Method for detecting operating anomalies of industrial equipment from anomaly scores and corresponding installation

Abstract

Method for detecting operating anomalies of industrial equipment from anomaly scores and corresponding installation Method for controlling industrial equipment to detect operating anomalies of the industrial equipment, comprising the following steps: a) obtaining an initial model adapted to calculate anomaly scores from a plurality of operating parameters, the initial model resulting from a machine learning process, b) obtaining a plurality of values ​​defining at least one operating point of the industrial equipment, c) calculating the anomaly score associated with the operating point, d) displaying an image showing anomaly scores calculated by the initial model, as a function of at least two operating parameters, e) determining, by a user, at least one desired correction in the displayed image, f) from the initial model and the desired correction,obtaining a corrected model, g) displaying a corrected image showing the corrected anomaly scores as a function of the two coordinates, steps b) to g) being optionally iterated, h) obtaining a plurality of values ​​of the operating parameters defining an operating point, and i) calculating the anomaly score associated with said operating point using the corrected model obtained.,

Description

Description Title of the invention: Method for detecting functional anomalies- operation of industrial equipment from scores anomaly and corresponding installation

[0001] — The present invention relates to a method for controlling industrial equipment to detect operating anomalies of industrial equipment, including obtaining a suitable model for calculating anomaly scores from a plurality of operating parameters representative of operating states of industrial equipment, the anomaly scores being respectively representative of probabilities that the operating states are abnormal, the initial model resulting from a machine learning process.

[0002] The invention also relates to a corresponding installation, comprising industrial equipment, sensors adapted to provide the plurality of parameters of operation, and a control system adapted to implement some of the steps of such a process.

[0003] …— Industrial equipment is for example a production line, a com- pressure of a fluid, or any equipment for industrial purposes and which we wish to monitor health, that is, know if it is functioning normally or if its func- operation is abnormal and requires, for example, a shutdown, maintenance, or any act of administration of industrial equipment. Operating parameters are for example pressures, temperatures, vibration levels, etc.

[0004] — In this field, two main approaches are known.

[0005] — According to a first main approach, operating parameters are identified by experts and monitored to ensure that these parameters evolve within ranges considered normal. The advantage of this first strategy is that it is based on a physical knowledge of industrial equipment and more ge- generally on physical laws and standards. For example, vibrational levels maximum acceptable values ​​are defined for a rotating machine, according to a standard, in depending on the power of the machine.

[0006] — However, a first drawback of this strategy is that the context of use usual rotating machine is not integrated, while the machine operates ge- generally within a reduced range of the normal operating range. Thus, This strategy generally does not allow early detection of a change behavior linked to a degradation of industrial equipment, as long as the pa- meters remain within the normal range.

[0007] — Furthermore, monitoring of the joint evolution of several parameters of operation can also be considered, but its implementation sometimes proves to be complex, even prohibitive, particularly in the case of a large number of operating parameters to be monitored. This leads machine operators to monitor only a small number of operating parameters, and separately, which makes monitoring very partial. According to a second main approach, the normal behavior(s) of the machine are learned from one or more operating parameters using learning algorithms. This creates a model providing, for each possible operation of the machine, an anomaly score representative of the probability that the operation is abnormal. The anomaly score is for example a probability between 0 and 1, the value "1" meaning that the operation is abnormal with maximum probability, the value "0" meaning that the operation is normal with maximum probability. The advantage of this second approach is to know, after a learning phase on a data history, the normal operations defined by several operating parameters, whether by their individual values, but also in relation to each other. This allows for very early detection of behavioral anomalies. However, a major drawback of this second approach is that obtaining a high-performance model, combining both good sensitivity and good specificity, requires a large data history, representative of all normal behaviors of the machine. In particular, since the relationships between sensors are monitored, the size of this history is closely linked to the number of monitored sensors. Thus, it is unimaginable to have this representative history and obtain an optimal model for calculating anomaly scores. In practice, industrial equipment monitoring is carried out while only an imperfect model is available. In order