Preventive method for thermal monitoring of a fixed internal structure in an aircraft turbomachine nacelle

Abstract

The invention relates to a method for thermal monitoring of a fixed internal structure in an aircraft turbomachine nacelle, the method comprising steps of: acquiring and storing, by a processor onboard the aircraft, temperature measurement data from a plurality of temperature sensors (SNk) distributed over a surface of the fixed structure (1); defining a mesh of the surface of the fixed structure; determining by the processor a temperature (Ti,j) of each mesh (Mi,j) as a function of the temperature measurements and the respective positions of the temperature sensors; for each mesh whose temperature is above a corresponding threshold value, calculating a damage index of the mesh as a function of the difference between the mesh temperature and the corresponding threshold value, and the time during which the mesh temperature remained above the corresponding threshold value. Figure 5

Description

Title of the invention: Preventive method for thermal monitoring of a fixed internal structure in an aircraft turbomachine nacelle technical field

[0001] The present invention relates to a monitoring method for defining preventively maintenance operations carried out in the nacelle of a turbomachine equipping an aircraft. State of the art

[0002] A turbojet nacelle generally has a substantially tubular structure comprising an air inlet upstream of the turbojet, an intermediate assembly designed to surround a turbojet fan, and a rear assembly that may incorporate thrust reversing means and is designed to surround the combustion chamber and all or part of the compressor and turbine stages of the turbojet. The nacelle is generally terminated by an exhaust nozzle whose outlet is located downstream of the turbojet.

[0003] Modern nacelles are designed to house a turbofan engine capable of generating, on the one hand, a hot air flow (also called the primary flow) from the turbofan engine's combustion chamber, circulating within a space delimited by a substantially tubular compartment called the "core" compartment, and on the other hand, a cold air flow (called the "secondary flow") from the fan, circulating outside the turbofan engine through an annular passage, called the "vein," formed between an internal structure defining a turbofan fairing and an internal wall of the nacelle. Both airflows are ejected from the turbofan engine through the nozzle at the rear of the nacelle.

[0004] The core compartment comprises an external casing serving as a housing, called the internal fixed structure (IFS) of the nacelle, generally made of composite material. This internal fixed structure is subjected to significant thermal stresses. To thermally protect this internal IFS, thermal protection panels are known to be used, in particular to isolate the nacelle components from the engine environment, thereby maintaining them at acceptable temperatures and thus maximizing their lifespan. These thermal protection panels also provide fire protection. They can be used in other areas of the nacelle where there is a risk of fire.

[0005] Thermal protection panels generally include at least one insulating layer, which may be made from silica fibers, ceramics, or a microporous material. The layer may be fixed between strips, usually made of stainless steel.

[0006] During maintenance inspections on turbofan engines, numerous IFS components were found to have defects due to localized overheating of the composite structure. This overheating results from prior deterioration of the thermal protection, to the point that it can no longer protect the composite structure. Under these conditions, the composite structure can be subjected to a heat flux with an excessive temperature. This flux can have a significant impact on the mechanical strength of a component of the structure, necessitating repair or even complete replacement of the component. Late detection of such overheating therefore potentially leads to more complex and extensive replacements or repairs of the composite structure, and longer and more frequent aircraft grounding.However, at present, there is no solution for detecting heating of the IFS composite structure in flight.

[0007] It is therefore desirable to propose a system that can more directly detect overheating of a composite structure such as that of the IFS during flight, analyze it, and define a maintenance operation within a timeframe compatible with the estimated damage. By optimizing maintenance operations in this way, it is possible to reduce the number of scrapped parts and the duration and frequency of aircraft grounding. The operational reliability of the aircraft can thus be improved. Summary

[0008] Embodiments relate to a method for thermal monitoring of a fixed internal structure in an aircraft turbomachine nacelle, the method being implemented by a processor and comprising steps consisting of: acquiring and storing, by a processor on board the aircraft, temperature measurement data from a plurality of temperature sensors distributed over a surface of a fixed structure of the nacelle; defining a digital mesh of the surface of the fixed structure comprising a plurality of cells; determining by the processor a temperature of each cell as a function of the temperature measurement data and the respective positions of the temperature sensors on the surface of the fixed structure;for each mesh whose temperature is greater than a corresponding threshold value, calculate by the processor a damage index of the mesh as a function of a difference between the temperature of the mesh and the corresponding threshold value, and a time; during which the temperature of the mesh remained above the corresponding threshold value.

