Predictive monitoring device for the operation of a unit estimating a physical quantity, system and associated processes

Abstract

Predictive monitoring device for the operation of a physical quantity estimation unit, system and associated methods The present invention relates to a predictive monitoring device (16) for the operation of a physical quantity estimation unit (14) receiving data acquired by a detection unit (12) and applying an estimation model to the acquired data, the predictive monitoring device (16) comprising: - a test data generation unit (20) applying a transformation to the acquired data, - a calculation unit (22) estimating a value of the physical quantity by applying the model to the test data, - an analysis unit (24) of the test value performing a comparison between the test value and an expected value according to a performance evaluation criterion of the model, and - a warning unit (26) issuing an alert based on the result of the comparison.Figure for the abbreviation: figure 1.

Description

Title of the invention: Device for predictive monitoring of the operation of a unit for estimating a physical quantity, system and associated methods

[0001] The present invention relates to a device for predictive monitoring of the operation of a unit for estimating a physical quantity. It also relates to an estimation system comprising such a predictive monitoring device. The present invention relates to associated monitoring and estimation methods.

[0002] As part of the development of critical systems involving restrictive security objectives and requirements, verifications of the robustness and stability of the systems developed are carried out during the design phase. These verifications are carried out in particular through test campaigns and tooled sensitivity analyses.

[0003] Artificial intelligence technologies are increasingly being used to develop data-driven machine learning models in critical systems.

[0004] These models are in particular estimation models capable of giving an estimated value of the physical quantity over a predefined area of ​​use.

[0005] The predefined domain of use is, by definition, the set of operating conditions under which the estimation model is specifically designed to operate, including, but not limited to, environmental, geographical and temporal restrictions.

[0006] An estimation model is considered robust when the estimation model can continue to operate correctly despite abnormal inputs and conditions outside the predefined domain of use

[0007] The estimation model is considered stable if it can continue to function correctly despite disturbances to its inputs within the predefined domain of use.

[0008] Robustness and stability are more difficult to verify for systems designed by data-driven machine learning. Indeed, few formal methods exist and they are often limited to certain model classes.

[0009] Furthermore, it is not possible to infinitely multiply the number of data in the training data sets for obvious reasons of availability of this data and viability of the industrial cost of the verification activities.

[0010] There is therefore a need for a predictive monitoring device for the func operation of a unit for estimating a physical quantity allowing the robustness and stability of a tool obtained by machine learning to be controlled.

[0011] To this end, the description describes a device for predictive monitoring of the operation of a unit for estimating a physical quantity,

[0012] the estimation unit receiving data acquired by a detection unit and applying an estimation model to acquired data to obtain an estimated value, the estimation model being capable of giving an estimated value of the physical quantity over a predefined area of ​​use,

[0013] the predictive monitoring device comprising:

[0014] - a test data generation unit, the generation unit receiving the data acquired by the detection unit and applying a transformation on the acquired data to obtain test data, the test data being data observable by the detection unit and on which the application of the estimation model gives an expected value,

[0015] - a calculation unit estimating a value of the physical quantity by applying to the test data the same estimation model as the estimation unit to obtain a test value,

[0016] - a unit for analyzing the test value, the analysis unit performing at least one comparison between the test value and an expected value according to at least one criterion to obtain a comparison result, each criterion being a criterion for evaluating the performance of the estimation model, and

[0017] - a warning unit issuing an alert depending on the result of the com parison.

[0018] According to particular embodiments, the predictive monitoring device has one or more of the following characteristics, taken in isolation or in all technically possible combinations:

[0019] - a transformation that the generation unit is capable of applying is a trans training introducing noise into the acquired data.

[0020] - a transformation that the generation unit is capable of applying is a function of generation of abnormal data, abnormal data being data not belonging to the predefined domain of use.

[0021] - a criterion used by the analysis unit is a model robustness criterion estimation and the expected value being an indication of the presence of abnormal data as input to the estimation model.

[0022] - a criterion used by the analysis unit is a stability criterion and the expected value is the estimated value.

[0023] - the warning unit is capable of issuing an alert when, for a number of times greater than a predefined threshold, the result of the comparison corresponds to a criterion not filled.

[0024] The description also describes a system for estimating a physical quantity, the estimation system comprising:

[0025] - an estimation unit suitable for receiving data acquired by a unit of detection and being capable of applying an estimation model to acquired data to obtain an estimated value, the estimation model being capable of giving an estimated value of the physical quantity over a predefined area of ​​use, and

[0026] - a predictive monitoring device for the estimation unit, the monitoring device surveillance being as previously described.