to improve the model, one or more re-parameterizations of the model are carried out over time, in order to take into account the good and bad anomaly detections made. By re-parameterization, we mean here a modification of the parameters defining the model. However, these reconfigurations are complex and tedious, due to their frequency and the amount of information to be exchanged. As a result, they are not always well accepted by a user of the control system, typically a technician, who must qualify all the anomalies, and for whom the reconfigurations can cause a significant work overload. Another disadvantage of this second approach is that it does not incorporate the physical or technical knowledge that one may have of industrial equipment, which may eventually lead to false alarms in trivial cases of operation. operation. An aim of the invention is to remedy all or part of the above drawbacks, by providing a control method for detecting operating anomalies of industrial equipment, which is more reliable and / or easier to implement. To this end, the invention relates to a method for controlling industrial equipment to detect operating anomalies of the industrial equipment, comprising the following steps: a) obtaining an initial model suitable for calculating anomaly scores from a plurality of operating parameters representative of operating states of the industrial equipment, the anomaly scores being respectively representative of probabilities that the operating states are abnormal, the initial model resulting from a machine learning process, b) obtaining a plurality of values ​​of the operating parameters defining at least one operating point of the industrial equipment, c) calculation, using the initial model, of the anomaly score associated with the operating point, d) displaying at least one image showing anomaly scores calculated by the initial model, as a function of at least two of the operating parameters forming two coordinates of the image, the image showing the anomaly score of the operating point, €) determination, by a user, of at least one desired correction in the displayed image, the desired correction relating to at least one correction point, or at least one correction zone, f) from the initial model and the desired correction, obtaining a corrected model adapted to calculate, from the plurality of operating parameters, corrected anomaly scores, g) display of a corrected image showing the corrected anomaly scores as a function of the two coordinates, steps b) to g) being optionally iterated over time, the corrected model of one iteration becoming the initial model of a following iteration, (h) obtaining a plurality of values ​​of the operating parameters defining an operating point of the industrial equipment, and 1) calculation of the anomaly score associated with said operating point using the corrected model obtained during the last of the iterations. According to particular embodiments, the method comprises one or more of the following characteristics, taken in isolation or in all technically possible combinations: - step a) includes a machine learning sub-step to obtain the initial model; - in step h), and optionally in step b), measurements are carried out by a plurality of sensors of the industrial equipment to obtain the values ​​of the operating parameters: - the method comprises a step j) of development by the user or a machine, from said anomaly score obtained in step i), of an administration decision for the industrial equipment intended to modify the operating state of the industrial equipment; - in step e), the desired correction relates to an area of ​​the displayed image; and in step f), the calculation of the corrected model comprises a summation of the initial model and a penalty function having as variables at least said coordinates of the image, the penalty function being adapted to carry out in the corrected image a correction of predefined shape, for example spherical, elliptical, rectangular, or by band; - in step e), the desired correction relates to at least one correction point belonging to the displayed image; and in step f), the calculation of the corrected model comprises a summation of the initial model and a penalty function having as variables at least said coordinates of the image, the penalty function comprising a term associated with the correction point, the term being, in absolute value, maximum for said correction point and decreasing as a function of a distance from said correction point; - said term is proportional to a Gaussian function having said distance as a variable; - said term comprises a preselected parameter to define a propagation of the desired correction in the corrected image, the term being of the form a K(* (y) ) es SR II where: X is a current point having as coordinates the operating parameters, and where the penalty function is calculated, Ÿ is the correction point, a is a coefficient, K is a decreasing function, with K(0) = 1, d(X,Y) is the distance between the correction point Y and the current point X, and y is the preselected parameter; and - the method comprises a step f1) of determining the preselected parameter from a data set comprising a plurality of operating points having respectively as coordinates values ​​of the operating parameters, the determination comprising the following sub-steps: - for increasing values ​​of a parameter of rank k, and for each of said points operating point, calculating a distance between said operating point and its kth