[0009] Thanks to these provisions, it is possible to determine in real time, with a spatial resolution dependent on the mesh used, the temperatures experienced locally by the composite structure. Using the thermal properties of the composite structure, it is possible to determine and anticipate potential damage to the composite structure in the event of local or more widespread overheating. Anticipating such damage makes it possible to determine maintenance operations to be carried out while reducing the scope and therefore the duration of these operations. This results in improved operational reliability of the aircraft.

[0010] According to one embodiment, the method includes a calculation of a remaining life of the composite structure as a function of the calculated damage indices.

[0011] Determining the remaining service life allows maintenance operations to be planned.

[0012] According to one embodiment, the method includes the application by the processor of a spatial confidence test to the temperature measurement data, the spatial confidence test comprising steps consisting of: considering a set of sensor pairs, each associating two sensors from the plurality of temperature sensors, calculating for each sensor a spatial confidence index as a function of temperature deviations calculated on the basis of the measurement data from each sensor pair to which the sensor belongs at a given time and a distance between the sensors of the sensor pair; and combining the measurement data from each sensor with the spatial confidence index calculated for the sensor to decrease the influence of the measurement data from the sensor as a function of the corresponding confidence index in the determination of the temperature of each mesh.

[0013] This provision makes it possible to eliminate false measurements and thus increase the reliability of the measurements.

[0014] According to one embodiment, the method includes the application by the processor of a temporal confidence test to the temperature measurement data, the temporal confidence test comprising steps consisting of: calculating for each sensor a temporal confidence index as a function of a temperature deviation calculated on the basis of the measurement data from the sensor taken at different times; and combining the measurement data from each sensor with the temporal confidence index calculated for the sensor in order to decrease the influence of the measurement data from the sensor as a function of the corresponding confidence index in the determination of the temperature of each mesh.

[0015] This provision also makes it possible to eliminate false measurements and thus increase the reliability of the measurements.

[0016] According to one embodiment, the method includes the calculation by the processor of a confidence coefficient for each sensor, combining the spatial confidence index and the temporal confidence index, the measurement data from each sensor being combined with the confidence coefficient calculated for the sensor to reduce the influence of the measurement data from the sensor in the determination of the temperature of each mesh.

[0017] By combining temporal and spatial comparisons, the detection of false measurements is more precise and more reliable.

[0018] According to one embodiment, the temperature of each mesh is determined as a function of curvilinear distances, following a curvature of the composite structure, between a center of the mesh and the respective positions of at least a part of the temperature sensors.

[0019] The accuracy of the temperatures of each mesh is thus improved.

[0020] According to one embodiment, the process includes a refining step of the respective mesh temperatures taking into account the evolution of mesh temperatures over a time interval, in order to reduce the impact of noise present in the measurement data from the sensors.

[0021] According to one embodiment, the damage index of each mesh is determined as a function of a sum of differences between mesh temperatures exceeding the corresponding temperature threshold value and the temperature threshold value.

[0022] The mesh damage index thus calculated is more accurately representative of the state of the mesh.

[0023] Embodiments may also relate to an on-board monitoring system for an aircraft turbomachine nacelle, comprising: a set of temperature sensors distributed over a surface of a fixed composite structure of the nacelle, and a processor connected to the set of sensors, the monitoring system being configured to implement the method as previously defined.