[0027] According to particular embodiments, the estimation system has one or more of the following characteristics, taken in isolation or in all technically possible combinations:

[0028] - the estimation system further comprises a detection unit specific to acquire acquired data.

[0029] - the estimation system further comprises a control unit capable of transmitting an output value for the estimation system depending on both the estimated value and an output from the alert unit.

[0030] The description also describes a method for predictive monitoring of the operation of a unit for estimating a physical quantity,

[0031] the estimation unit receiving data acquired by a detection unit and applying an estimation model to acquired data to obtain an estimated value, the estimation model being capable of giving an estimated value of the physical quantity over a predefined area of ​​use,

[0032] the method comprising:

[0033] - a test data generation step implemented, the generation step comprising receiving the data acquired by the detection unit and applying a transformation to the acquired data to obtain test data, the test data being data observable by the detection unit and on which the application of the estimation model gives an expected value,

[0034] - a calculation step comprising the estimation of a value of the physical quantity by applying the same estimation model to the test data as the estimation unit to obtain a test value,

[0035] - a step of analyzing the test value, the analysis step comprising at least one comparison between the test value and an expected value according to at least one criterion to obtain a comparison result, each criterion being a criterion for evaluating the performance of the estimation model, and

[0036] - a warning step comprising the emission of an alert depending on the result of the comparison.

[0037] The description also relates to a method for estimating a physical quantity comprising:

[0038] - an estimation step, the estimation step being implemented by a unit estimation and comprising the reception of data acquired by a detection unit and the application of an estimation model on acquired data to obtain an estimated value, the estimation model being capable of giving an estimated value of the physical quantity over a predefined area of ​​use, and

[0039] - a step of implementing a predictive monitoring method such as pre previously described.

[0040] In the present description, the expression “suitable for” means indifferently “adapted for”, “adapted to” or “configured for”.

[0041] Characteristics and advantages of the invention will appear on reading the description which follows, given solely by way of non-limiting example, and made with reference to the appended drawings, in which:

[0042] - [Fig.l] [Fig.l] is a schematic representation of an estimation system of a physical quantity including a predictive monitoring device, and

[0043] - [Fig.2] [Fig.2] is a schematic representation of the monitoring device predictive of [Fig.l].

[0044] [Fig.l] describes an estimation system 10 comprising a detection unit 12, an estimation unit 14, a predictive monitoring device 16 and a control unit 18.

[0045] The estimation system 10 is capable of estimating a physical quantity.

[0046] The physical quantity can be of any type, the estimation system 10 thus serving as a measurement system.

[0047] The detection unit 12 is capable of acquiring data.

[0048] In this respect, the detection unit 12 comprises one or more sensors capable of collecting data.

[0049] The data collected are thus measurements.

[0050] According to the example described, the detection unit 12 is a camera, the camera providing images.

[0051] The data acquired by the detection unit 12 are thus images.

[0052] Alternatively or additionally, the detection unit 12 may comprise any type of sensors, and in particular sensors capable of collecting signals of the audio, video, time series, tabular data or any other type of signals.

[0053] The estimation unit 14 is a unit for estimating a physical quantity.

[0054] The estimation unit 14 is suitable for receiving the data acquired by the unit of detection 12.

[0055] The estimation unit 14 is capable of applying an estimation model to data acquired to obtain an estimated value, which corresponds to the value of the physical quantity sought.

[0056] The estimation model is a model derived from artificial intelligence and in particular a model obtained by machine learning.

[0057] As a result, the estimation model has been trained on necessarily incomplete data, which justifies the presence of the monitoring device 16.

[0058] For example, the estimation model is a neural network.

[0059] The estimation model is capable of giving an estimated value of the physical quantity over a predefined area of ​​use.

[0060] The predefined domain of use is, by definition, the set of operating conditions under which the estimation model is specifically designed to operate, including, but not limited to, environmental, geographical and temporal restrictions.

[0061] The predefined domain of use is often designated by the abbreviation ODD which refers to the English term “operational design domain”.

[0062] To determine the predefined field of use of a model derived from artificial intelligence, it is appropriate to measure the performance of the model under operational conditions as close as possible to real conditions (i.e. nominal and abnormal conditions).