nearest neighbor in the plurality of operating points, - for each of the values ​​of the parameter of rank k, calculation of an average distance equal to an average of the distances obtained for the operating points, - calculation of a first derivative of the average distance with respect to the parameter of rank k and, if the first derivative has at least one first peak, choice of a value of the parameter of rank k located just before the first peak, the preselected parameter then being defined as the average distance associated with said value of the parameter of rank k, - if the first derivative does not have a first peak, calculation of a second derivative of the average distance with respect to the parameter of rank k and choice of a value of the parameter of rank k for which the second derivative becomes stable, the pre-selected parameter then being defined as the average distance associated with said value of the parameter of rank k. The invention also relates to an installation comprising: - industrial equipment, - sensors adapted to provide a plurality of operating parameters representative of operating states of the industrial equipment, and - a control system adapted to detect operating anomalies of the industrial equipment, the control system comprising at least a central unit, a memory, a screen and a keyboard, the memory comprising instructions which when read by the central unit cause the installation to carry out steps a) to d) and f) to i) of a method as described above, the screen being adapted to display the image showing anomaly scores or the corrected image. The invention will be better understood from reading the following description, given solely by way of example and with reference to the appended drawings, in which: [Fig.1] [Fig.1] is a schematic view of an installation according to the invention, [Fig.2] [Fig.2] is a schematic view illustrating the steps of a method according to the invention, some of which are implemented by the installation shown in [Fig.1], [Fig.3] [Fig.3] is a graph giving anomaly scores, in value levels, of historical operating points of industrial equipment of the installation represented in [Fig.1], as a function of two of the operating parameters, [Fig.4] [Fig.4] is an image analogous to the graph shown in [Fig.3], showing anomaly scores, in value levels, provided by the initial model of the process shown in [Fig.2], the anomaly scores having been obtained by machine learning using historical operating points, [Fig.5] [Fig.5] is a graph similar to that shown in [Fig.3], showing corrections desired by a user on correction points taken from the historical operating points shown in [Fig.3], [Fig.6] [Fig.6] is a graph analogous to that shown in [Fig.4], showing a penalty function obtained from the desired corrections shown in [Fig.5], the preselected parameter for defining the spread of the correction being 0.25, [Fig.7] [Fig.7] is a graph analogous to that shown in [Fig.4], showing corrected anomaly scores provided by a corrected model of the process shown in [Fig.2], the corrected anomaly scores being obtained from the anomaly scores shown in [Fig.4] and the penalty function shown in [Fig.6], [Fig.8] [Fig.8] is a graph analogous to that shown in [Fig.3], giving corrected anomaly scores for the historical operating points shown in [Fig.3], [Fig.9] [Fig.9] is a graph similar to that shown in [Fig.6], showing a correction desired by the user on elliptical and rectangular areas of the graph shown in [Fig.4], [Fig.10] [Fig.10] is a graph analogous to that shown in [Fig.7], showing corrected anomaly scores provided by the corrected model, the corrected anomaly scores being obtained from the anomaly scores shown in [Fig.4] and the correction shown in [Fig.9], [Fig.11] [Fig.11] is a variation of the graph shown in [Fig.9], showing a user-desired correction across a strip of the image shown in [Fig.4], [Fig.12] [Fig.12] is a graph analogous to that shown in [Fig.6], showing a penalty function obtained from a combination of the penalty functions shown in Figures 6 and 9, [Fig.13] [Fig.13] is a variant of the graph shown in [Fig.6], showing a penalty function obtained from the desired corrections shown in [Fig.5], the preselected parameter for defining the spread of the correction being 0.05 instead of 0.25, [Fig.14] [Fig.14] is a graph analogous to that shown in [Fig.7], showing corrected anomaly scores obtained from the anomaly scores shown in [Fig.4] and the penalty function shown in [Fig.13], [Fig.15] [Fig.15] is a variant of the graph shown in [Fig.6], showing a penalty function obtained from the desired corrections shown in [Fig.5], with the preselected parameter for defining the spread of the correction being 1.0 instead of 0.25, [Fig.16] [Fig.16] is a graph analogous to that shown in [Fig.7], showing corrected anomaly scores obtained from the anomaly scores shown in [Fig.4] and the penalty function shown in [Fig.15], [Fig.17] [Fig.17] is a variant of the graph shown in [Fig.6], showing a penalty function obtained from the desired corrections shown in [Fig.5], with the preselected parameter for defining the spread of the correction being 5.0 instead of 0.25, [Fig.18] [Fig.18] is a graph analogous to that shown in [Fig.7], showing corrected anomaly scores obtained from the anomaly scores shown in [Fig.4] and the penalty function shown in [Fig.17], [Fig.19] [Fig.19] is a three-part graph showing, as a function of the rank parameter k, at the top, the average distance obtained for the historical