[0024] According to one embodiment, the monitoring system includes a transmission interface connected to the processor to communicate with an external monitoring system. Brief description of the figures

[0025] The present invention will be better understood with the aid of the following description of exemplary embodiments with reference to the accompanying figures, in which identical reference signs correspond to structurally and / or functionally identical or similar elements.

[0026] [Fig. 1] Figures IA and IB schematically represent, respectively in cross-section and in a projected view on a plane, an internal composite structural element of a turbojet nacelle associated with thermal protection, according to one embodiment,

[0027] [Fig. 2] Figure 2 schematically represents an on-board part of a monitoring system for a fixed composite structure according to one embodiment,

[0028] [Fig. 3] Figure 3 represents steps of a process executed by the monitoring system, according to one embodiment,

[0029] [Fig. 4] Figure 4 is a projected view onto a plane of the internal composite structural element, illustrating a step of the process, according to one embodiment,

[0030] [Fig. 5] Figure 5 is a projected view onto a plane of the internal composite structural element, illustrating a step of the process, according to one embodiment,

[0031] [Fig. 6] Figure 6 represents a temperature variation curve in an area of ​​the internal composite structure, illustrating a step of the process, according to one embodiment,

[0032] [Fig.7] Figure 7 shows curves of variation of the structural deflection of parts of the structural element as a function of time and temperature experienced by the part of the structural element, illustrating a step of the process, according to one embodiment,

[0033] [Fig.8] Figure 8 represents an example of a corrected service life variation curve of a part of the structural element as a function of time, illustrating a step of the process, according to one embodiment. Detailed description

[0034] Figures IA and IB represent a thermal protection layer 2 covering a fixed composite structure 1 such as that of a turbojet nacelle. The layer 2 may be formed of several juxtaposed panels. In one embodiment, the thermal protection layer 2 is associated with temperature sensors SN distributed on the external surface of the layer 2. The sensors may also be integrated within the layer 2, or distributed on the surface of the composite structure 1, covered by the layer 2, as illustrated in Figure IA. The sensors SN may include temperature sensors and optionally, pressure sensors.

[0035] Figure 2 shows an embedded part of a monitoring system adapted for monitoring a fixed composite structure. The embedded system includes a PRC processor, a TXI transmission interface connected to the PRC processor for communicating with an external monitoring system, for example, a ground-based system, when the aircraft is on the ground. The PRC processor is connected to the SN temperature sensors directly or indirectly via wired or wireless connections. The PRC processor is configured to process signals from SN sensors.

[0036] SN temperature sensors may be of the type having an RFID (Radio-Frequency Identification) communication interface communicating with an RFID reader connected to the PRC processor and one or more antennas, for example, of the UHF type. SN temperature sensors may also include Bragg gratings on optical fibers distributed over the composite structure 1, the PRC processor being connected to the optical fibers via an optoelectronic interface circuit (OPI). Each optical fiber transmits light pulses, a portion of the incident light being reflected by each Bragg grating at the Bragg wavelength, while the remainder of the incident light is transmitted through the optical fiber. When the optical fiber undergoes deformation or a temperature change, the Bragg wavelength shifts.By detecting and measuring such a shift, it is possible to obtain localized temperature measurements along the fiber. The sensors can also include thermocouples distributed over the composite structure 1.

[0037] The distribution of the SN sensors on the fixed composite structure 1 is determined based on the potential thermal effects in the event of damage to the thermal protection layer 2. In one embodiment, the temperature sensors are positioned so as to detect a heat flux at an excessive temperature on the composite structural part, such an excessive temperature being able to reveal a breach in the thermal protection. For this purpose, a measured temperature can be considered excessive when the corresponding heat flux has sufficient energy to damage the composite structural part.

[0038] According to one embodiment, the temperature sensors are preferably positioned on areas of the composite structure subjected to the greatest mechanical stresses. Indeed, these areas are predominant in the material health of the composite structure 1 as a whole.