[0063] Using the results of the model's inference on input data containing nominal values ​​and anomalies, the predefined domain of use ODD of the model is characterized by the measured parameter ranges (e.g., image brightness) for which the model provides a correct output within an acceptable margin of error.

[0064] This predefined ODD usage domain of the model will serve as a reference for the monitoring device 16, because it corresponds to the domain in which the estimation model guarantees the expected behavior and performance level.

[0065] The monitoring device 16 is capable of monitoring the operation of the estimation unit 14.

[0066] The monitoring device 16 makes it possible to check that the estimation unit 14 is functioning correctly or more precisely will function correctly as explained in the remainder of the description.

[0067] In this context, the expression "functioning correctly" refers to correct functioning of the estimation unit in the future (in production or in operation), which gives the monitoring carried out by the monitoring device 16 a predictive aspect. In this sense, the monitoring device 16 is a predictive monitoring device.

[0068] The monitoring device 16 comprises a generation unit 20, a unit calculation 22, an analysis unit 24 and a warning unit 26.

[0069] The generation unit 20 is a test data generation unit.

[0070] The generation unit 20 being capable of receiving the data acquired by the detection unit 12.

[0071] The data received by the generation unit 20 are therefore the same data as the data received by the estimation unit 14.

[0072] The generation unit 20 and the estimation unit 14 therefore use the same input data.

[0073] The generation unit 20 is capable of applying a transformation to the acquired data to obtain test data.

[0074] In this sense, test data is acquired data that has been disturbed.

[0075] Each test data is data that can be observed by the detection unit 12 and for which the application of the estimation model will give an expected value. This “observable” character means that the data are data that can be observed under real operating conditions such as the failure of a sensor or an operating condition outside the predefined area of ​​use.

[0076] Data is data that can be observed when it is representative of an environment that may occur in a real case.

[0077] Observing snow in the middle of summer on a beach where the temperature has been 40°C for several weeks is, for example, not something that can be observed in real life.

[0078] The expected value is usually either the value of the physical quantity or an output error code, the case having never been encountered by the model.

[0079] Several examples of transformation illustrating a possible operation of the generation unit 20 are now described.

[0080] According to a first use case relating to monitoring the stability of the estimation unit 14, the generation unit 20 is capable of applying a transformation introducing noise into the acquired data. The test data are then noisy data.

[0081] By applying such a transformation, the generation unit 20 is able to ensure that the noisy data remains within the predefined domain of use.

[0082] According to a second use case relating to the monitoring of the robustness of the estimation unit 14, the generation unit 20 is capable of applying an abnormal data generation function.

[0083] In this context, abnormal data is data that does not belong to the predefined domain of use.

[0084] In such a case, this implies that the estimation model taking abnormal data as input must, in normal operation, interpret such data as an anomaly not to be treated. Typically, the estimation model will output an error code or a classification of the abnormal data as unknown data.

[0085] For example, in the case of acquired data that are images, examples of transformation are rotations and / or translations of the original image.

[0086] It is also possible to apply a transformation enabling the output of images to be obtained that are deliberately blurred, colored, obscured, too bright, too opaque, containing noise or a simulation of “dead” pixels or lines or columns of “dead” pixels.

[0087] According to another example, the acquired data are a set of measurements whose temporal variation is monitored. The acquired data are then time series.

[0088] In such an example, a transformation is, for illustration, a function modulating amplitude, frequency or phase shift.

[0089] It is also possible to consider changes in frequency or seasonality, the addition of breaks or outlier observations.

[0090] For acquired data that would be tabular data, a transformation may, for example, aim to add aberrant observations and / or spurious correlations between the initial acquired data.

[0091] The generation unit 20 is controlled by a control unit capable of inhibiting its operation.

[0092] The control unit inhibits the generation unit when the data acquired at the input of the estimation unit 14 does not belong to the predefined domain of use. In fact, by construction, the estimation unit 14 is not suitable for estimating a quantity from data outside the predefined domain of use and the evaluation of robustness or stability in this context is meaningless. The calculation unit 22 is a unit suitable for estimating a value of the physical quantity by applying the same estimation model as the estimation unit 14 to the test data. The value estimated by the calculation unit 22 is thus a test value.

[0093] The calculation unit 22 is therefore a digital twin of the estimation unit 14.

[0094] In operation, the only difference is that the calculation unit 22 operates on input data different from those of the estimation unit 14. These input data are different in the sense that the input data of the calculation unit 22 are test data generated from data actually acquired to reproduce the data which could be acquired in the future.

