, of which the pair (XCT,... \ corresponding CT) at flight C, and from the pair of matrices (Xyjy ... Y the condition Cl allowing Y to live) choosing a prediction model Hc to build the HVi model may be a condition on the distance between the matrices Xy^ and XCT.

[0106] For example, the distance between the matrices XyIT and XCy is less than a threshold S3, S3 being for example a positive real number or zero.

[0107] For example, the distance between the matrices XyjT and Xyy is minimal compared to the respective distances between each matrix XyT for each vol V and the matrix ^VIT •

[0108] According to an embodiment in which the stabilized training data included in the set S are constructed from a plurality of matrix pairs (XyS,... 1 for each vol V, of which the pair (Xcs,... \ corresponding Wys) ^cs) For flight C, the condition Cl allowing the selection of the stabilized prediction model Hc to construct the model HVi can be a condition on the cost function minimizing, for flight C, the difference between the matrix Hv(f(\\) and the matrix X-CSr ^-vc) The condition Cl could be, for example: the stabilized prediction model Hc chosen is the model having the minimum cost function relative to the other cost functions of each stabilized prediction model Hv for each flight V of each turbomachine TM of the stabilized assembly S.

[0109] According to an embodiment in which the stabilized training data included in the set S are constructed from several pairs of matrices, for example the set of the following pairs of matrices Y i< v — N, ^vs) Given that N is an integer strictly greater than 1, and according to which stabilized prediction models (Hv) i< v — were estimated in step 103 of process 100, the stabilized prediction model HVi can be constructed from a weighted average of k stabilized prediction models among the N stabilized prediction models (Hv) i <v N, k being an integer between 1 and N, the condition Cl being on the number of k, for example the k stabilized prediction models among the set of N stabilized prediction models (Hv) R v — N having the k smallest cost functions among the N cost functions.

[0110] According to an embodiment in which the stabilized training data included in the set S are constructed from a plurality of matrix pairs (XyS, \ for each vol V, of which the pair (Xqg, \ corresponding ^vs) rCS) at flight C, the condition Cl allowing the choice of the stabilized prediction model Hc to build the model HVi can be a condition on the generalization error of the model Hc on the data included in the set S with the exception of the data concerning flight C.

[0111] The term "model generalization error" refers to the ability of the model to make robust predictions on new data not used during training.

[0112] According to the preceding embodiment, the condition Cl can be for example: the stabilized prediction model Hc chosen is the model having the minimum generalization error compared to the other respective generalization errors of each stabilized prediction model Hv for each flight V of each turbomachine TM of the stabilized assembly S.

[0113] According to an embodiment in which the stabilized training data included in the set S are constructed from several pairs of matrices, for example the set of the following pairs of matrices ((Xy Y i< v — N , ^VS) Given that N is an integer strictly greater than 1, and according to which stabilized prediction models (Hv) i< v — were estimated in step 103 of process 100, the stabilized prediction model HVi can be constructed from a weighted average of k stabilized prediction models among the N stabilized prediction models (Hv) i <v N, k being an integer between 1 and N, the condition Cl being on the number of k, for example the k stabilized prediction models among the set of N stabilized prediction models (Hv) i< v — having the k smallest generalization errors among the N generalization errors.

[0114] The method 100 includes an estimation step 105 of a matrix YSI of output parameter values ​​relating to the turbomachine of interest for at least one flight instant of the flight of interest as a function of the model Hvias associated with the vector to the matrix Xscet to the transient prediction model f.

[0115] According to one embodiment, the matrix YSiest is defined such that YS1 = HVi ( f(Xsc), X sc).

[0116] The YSi matrix corresponds to the steady-state behavior of the turbomachine of interest for the flight of interest.

[0117] The process 100 may further include a step 106 margin calculation, the margin being proportional to the difference between the matrix of interest YSi and a matrix Ymp, the matrix Yaip being the output of a physical model representing the worst case for the input Xsc. Thus, according to one embodiment, the margin calculation is carried out such that: margin = YSi - Ymp.

[0118] Monitoring the condition of the turbomachine of interest can be achieved through margin calculation.