operating points represented in [Fig.3], in the middle, the first derivative of the average distance with respect to the rank parameter k, and at the bottom, the second derivative of the average distance with respect to the rank parameter k, [Fig.20] [Fig.20] is a variant of the graph shown in [Fig.3], showing historical operating points forming not several distinct groups, but only one, and [Fig.21] [Fig.21] is a graph similar to that shown in [Fig.19], but obtained for the historical operating points shown in [Fig.20], and not those shown in [Fig.3]. Facility With reference to [Fig. 1], an installation 10 according to the invention is described. Installation 10 is, for example, located on a site producing energy, raw materials, goods, or intermediate products intended for industry. The installation 10 comprises industrial equipment 12 and sensors 14 adapted to provide a plurality of operating parameters C1, C2 … CN, representative, according to their possible values, of operating states of the industrial equipment. The installation 10 also comprises a control system 16 adapted to detect operating anomalies of the industrial equipment 12. The industrial equipment 12 is for example a compressor, a turbine, a boiler, or any industrial equipment known per se. There are N sensors 14, with N equal to one, two, three or more. According to a particular embodiment, the or at least one of the sensors 14 can provide several operating parameters. For example, a multi-axis accelerometer provides accelerations along three axes, i.e. three operating parameters. Operating parameters are for example temperatures, pressures, vibration levels, or other quantities typical of industrial equipment. The operating parameters are, for example, instantaneous values, or averaged over time. The operating parameters are advantageously standardized, i.e., for example, divided by a reference value. For example, a pressure P is replaced by a pressure P*=P / P0, where: P* is the normalized pressure, P is the pressure in bar, varying for example between 1 and 2 bar, PO is a reference pressure, for example 1.5 bar. More generally, P* = aP+b, where a and b are parameters, which are advantageously calculated automatically. The control system 16 is located close to the equipment 12, or alternatively remotely. The control system 16 is adapted to be used by at least one user 18 (hereinafter called “the user”, but it may be a team, or successive users). The control system 16 is advantageously adapted to act directly (arrow 19) on the industrial equipment 12, for example to stop it, start it, or modify an operating instruction. The control system 16 is advantageously adapted to receive and store historical operating data, or to receive values ​​of the operating parameters C1, C2 … CN. The control system 16 comprises at least a central unit 20, a memory 22, a screen 24 and a keyboard 26. The user 18 is typically a technician responsible for controlling the industrial equipment 12, for example on site, or alternatively remotely. The memory 22 comprises instructions which, when read by the central unit 20, cause the installation 10 to carry out steps a) to d) and f) to i) of a method according to the invention which will be described below with reference to [Fig.2]. Step e) of this method is carried out by the user 18, who enters one or more desired corrections 28 into the control system 16. The screen 24 is for example adapted to display images such as represented in FIGS. 3 to 18, showing anomaly scores obtained from the operating parameters C1, C2, etc. CN, or showing desired corrections, as a function of two of the operating parameters, in the example the parameters C1 and C2. Advantageously, the user 18 can select the pair of parameters serving as coordinates for the images that he wishes to see displayed. According to a variant not shown, the images are produced as a function of three of the operating parameters, but such three-dimensional images are more difficult for the user 18 to read and interpret. Process With reference to [Fig. 2], we will now describe a method according to the invention, which is implemented in the example by the installation 10 and the user 18 (or users). The method is suitable for controlling the industrial equipment 12, in particular by detecting anomalies in its operation in the form of anomaly scores displayed on the screen 24. The method comprises a step a) of obtaining an initial model F suitable for calculating anomaly scores from the operating parameters C1, C2 … CN, a step b) of obtaining a plurality of values ​​30 of the operating parameters defining at least one operating point M ([Fig. 3]) of the industrial equipment 12, and a step c) of calculating, using the initial model, the anomaly score F(M) associated with the operating point M ([Fig. 3]). The method comprises a step d) of displaying at least one image 32 ([Fig.4]) showing anomaly scores calculated by the initial model F, as a function of at least two of the operating parameters, here the parameters C1 and C2, forming two coordinates of the image 32, the image 32 showing the anomaly score of the operating point M and possibly being accompanied by an alert message. The method comprises a step e) of determining, by the user 18, at least the desired correction 28 (figures 5, 9, 11 and 12) to be made to the displayed image 32, the desired correction relating to at least one correction point ([Fig.5], showing several point corrections), or at least one correction zone (figures 9 and 