[0095] The calculation unit 22 is thus used to determine the behavior of the estimation model in the presence of the test data, that is to say input data modified to simulate noisy input data or abnormal input data.

[0096] The analysis unit 24 is a test value analysis unit.

[0097] According to the example described, the analysis unit 24 comprises two evaluation sub-units.

[0098] The first evaluation sub-unit 28 is a sub-unit for evaluating the robustness of the estimation model.

[0099] As previously stated, the estimation model is considered robust when the estimation model can continue to operate correctly despite abnormal inputs and conditions outside the predefined domain of use.

[0100] Robustness assessment consists of determining whether the estimation model is robust or not according to a criterion.

[0101] This criterion is, for example, the percentage of incorrect operation in the presence of a predefined number of tests of the estimation model with abnormal inputs.

[0102] The first evaluation sub-unit 28 performs a comparison between the test value (at the output of the calculation unit 22) and an expected value (at the output of the estimation unit 14) which is an indication of the presence of abnormal data at the input of this estimation model.

[0103] The second evaluation sub-unit 30 is a sub-unit for evaluating the stability of the estimation model.

[0104] According to the previous definition, it is recalled here that the estimation model is considered stable if it can continue to function correctly despite disturbances to its inputs within the predefined domain of use.

[0105] An example of a stability criterion is the concept of "bounded inputs, bounded outputs" or "invariance of the output class".

[0106] The second evaluation sub-unit 30 performs a comparison between the test value (at the output of the calculation unit 22) and an expected value which is the output value of the estimation unit 14.

[0107] The alert unit 26 is capable of issuing an alert based on the result of the comparison.

[0108] In the example described, the alert unit is capable of issuing an alert when, for a number of times greater than a predefined threshold, the result of the comparison corresponds to an unfulfilled criterion.

[0109] The control unit 18 is capable of emitting an output value for the estimation system 10 depending both on the estimated value and on the output of the alert unit 26.

[0110] For example, in the simplest implementation, the control unit 18 will output the acquired estimated value determined by the estimation unit 14 when the alert unit does not issue an alert and otherwise will transmit the alert message issued by the alert unit instead of the output of the estimation unit 14.

[0111] A specific example is now described to illustrate the operation of the dif- different subunits of the analysis unit 24.

[0112] An estimation model is considered which would be capable of classifying the input data according to three output classes, a first class Cl, a second class C2 and an anomaly class CA.

[0113] The first evaluation sub-unit 28 then evaluates the robustness of the evaluation model according to the following decision table:

[0114] [Tables 1] Estimation unit output 14 Calculation unit output 22 Comparison result Estimated status for the robustness of the estimation model Cl Cl non-compliance with the criterion yes Cl C2 non-compliance with the criterion yes Cl CA compliance with the criterion no C2 Cl non-compliance with the criterion yes C2 C2 non-compliance with the criterion yes C2 CA compliance with the criterion no CA Cl not carried out not evaluated CA C2 not carried out not evaluated CA CA not carried out not evaluated

[0115] In the preceding table, a “yes” status corresponds to an estimation unit 14 operating locally (i.e. for these particular test data) in a non-robust manner and conversely, a “no” status corresponds to an estimation unit 14 operating locally in a robust manner.

[0116] The case of already abnormal acquired data is considered to be of little relevance since the generation unit 20 does not modify the data. This does not allow robustness to be tested.

[0117] The second evaluation sub-unit 30 then evaluates the stability of the evaluation model according to the following decision table:

[0118] [T ableaux2] Output of estimation unit 14 Output of calculation unit 22 Comparison result Estimated status for the stability of the estimation model Cl Cl compliance with criterion no Cl C2 non-compliance with criterion yes Cl CA non-compliance with criterion yes C2 Cl non-compliance with criterion yes C2 C2 compliance with criterion no C2 CA non-compliance with criterion yes CA Cl not carried out not evaluated CA C2 not carried out not evaluated CA CA not carried out not evaluated

[0119] In the preceding table, a “yes” status corresponds to an estimation unit 14 operating locally (i.e. for these particular test data) in a non-stable manner and conversely, a “no” status corresponds to an estimation unit 14 operating locally in a stable manner.

[0120] Here again, the case of already abnormal acquired data is considered to be of little relevance since the generation unit 20 does not modify the data. In other words, we are not interested in the stability of the estimation model 14 outside its predefined domain of use (ODD).

[0121] Based on these elements, it is the warning unit 26 which will determine whether or not an alert should actually be issued.