Claims

1. Demands A computer-implemented method (100) for monitoring the health status of an aircraft turbomachine of interest for an aircraft flight of interest VI, based on a setpoint matrix Xsc comprising at least one value of at least one input parameter relating to the turbomachine of interest, the method comprising the following steps: - Construction (101) of a training database comprising: • a so-called transient set T comprising transient learning data recorded during at least one recording instant of one or more flights V of a turbomachine TM operating in transient regime and of the flight of interest VI of the turbomachine of interest TMI operating in transient regime, • a so-called stabilized set S comprising stabilized learning data recorded at each instant of flight recording of each flight V, with the exception of the flight of interest VI, of each turbomachine TM operating in stabilized regime; - Estimation (102) of a transient prediction model f on the transient training data of the transient set T; - For each flight V of each turbomachine TM, with the exception of the flight of interest VI, estimation (103) of a stabilized prediction model Hv from the transient prediction model f and the stabilized training data of each instant of flight recording of flight V, included in the stabilized set S; - Construction (104) of a stabilized HVI prediction model for the flight of interest VI, from at least one stabilized HC prediction model estimated for a flight C, flight C being one flight among the flight(s) V;

2.

3. - Estimation (105) of a matrix of interest YSI representing the behavior of the turbomachine of interest TMI in stabilized regime for the flight of interest VI from the stabilized prediction model HVI, the transient prediction model f and the setpoint matrix Xsc, - Monitoring the health status of the turbomachine of interest from margin calculations dependent on the matrix of interest YSi. Method (100) according to any one of the preceding claims wherein the TM turbomachine is the turbomachine of interest. Method (100) according to any one of the preceding claims, wherein: turbomachine TM, from a couple ( The transient learning data for the set T are constructed, for each flight V of each %VT the matrix Xy?' comprising at least one row and at least one column, each row corresponding to an input parameter of the TM turbomachine in transient regime, each input parameter of each row being different from the other input parameters of the other rows, each column corresponding to a recording instant of flight V, each recording instant of each column being different from the other recording instants of the other columns, • the Wyp matrix comprising at least one row and at least one column, each row corresponding to an output parameter of the turbomachine TM in transient regime, each output parameter of each row being different from the other input parameters of the other rows, each column corresponding to a recording instant of flight V, each recording instant of each column being different from the other recording instants of the other rows, the stabilized training data of the set S are constructed, for each flight V of each turbomachine TM, with the exception of the flight of interest, from the torque ( • the matrix Xyg comprising at least one row and at least one column, each row corresponding to an input parameter of the turbomachine TM in steady state, each input parameter of each row being different from the other input parameters of the other rows, each column corresponding to a time of flight V recording, each recording instant of each column being different from the other recording instants of the other columns, • the matrix comprising at least one row and at least one column, each row corresponding to an output parameter of the turbomachine TM in stabilized regime, each output parameter of each row being different from the other input parameters of the other rows, each column corresponding to a recording instant of flight V, each recording instant of each column being different from the other recording instants of the other columns.

4. A method (100) according to claims 2 and 3 wherein, for the flight of interest VI, the assembly T is constructed from the pair i^VIT' ^viT

5. A method (100) according to claim 3, wherein the transient training data of set T comprises: • A matrix Xy resulting from a horizontal concatenation of each matrix Xyp, • A matrix Yp resulting from a horizontal concatenation of each matrix Yp, and wherein the transient prediction model f is defined such that f(Xp) = Yr

6. A method (100) according to claim 3 wherein, for each flight V of each turbomachine TM of the stabilized assembly S, with the exception of the flight of interest VI of the turbomachine of interest TMI: - the stabilized training data of the assembly S includes the torque (y) - the stabilized prediction model Hv is defined such that, 1 y 1= m ^vsr ^vs) ^vs

7. A method (100) according to any one of the preceding claims wherein the stabilized prediction model HVi is constructed from at least the stabilized prediction model Hcchoisi according to a condition Cl.

8. Method (100) according to the preceding claim in combination with claim 4 wherein condition Cl is: - The distance between and is minimal, where corresponds to flight C and Yyip corresponds to flight of interest VI; - The cost function of the stabilized prediction model Hcest is minimal; or - The generalization error of the stabilized prediction model Hcest is minimal.

9. Method (100) according to any one of the preceding claims comprising a margin calculation step (106), the margin being proportional to the difference between the matrix of interest YSiet a matrix Y mp, the matrix Y mp being the output of a representative worst-case physical model for the input Xsc.

10. Product (100) computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method (100) according to any one of the preceding claims.

Images