11 showing correction zones). The correction can also combine point and zone corrections ([Fig.12]). The method comprises a step f) of obtaining, from the initial model F and the desired correction 28, a corrected model Fp adapted to calculate, from the plurality of operating parameters C1, C2 … CN, corrected anomaly scores. The method comprises a step g) of displaying a corrected image 36 (figures 7 and 10) showing the corrected anomaly scores, as a function of the two parameters C1 and C2. Steps b) to g) are optionally iterated over time, the corrected model Fp of one iteration becoming the initial model F of a following iteration, so as to make successive corrections which accumulate. The method comprises a step h) of obtaining a plurality of values ​​38 of the operating parameters 14 defining an operating point of the industrial equipment 12, for example the point M, and a step i) of calculating the anomaly score Fp(M) associated with said operating point using the corrected model, or that obtained during the last of the iterations. An alert message is possibly issued by the control system 16 depending on the value of Fp(M). Advantageously, the method comprises a step j) of development by the user 18 or a machine, for example the control system 16, from the anomaly score obtained in step h), of an administration decision 40 of the industrial equipment 12, intended to modify the operating state of the industrial equipment 12, for example to stop it or to change an operating instruction. Step a) corresponds to an initial start-up of the method. Steps b) to g), possibly iterated, correspond to a period, more or less long, of taking into account one or more feedbacks from the user 18 on anomaly scores, in particular one-off, and / or of taking into account his knowledge of the industrial equipment 12, in particular by one or more feedbacks on areas of the displayed images. Steps h) to j) are rather exploitation steps once the model F has been made reliable by the feedback(s) from the user 18. It is understood that steps b) to g) allow for the reliability of anomaly scores in a flexible manner, without requiring a tedious re-parameterization of the F model. Step a) —- Initial model Anomaly scores are representative of probabilities that operating states (defined by operating parameters C1, C2 … CN) are abnormal. For example, anomaly scores have a value in the interval [0,1], with the value 1 representing definitely abnormal operation, and the value 0 meaning definitely normal operation. At each point X defined by the operating parameters, the initial model F obtained in step a) (or the corrected model obtained during the previous iteration) provides an anomaly score F(X). In [Fig.4], image 32 represents the anomaly scores F(X) as a function of the coordinates C1 and C2 of each point. In the example, parameter C1 varies between -10 and +10, and parameter C2 varies between -6 and +8. In a manner known per se, these are advantageously standardized values, in particular so as not to favor one operating parameter over another in the distance calculations which will be explained below. Anomaly scores are represented by contour lines (as for altitude on geographic maps). In the example, the levels 0.0, 0.2. 1.0 were chosen in [Fig.4]. In practice, of course, we will use a color gradient, which is easier to read. The initial model F results from a machine learning process. The initial model F was for example obtained previously and is found in the memory 22 of control system 16. According to a particular embodiment, step a) comprises a sub-step k) of automatic learning to obtain the initial model F. For example, historical data 42 represented in [Fig. 3] are used, comprising a set of operating points. Using automatic learning known per se (one can cite for example the one-class SVM, PCC, KNN methods, etc.), areas encompassing these points are determined. For each point X, a probability of anomaly is determined, generally as a function of the distance of point X from these areas. The learning process results in the initial model F, of which the image 32 in [Fig. 4] provides a partial representation. This is a functional, probabilistic translation of the discrete historical data 42 represented in [Fig. 3]. Step b) and h) - Measurements In step h), and optionally in step b), measurements are advantageously carried out by the sensors 14 of the industrial equipment 12 to obtain the values ​​of the operating parameters C1 to CN. Alternatively, the values ​​of the operating parameters Cl to CN come for example from memory 22 and correspond for example to test cases, or to old measured values. Steps c) and d) —- Displaying anomaly scores In step c), an anomaly score F(M) associated with operating point M is calculated using operating parameters C1 to CN. In [Fig.4], F(M) has a value slightly greater than 0.4. For example, if the anomaly score F(M) is greater than a predefined threshold, an audible and / or visual alert is issued. In step d), image 32 visible in [Fig.4] is displayed. Image 32 shows the anomaly scores F(X) for any point X located in a plane, point X being defined by its coordinates C1 and C2 in this plane. E f) — j He wished: the use According to a first embodiment, in step e), the correction 28 desired by the user 18 takes the form of one or more point corrections. These corrections relate for example to the operating point M, or to other correction points Y. For example, [Fig.3] shows, at the bottom left of the graph, a number of unlabeled