[0122] In this sense, the warning unit 26 synthesizes the information coming from the two evaluation sub-units 28 and 30.

[0123] For this, the warning unit 26 can carry out the analysis over a measurement period (a measurement here is the output of the evaluation sub-units, i.e. the emission or not of an alert for stability or robustness).

[0124] The warning unit 26 will accumulate the measurements and when this number of measurements is greater than a threshold N, the warning unit 26 implements, for example, a synthesis technique as described below.

[0125] For measurements k to kN (k being an integer greater than or equal to N), the number of alarms issued is compared to a threshold, exceeding the threshold leading to the issue of an alarm for the property (stability or robustness) considered.

[0126] Alternatively, it is possible to consider the percentage of alarms issued instead of the number of alarms issued.

[0127] The monitoring device 16 which has just been described makes it possible to verify the correct functioning of the estimation unit 14 during the phase of use of the estimation unit 14 in production or in operation in the real environment.

[0128] More specifically, the monitoring device 16 makes it possible to determine whether the estimation model used by the estimation unit 14 maintains satisfactory robustness and stability throughout its life cycle in production or in operation.

[0129] The monitoring device 16 therefore performs predictive control of robustness and stability using a hybrid approach in the sense that it is based on both data and rules (an approach sometimes referred to as “data and rule driven”).

[0130] This predictive control is based on the generation unit 20 serving as an online generator of adversarial examples which will be used to produce disturbed or abnormal inputs, on the calculation unit 22 which is a digital twin of the estimation unit 14 and on decision tables which will produce a verdict on the robustness and stability of the estimation model.

[0131] The predictive nature of the control allows for preventive intervention, which is not possible with existing techniques which are reactive.

[0132] The fact that the control can be carried out in real time and under real operating conditions is also innovative insofar as it makes it possible to continue in production the verification of the robustness and stability of the estimation model in order to complete the verification effort already made in the factory during the development phase and which is in practice often incomplete.

[0133] This approach is particularly suitable for estimation models carrying out a “classification” or “regression” type operation but can be extended to other types of estimation.

[0134] The search for robustness and stability is found in many areas of industry, so that the monitoring device 16 can advantageously be used in areas such as aeronautics, space, defense, energy, land transport, health or the chemical industry.

[0135] It will be clearly understood that the monitoring device 16 is also of interest even if it only performs one of the two analyses (robustness or stability).

[0136] In this respect, in the most general case, the analysis unit 24 of the test value performs at least one comparison between the test value and an expected value according to at least one criterion to obtain a comparison result, each criterion being a criterion for evaluating the performance of the estimation model and the warning unit 26 issuing an alert according to the result of the comparison.

[0137] In the example which has just been described, the generation unit 20, the calculation unit 22, the analysis unit 24, the warning unit 26 are each produced in the form of software, or a software brick, executable by the processor.

[0138] The memory of the monitoring device 16 is then able to store generation software, calculation software, analysis software, and warning software. The processor is then able to execute each of the generation software, calculation software, analysis software, and warning software.

[0139] All these units can be developed using software technologies such as C++, Java, C, "Rust" (registered trademark), "Go" (registered trademark) or "Python" (registered trademark). The choice of programming language will be defined by the level of criticality of the use of the estimation system 10. The platform where these units will be executed will depend on the use case. It can be a simple traditional server or a station with graphics processing units (GPU) locally or in edge computing, or a distributed platform.

[0140] In a variant not shown, the generation unit 20, the calculation unit 22, the analysis unit 24 and the warning unit 26 are each produced in the form of a programmable logic component, such as an FPGA (Field Programmable Gate Array) or an integrated circuit, such as an ASIC (Application Specific Integrated Circuit).

[0141] When the monitoring device 16 is produced in the form of one or more software programs, that is to say in the form of a computer program, also called a computer program product, it is furthermore capable of being recorded on a medium, not shown, readable by a computer. The computer-readable medium is for example a medium capable of storing electronic instructions and of being coupled to a bus of a computer system. By way of example, the readable medium is an optical disk, a magneto-optical disk, a ROM memory, a RAM memory, any type of non-volatile memory (for example FLASH or NVRAM) or a magnetic card. A computer program comprising software instructions is then stored on the readable medium.

[0142] What has just been described in terms of physical implementation is also valid for the estimation unit 14 and the control unit 18.

[0143] These remarks relating to the physical implementation of the generation units 20, the calculation unit 22, the analysis unit 24 and the warning unit 26 are also valid for the sub-units that these units contain.