[0039] To reduce the number of sensors to be distributed, these can be positioned on areas of the composite structure likely to be subjected to the highest temperatures. These areas can be identified beforehand, for example, using a thermal imaging camera. As a result, the distribution of the SN sensors on the structure is not necessarily uniform, as illustrated in Figure IB.

[0040] Figure 3 shows steps S1 to S8 of a process executed by the monitoring system, and in particular the PRC processor, according to one embodiment. In step S1, the processor receives temperature measurement signals Tk (k = 1, ..., m) from m temperature sensors SN1 to SNm. These signals are sampled into measurements Tkjt per unit of time t, according to a sampling frequency. In step S2, the The PRC processor constructs a temperature map at each unit of time t, considering a mesh dividing the surface of the composite structure 1 into cells Mki, Mij, ..., for example, as shown in Figure 4. The PRC processor calculates, for each unit of time t, a temperature T;j >t at the center of each cell M;j, based on temperature samples Tkjt from sensors SNk. To do this, it uses a distance D;j >k between each sensor k and the center of each cell Mi. The distances >k, which can be provided by a database, are calculated during a calculation step S10 by a DC distance calculation function, based on the position Pk of each sensor SNk and the position of the center MCij of each cell Mi. The temperature T;j >t provided for each cell can be obtained, for example, by a weighted sum of the temperatures Tk,t.In this weighted sum, each temperature Tk t is multiplied by a coefficient FD that varies according to the distance Dij k and a parameter representing the temperature propagation in the composite structure. Each of the distances D; j >k can be curvilinear distances, i.e., following the curvature of the composite structure between the sensor SNk and the center of the mesh M^. In calculating the temperature of each mesh Mij, only a subset of the sensors SNk can be considered, for example, only those sensors located at a distance Dijjk from the mesh, less than a threshold distance value.

[0041] Figure 5 illustrates the thermal map TM(t) obtained at the end of step S2. The TM(t) map extends over the entire surface of the monitored composite structure 1 and shows the temperature Tjj >t calculated for each mesh Mij for the temperature samples Tkjt taken at time t.

[0042] The thermal map TM(t) can be refined by taking into account its evolution during the time interval from t - e to t between time t and the time corresponding to the previous temperature sample, in order, for example, to reduce the impact of noise in the measurements. For this purpose, the PRC processor can use the following equation: 100431 TjtFEjTjt-eMe))* 1 )

[0044] in which e defines the duration of the time interval considered and v is a function determining a weighting according to the duration of the interval e.

[0045] In step S3, the PRC processor compares the temperature >t of each unit cell M^ at each instant t to a maximum threshold temperature value TTH, defined for the unit cell Mij. The threshold values ​​TTHij can be read from the database DB. Below the threshold temperature TTHij, the unit cell is considered not to be undergoing any degradation related to excessive temperature. Above this threshold temperature, the unit cell is considered to be in a state of overheating. If none of the values If the TTH threshold is not exceeded, the PRC processor can evaluate a SOH "health" state of the composite structure 1 at step S4. If one or more of the TTHi threshold values ​​are reached or exceeded (at least one Mij mesh is considered to be in an overheating state), the PRC processor executes step S5.

[0046] In step S5, the PRC processor calculates a damage index that depends on the amplitude of the overheating and the time during which each cell remained in a state of overheating. The calculation of the damage index for a cell is illustrated by Figure 6, which shows a curve Cl of the temperature variation of a cell as a function of time t. In the example in Figure 6, the temperature Tjj of cell Mjj is approximately 55°C at time t0, reaches the threshold value TTHij of 120°C at time t1, and continues to increase until it reaches 150°C at the current time t2. The damage index Ejj for the mesh Mjj can be estimated as a function of the area (t2 -tl)-(T;j(t2) - Tij(tl)) of the rectangle RL This area is also equal to (t2 - tl)-(T;j(t2) -TTHij(tl)).