points, to which are associated, for example, anomaly scores of 0.5, meaning that the initial model F cannot determine whether these operating points are normal or abnormal. The user decides, for example, that these points correspond to abnormal operating modes, and labels these points as abnormal, as visible on the [Fig.5], showing a positive correction for these points, which will increase their anomaly scores. In step f), the calculation of the corrected model Fp comprises a summation of the initial model F and a penalty function Pen having as variables at least the coordinates C1, C2 of the image, advantageously all the operating parameters CIa to CN. For each point X, the corrected anomaly score is given by the formula: Fp(X)=F(X)+Pen(X)where: F is the initial model, Pen is the desired correction, Fp is the corrected model. If Fp(X) is greater than 1.0 or less than 0.0, a normalization is advantageously carried out, so that the corrected anomaly score Fp(X) remains in the interval [0,11]. If the user decides to make a single point correction at a point Y, then, for any point X, the penalty function Pen is advantageously, in absolute value, maximum for said correction point Y, and decreasing as a function of a distance d(X,Y) between the correction point Y and the point X. For example: A(XF) \ where: Pen(X) =ak($55) £E) ) where: à is a coefficient such that the penalty is advantageously worth a correction b desired by the user at point Y, the correction b being between -1 and +1 in the example, with a=b in the case of a single point correction. K is a decreasing function, with for example K(0) = 1, d(X,Y) is for example the Euclidean distance between points X and Y, y is a preselected parameter whose meaning will be explained below. Alternatively, other distances than the Euclidean distance could be used. The correction is maximal in absolute value at point Y, and its absolute value |Per(F)} is |a|. The further away from point Y, the more the value |Per(X)| decreases. In [Fig.5], the user has in fact decided to make several point corrections at a plurality of points Y. The corrections are positive for a cloud of points located at the bottom left, and negative for another cloud of points located on the right. For any point X, the penalty function Pen is then given by the formula: d(XY;) ù: Pen(X)= Ya k(2572)1° i is an integer used to index the points Y;, a; is a coefficient advantageously calculated so that the penalty at point Y; is worth a correction b; for example between -1 and +1 requested by the user, that is to say that, for any point labeled Y; : Pen(Y;) = b, [Fig.6] shows the values ​​of Pen(X) in the example, on a scale from -1.0 to +100. The process allows any operating point to be labeled. For example, the user labels a point as abnormal by choosing a correction b of +0.5 or as normal by choosing a correction b of -0.5. Advantageously, if the user is not sure of the correction he wants, he can choose corrections less than 0.5 in absolute value, so that the correction has only a relative effect on the corrected anomaly score. Conversely, if the correction point has a low initial anomaly score (therefore considered rather normal), while the user considers that this operating point is abnormal, the user can choose a correction greater than +0.5, or even a correction of +1.0. Normalization advantageously brings the Fp(X) value back into the interval [0,1] in all cases. [Fig.7] shows the values ​​of Fp(X), i.e. the corrected image 36. Advantageously, each term of the penalty function Pen associated with a correction point Y is proportional to a Gaussian function having the distance d(X,Y) as a variable. In this case, the function K takes for example the following form: K(x)=exp(-In (2) xord) where: X is a variable, exp is the exponential function, In is the natural logarithm, ord is a parameter greater than 1. This leads, for each term associated with a correction point Ÿ, to a penalty worth: port . 4 ad{xyy"* 'aexp| -ln(2) sat) Alternatively, other K functions are possible. The inventors observed that a Gaussian function is particularly advantageous. As we have seen, in the case of a single correction point Ÿ, Pen(ŸY) = a. Furthermore, if d(X,Y) = y, Pen (X) = a / 2 in the example. The preselected parameter y therefore represents, in the example, the distance for which the correction at point X is only half of what it is at correction point Y in the case of a single correction point Y. Alternatively, other definitions of the preselected parameter y are possible by modifying the function K. For example, the preselected parameter y can represent the distance for which the correction at point X is no more than a given fraction of what it is at correction point Y. The method therefore makes it possible, by preselecting the value of the parameter y, to control the "propagation" or "diffusion" of a point correction from the correction point Ÿ to points X located in the vicinity of the correction point Y. The lower the value of the parameter y, the more the correction applied decreases rapidly as one moves away from the correction point Y. In Figures 6 and 7, the value of the parameter y is, for example, 0.25. We will return to the effect of the parameter y and describe a method for choosing the value of this parameter. [Fig.8] shows the corrected anomaly scores Fp(Y;) of the Y; points represented in [Fig.3]. Steps e) and f) — correction on an area desired by the user According to a second embodiment (alternatively or in addition