[0144] It should be noted here that the units may contain more subunits than the subunits shown in [Fig.l].

[0145] In particular, the generation unit 20 could comprise a sub-unit for each use case, i.e. a first sub-unit applying the transformations adapted for stability assessment (first use case) and a second subunit applying the transformations adapted for robustness assessment (second use case).

Claims

Claims

1. A device (16) for predictively monitoring the operation of a unit (14) for estimating a physical quantity, the estimation unit (14) receiving data acquired by a detection unit (12) and applying an estimation model to the acquired data to obtain an estimated value, the estimation model being capable of giving an estimated value of the physical quantity over a predefined area of ​​use, the predictive monitoring device (16) comprising: - a unit (20) for generating test data, the generation unit (20) receiving the data acquired by the detection unit (12) and applying a transformation to the acquired data to obtain test data, the test data being data observable by the detection unit (12) and on which the application of the estimation model gives an expected value,- a calculation unit (22) estimating a value of the physical quantity by applying to the test data the same estimation model as the estimation unit (14) to obtain a test value, - an analysis unit (24) of the test value, the analysis unit (24) carrying out at least one comparison between the test value and an expected value according to at least one criterion to obtain a comparison result, each criterion being a criterion for evaluating the performance of the estimation model, and - a warning unit (26) issuing an alert depending on the result of the comparison.,

2. A predictive monitoring device according to claim 1, wherein a transformation that the generation unit (20) is capable of applying is a transformation introducing noise into the acquired data.

3. A predictive monitoring device according to claim 1 or 2, wherein a transformation that the generation unit (20) is capable of applying is an abnormal data generation function, abnormal data being data not belonging to the predefined field of use.

4. A predictive monitoring device according to any one of claims 1 to 3, wherein a criterion used by the analysis unit (24) is a robustness criterion of the estimation model and the expected value being an indication of the presence of abnormal data in the input of the estimation model.

5. A predictive monitoring device according to any one of claims 1 to 4, wherein a criterion used by the analysis unit (24) is a stability criterion and the expected value is the estimated value.

6. Predictive monitoring device according to any one of claims 1 to 5, in which the warning unit (26) is capable of issuing an alert when, for a number of times greater than a predefined threshold, the result of the comparison corresponds to an unfulfilled criterion.

7. System (10) for estimating a physical quantity, the estimation system (10) comprising: - an estimation unit (14) capable of receiving data acquired by a detection unit (12) and being capable of applying an estimation model to acquired data to obtain an estimated value, the estimation model being capable of giving an estimated value of the physical quantity over a predefined area of ​​use, and - a predictive monitoring device (16) of the estimation unit (14), the monitoring device (16) being according to any one of claims 1 to 6.

8. An estimation system according to claim 7, the estimation system (10) further comprising a detection unit (12) capable of acquiring acquired data.

9. An estimation system according to claim 7 or 8, wherein the estimation system (10) further comprises a control unit (18) adapted to emit an output value for the estimation system (10) depending both on the estimated value and on an output of the alert unit (26).

10. Method for predictive monitoring of the operation of an estimation unit (14) of a physical quantity, the estimation unit (14) receiving data acquired by a detection unit (12) and applying an estimation model to the acquired data to obtain an estimated value, the estimation model being capable of giving an estimated value of the physical quantity over a predefined area of ​​use, the method comprising: - a step of generating test data implemented, the generation step comprising receiving the data acquired by the estimation unit (14) in order ... detection (12) and applying a transformation to the acquired data to obtain test data, the test data being data observable by the detection unit (12) and on which the application of the estimation model gives an expected value, - a calculation step comprising the estimation of a value of the physical quantity by applying to the test data the same estimation model as the estimation unit (14) to obtain a test value, - a step of analyzing the test value, the analysis step comprising at least one comparison between the test value and an expected value according to at least one criterion to obtain a comparison result, each criterion being a criterion for evaluating the performance of the estimation model, and - a warning step including the issue of an alert based on the result of the comparison.

11. Method for estimating a physical quantity comprising: - an estimation step, the estimation step being implemented by an estimation unit (14) and comprising the reception of data acquired by a detection unit (12) and the application of an estimation model to the acquired data to obtain an estimated value, the estimation model being capable of giving an estimated value of the physical quantity over a predefined area of ​​use, and - a step of implementing a predictive monitoring method according to claim 10.

Images