[0047] According to another embodiment, the damage index Ejj for the unit cell Mjj can be estimated using a DF function applied to the area OTSij between the temperature variation curve Cl and the line corresponding to the threshold value TTHij. The area OTSij can be estimated for the unit cell by the following sum:

[0048] 0TSi .(tj-TTHij) <2)

[0049] In the next step S6, the PRC processor evaluates a structural degradation SAB(t) of the composite structure (1) at time t by applying a FA function to the indices Eii(t) calculated for each of the cells of the composite structure. The structural degradation SAB(t) makes it possible to quantify the extent and impact of overheating damage on the properties of the composite structure 1 relative to a nominal state. The structural degradation can be expressed as a percentage equal to (100% - residual mechanical strength), the residual mechanical strength representing the degradation of the mechanical properties of a part relative to an initial state, for example, at the end of the production line. The calculation of the structural degradation SAB(t) can use models of the behavior of the material forming the composite structure to identify and estimate the impact of overheating on its properties.These models, which are stored in the database DB, may include charts obtained by subjecting the materials of the composite structure and the entire composite structure to mechanical strength tests (for example, compression and shear strength tests).

[0050] To illustrate the influence of temperature on structural deflection, Figure 7 shows curves of the variation of structural deflection of parts of the composite structure as a function of time and the temperature experienced by each part considered of the composite structure. The structural degradation values ​​shown in Figure 7 can be stored as tables in the database DB, providing the structural degradation as a function of the damage index Ejj of each considered part of the composite structure. Figure 7 shows that the structural degradation degrades more or less rapidly with the temperature experienced, depending on the considered part of the composite structure.

[0051] In the next step S7, the PRC processor evaluates the health status SOH(t) of the material constituting the composite structure 1 at the current time t. To this end, the PRC processor determines a structural margin MS. In the case of a composite material, the structural margin MS can be obtained by the ratio between an allowable deformation value of the material and a corresponding deformation value actually observed in the composite structure under a given loading (stress + thermal) situation, the deformation value of the composite structure being multiplied by the structural deflection SAB(t). These values ​​are defined during the design of the composite structure and provided by the database DB.

[0052] The allowable deformation of a material is defined using charts provided by the database DB. In the case of a composite material, the allowable deformation depends on several parameters, including the number of carbon plies, the orientation of the carbon fibers, and the characteristics of the honeycomb structure. This allowable deformation can be expressed in pStrain. In the case of a composite structure equipping an aircraft, this structure is designed to have a structural margin greater than 1 for a given load. This ensures that the part will have a structural capacity compatible with the aircraft's service life. Thus, the calculated structural margin allows for a precise evaluation of the composite structure's ability to remain operational. If this margin is insufficient to ensure the structural integrity of the composite structure, the PRC processor calculates a remaining service life (RLT) for the composite structure in step S8.

[0053] The remaining service life (RLT) determines a maximum period before the next maintenance operation on the composite structure. The data estimated in steps S7 and S8, including the remaining service life (RLT) and maintenance recommendations, can be stored in the database (DB), for example, at the end of each flight mission. The remaining service life (RLT) varies according to the structural margin (MS). For example, the remaining service life (RLT) can be related to the structural margin (MS) by a linear function.

[0054] Figure 8 illustrates an example of the variation of the remaining lifetime (RLT) as a function of time. In the example of [Fig. 8], the composite structure underwent two temperature failures, namely an initial overheating of 200°C for 200 hours after approximately 15,000 hours of service, followed by a second overheating of 140°C for 3 600 hours after approximately 30,000 hours. The first overheating event reduced the remaining lifespan of the composite structure by approximately 60,000 hours out of its initial 120,000 hours. The second overheating event further reduced the remaining lifespan of the composite structure by approximately 25,000 hours, bringing it to almost zero.

[0055] At the end of a mission (a flight for an aircraft), the database DB can be updated with the structural margin MS and the remaining life RLT which have just been calculated and maintenance recommendations determined in particular according to the remaining life and a schedule of planned maintenance operations.