to the first embodiment), in step e), the correction 28 desired by the user 18 relates to a zone Z of the displayed image. This advantageously makes it possible to take into account knowledge of the operation of the industrial equipment 12, or the experience of the user, in particular the thresholds not to be crossed by certain of the operating parameters, or the zones of good operation. For example, in [Fig.9], the user wants to make a negative correction on an elliptical shaped zone Z1 (normal operation inside the ellipse) and a positive correction on a rectangular shaped zone Z2 (abnormal operation outside the rectangle). In step f), the calculation of the corrected model Fp comprises a summation of the initial model F and a penalty function Pen having as variables at least said coordinates of the image, advantageously all the operating parameters C1 to CN. The penalty function Pen is adapted to carry out in the corrected image 36 a correction of elliptical shape Fp in the zone Z1 and rectangular in the zone Z2. The corrected model is shown in [Fig.10] (which is therefore the sum of the images shown in figures 4 and 9). Alternatively or in addition, the correction has a band shape in a Z3 zone, as shown in [Fig.11]. This type of zone correction can be obtained using functions, or a sum of functions known per se to those skilled in the art. For example, an elliptic correction comes from the choice of the Euclidean distance in the Pen function defined previously. For a rectangular correction, we choose for example the distance induced by the lp norm: , with the parameter p sufficiently high. For a vertical band correction, the distance is for example calculated with only C1 as the operating parameter, unlike the previous cases. For example, the correction can be "abrupt" (with a drop-off when leaving the ellipse, rectangle or band), or "gentle" (with a transition zone whose extension can be advantageously configured). This can be adjusted via the ord parameter: the higher it is, the more abrupt the correction. In [Fig. 12], a penalty function resulting from the combination of the penalty functions shown in [Fig. 6] (from point corrections) and in [Fig. 9] (corrections by zones) is shown. The method according to the invention makes it possible to introduce several corrections, for example of different nature. Iteration of steps b) to g) As we have seen, in steps e) and f), the user can introduce one or more corrections, one-off or by zone. Steps b) to g) are possibly iterated over time, so that the user can, at each iteration, reintroduce one or more corrections which are cumulative with the previous ones. The new corrections may come from a new operating point or a new zone whose anomaly score the user wants to correct, or may come from a change of opinion of the user, taking into account his acquired experience, on a previously corrected operating point or zone. The corrected model Fp obtained at step f) of an iteration becomes the initial model F of the following iteration. The process therefore advantageously allows several simultaneous or successive corrections to be made over time, and also allows previous corrections to be returned to and modified. In step 1), the last corrected model Fp is used. Effect of parameter y - choice of this parameter As seen above, the preselected parameter y represents, in the case of a point correction desired by the user, a distance for which the correction at point X is only a fraction, for example half, of what it is at correction point Y. [Fig.6] represents the penalty function Pen(X) obtained from the point corrections 28 shown in [Fig.5], using y=0.25. [Fig.7] shows the corrected model Fp(X), which is the sum of the initial model F(X) shown in [Fig.4] and the penalty function Pen(X) shown in [Fig.6]. Figures 13 and 14 show respectively the penalty function Pen(X) and corrected model Fp(X) obtained with y=0.05. We note that the corrections do not do not diffuse in the corrected image ([Fig.14]) and have little effect. In particular, the corrections do not propagate to the operating points close to those that the user has chosen to label in [Fig.5]. Figures 15 and 16 show respectively the penalty function Pen(X) and corrected model Fp(X) obtained with y=1.0. Figures 17 and 18 show respectively the penalty function Pen(X) and corrected model Fp(X) obtained with y=5.0. In these two cases, the corrections are spread too widely and affect not only the point cloud located in the vicinity of the points labeled by the user, but other point clouds. In the case y=5.0, the information is spread so much in the corrected image that the corrected model Fp(X) is no longer discriminating (no longer allows to distinguish abnormal points from normal points), especially in the left part of image 36 shown in [Fig.18]. The parameter y is for example stored in memory 22 and has been advantageously chosen so that the corrections diffuse sufficiently, but not too much. According to a particular embodiment, the method comprises a step fl) of determining the preselected parameter y. Final stage of reading We start, for example, from the plurality of operating points Ÿ; represented in [Fig.3], having respectively as coordinates the values ​​of the operating parameters. For increasing values ​​of a parameter of rank k, and for each of said operating points Y,, a distance D,(Y;) is calculated between the operating point Y; and its k*th nearest neighbor in the plurality of operating points. For each of the values ​​of the