[0056] According to one embodiment, the PRC processor implements a learning loop configured to update the thermal reference data in the DB database, based on thermal readings from the sensors, and thus closely reflect the operational conditions of the composite structure equipping aircraft. The DB database can also be used to enrich a general database that aggregates the databases of an aircraft fleet, and the learning loop can be applied to the general database.

[0057] Furthermore, the assessment of damage during maintenance operations can be used to adjust the TTHLJ threshold values. This provision makes it possible to adapt the parameterization of the composite structure monitoring to the actual operating conditions and thus improve this monitoring.

[0058] According to one embodiment, spatial confidence tests (TCI) and temporal confidence tests (TC2) are applied to temperature measurement signal samples Tk(t) to determine a confidence coefficient (CCk(t)) for each sample Tk(t). The TCI confidence test comprises steps SI1, SI2. In step SI1, the PRC processor calculates, for each existing sensor pair (SNk, SNki), confidence indices (CSkjk i) as a function of the temperature measurement samples Tk(t) and Tk i(t) recorded at time t and the distances (DkjH) between the sensors SNk and SNkp of the sensor pair. Thus, each confidence index (CSk,ki) can be calculated by applying a function (FD1) to the temperature difference (Tk - Tki) and the distance (Dk>ki) between the sensors SNk and SNkp. The function (FD1) can be a polynomial comparison function, for example, in 1 / x.

[0059] Here too, each of the distances Dk >ki can be a curvilinear distance, that is to say, following the curvature of the composite structure between the sensors SNk and SNk i j.

[0060] Thus, when the temperature difference is small with a nearby sensor, the temperature measurement has a relatively high confidence index. When When the temperature difference is significant with a distant sensor, the temperature measurement has a slightly reduced confidence level. When the temperature difference is significant with a nearby sensor, the temperature measurement has a severely reduced confidence level.

[0061] In step S12, the PRC processor calculates, for each temperature value Tk(t) at time t, a spatial confidence coefficient Clk(t) by combining, using a function FC1, all the confidence indices CSkjk i obtained in step S1 for the value Tk(t). The function FC1 calculates, for example, the product of all the error coefficients CSkjk i determined for the measurement sample Tk(t), kl varying from 1 to the number m of sensors SNk, according to the following equation: 100621 ClXcsJ 3 '

[0063] The function FD1 and more generally the calculation of the confidence coefficient Clk(t) can be the subject of a learning loop so that the behavior of the confidence coefficient Clk(t) adapts to the real conditions encountered. According to one embodiment, the sensors SNk[ considered in equation (3) are restricted to those whose distance Dkjk i is less than a potentially variable threshold value.

[0064] The TC2 confidence test includes a step S13, during which the PRC processor calculates, for each value Tk(t), a temporal confidence coefficient C2k(t) as a function of the current temperature measurement sample Tk(t), and previous measurement samples Tk(t-1), ..., Tk(te), recorded at times t-1, ..., te by the SNk sensor. Thus, each confidence coefficient C2k can be calculated by applying a comparison function FT to the samples Tk(t), ..., Tk(te).

[0065] In step S14, the PRC processor combines the spatial confidence coefficient Clk(t) and the temporal confidence coefficient C2k(t) for each temperature sample value Tk(t) at time t using a combination function FC, to produce a confidence coefficient CCk(t) for each sample value Tk(t). The confidence coefficients CCk(t) thus obtained can be used in the calculations performed in step S2 by being combined (for example, multiplied) respectively by the sample values ​​Tk(t), in order to reduce the influence of each sample according to the corresponding confidence coefficient. In this case, the confidence coefficient CCk(t) can take values ​​between 1 if the corresponding sample value Tk(t) is completely reliable, and 0 if this value is determined to be completely erroneous.

[0066] It will be clear to those skilled in the art that the present invention is susceptible to various embodiments and applications. In particular, not all the steps for calculating the remaining service life of the composite structure are necessarily carried out by the on-board processor PRC, some of the final steps of the process can be executed by a ground computer from the thermal map of the composite structure, formed from the respective temperatures of the meshes.

Claims

Demands

1. 1. A method for thermal monitoring of a fixed internal structure in an aircraft turbomachine nacelle, the method being implemented by a processor (PRC) and comprising steps of: acquiring and storing, by a processor (PRC) onboard in the aircraft, temperature measurement data from a plurality of temperature sensors (SN) distributed over a surface of a fixed structure (1) of the nacelle; defining a digital mesh of the surface of the fixed structure comprising a plurality of cells (Mij); determining by the processor a temperature (Tjj) of each cell as a function of the temperature measurement data and the respective positions of the temperature sensors on the surface of the fixed structure;for each mesh whose temperature is above a corresponding threshold value (TTH,,), calculate by the processor a damage index (¾) of the mesh as a function of a difference between the temperature of the mesh and the corresponding threshold value, and of a time during which the temperature of the mesh remained above the corresponding threshold value.;

2. 2. Method according to claim 1, comprising a calculation of a remaining life (RLT) of the composite structure as a function of the calculated damage indices (¾).

3. 3. A method according to claim 1 or 2, comprising the application by the processor (PRC) of a spatial confidence test (TCI) to the temperature measurement data, the spatial confidence test comprising the steps of: considering a set of sensor pairs, each pair associating two sensors from the plurality of temperature sensors (SN), calculating for each sensor a spatial confidence index (Cl) as a function of temperature deviations calculated on the basis of the measurement data from each sensor pair to which the sensor belongs at a given time and a distance between the sensors in the sensor pair; and combining the measurement data from each sensor with the spatial confidence index (Cl) calculated for the sensor to decrease the influence of measurement data from the sensor as a function of the corresponding confidence index in determining the temperature (Tij) of each mesh (M^).

4. 4. A method according to any one of claims 1 to 3, comprising the application by the processor (PRC) of a temporal confidence test (TC2) to the temperature measurement data, the temporal confidence test comprising steps of: calculating for each sensor (SN) a temporal confidence index (C2) as a function of a temperature deviation calculated on the basis of the measurement data from the sensor taken at different times; and combining the measurement data from each sensor with the temporal confidence index (C2) calculated for the sensor to decrease the influence of the measurement data from the sensor as a function of the corresponding confidence index in the determination of the temperature (T^) of each mesh (Mij).

5. 5. Method according to claims 3 and 4, comprising the calculation by the processor of a confidence coefficient (CC) for each sensor (SN), combining the spatial confidence index (Cl) and the temporal confidence index (C2), the measurement data from each sensor being combined with the confidence coefficient (CC) calculated for the sensor to decrease the influence of the measurement data from the sensor in the determination of the temperature (T^) of each mesh (Mij).

6. 6. A method according to any one of claims 1 to 5, wherein the temperature (T^) of each mesh (M^) is determined as a function of curvilinear distances, following a curvature of the composite structure, between a center of the mesh and the respective positions of at least a portion of the temperature sensors (SN).

7. 7. A method according to any one of claims 1 to 6, comprising a step of refining the respective temperatures (T;j) of the meshes (M^) taking into account the evolution of the temperatures of the meshes during a time interval, in order to reduce a noise impact present in the measurement data from the sensors (SN).

8. 8. A method according to any one of claims 1 to 7, wherein the damage index (¾) of each mesh (Mij) is determined as a function of a sum of differences between temperatures (T^) of the mesh exceeding the corresponding temperature threshold value (TTH,,) and the temperature threshold value.

9. 9. On-board monitoring system for an aircraft turbomachine nacelle, comprising: a set of temperature sensors (SN) distributed over a surface of a fixed composite structure (1) of the nacelle, and a processor (PRC) connected to the set of sensors, the monitoring system being configured to implement the method according to any one of claims 1 to 8.

10. 10. Monitoring system according to claim 9, comprising a transmission interface (TXI) connected to the processor (PRC) for communicating with an external monitoring system.

Images