parameter of rank k, an average distance D, equal to the arithmetic mean of the distances D,(Ÿ;) obtained for the operating points Y; is calculated: v . where D,=+L, D(F,y N is the number of operating points Y,, and k is an integer that can take values ​​from 1 to N-1. D, is represented as a function of k on the graph at the top of [Fig.19]. Then we calculate a first derivative D," of Dy as a function of k. For example: D; =D;.1-Di- D;' is shown on the graph in the middle of [Fig.19]. In Figure 19, a second derivative D, = Di, ex” D; is also represented on the graph located at the bottom. If the first derivative Dy' has at least one first peak, we choose a value kO of the parameter of rank k located just before this peak, and the preselected parameter is there then defined as the average distance Dy, associated with this value kO of the rank parameter. We have: y = Dys. KO being the parameter of rank k located just before the first peak of D,'. In the example, we see in [Fig.19], in the graph vertically in the middle, that the function D,' presents several successive peaks P1, P2, P3… To be just before the first peak PI, we must choose for the rank parameter the value kO = 15. We obtain y = D,s = 0.25. It was observed that the parameter y thus obtained advantageously allows the point corrections to be diffused in the cloud of points located in the vicinity of the correction point, without going beyond. In the example of [Fig.19], the second derivative D; is not used and is advantageously not calculated, because the first derivative D,' has at least a first peak PI, due to the fact that the plurality of operating points Y; comprises several clouds of points distinct from each other ([Fig.3]). If, instead of starting from the plurality of operating points Ÿ; represented in [Fig.3], we start from the plurality of points Y; represented in [Fig.20], we obtain the graphs D,, D,' and D,' represented in [Fig.21]. We then note that curve D" does not present a first peak that can be isolated. This is due to the fact that the points Y; on [Fig.20] form a single cloud and not several. In this case, when the first derivative D,” does not have a peak, we calculate the second derivative D,” and we choose a value k1 of the parameter of rank k for which the second derivative D,” becomes stable, and the preselected parameter y is then defined as the average distance Dy, associated with this value k1 of the parameter of rank k. We have: y = Du, k1 being the parameter of rank k for which the second derivative D,' becomes stable. In the example of [Fig.21], we see that the function D;"" stabilizes quite quickly after a value k1 = 6. We obtain y = D,5 = 0.50. It has been observed that the parameter y thus obtained advantageously makes it possible to diffuse the point corrections towards significant points in the vicinity of the correction point, without “contaminating” the set of points Y. Thanks to the characteristics described above, the method is more reliable, because the anomaly scores are progressively made more reliable by the point or zone corrections requested by the user. The method is also easy to implement, because it allows one or more simultaneous corrections of the anomaly scores to be made, and possibly successive corrections to be made which are cumulative, without carrying out a complete re-parameterization of the initial model. The control of the industrial equipment 12 is so improved. According to the particular embodiments described above, the method makes it possible to label an actual operating point, or simply a displayed point, as “normal” or “abnormal”. The correction can be punctual (with diffusion) or immediately concern a given area of ​​the displayed image. The correction is positive (towards “abnormality” in the example) or negative (towards “normality”) and its value (between -1 and +1 in the example) can be modulated according to the user’s degree of certainty. For area corrections, by means of a suitable shape, such as elliptical, rectangular or strip, the user can easily translate his experience or knowledge of the operation of industrial equipment 12 into the form of correction. For a one-time correction (from a correction point), the value of the parameter y makes it possible to control the diffusion of the correction from the correction point to its neighbors. The parameter y can be preselected, or advantageously determined by a method based on the plurality of historical operating points available.

Claims

Claims

1. A method of monitoring industrial equipment (12) to detect malfunctions of industrial equipment (12), putting implement a control system (16) adapted to be used by at least at least one user (18), the method comprising the following steps: a) obtaining, in a memory (22) of the control system (16), a initial model (F) adapted to calculate anomaly scores from of a plurality of operating parameters (C1, ...CN) represent- indicators of operating states (X) of industrial equipment (12), the anomaly scores being respectively representative of probabilities that the operating states (X) are abnormal, the initial model (F) resulting from a machine learning process, b) obtaining a plurality of values ​​(30) of the func- operation (C1, ...CN) defining at least one operating point- operation (M) of industrial equipment (12), c) calculation, using the initial model (F), of the anomaly score associated with the operating point (M), d) display, on a screen (24) of the control system (16), of at least an image (32) showing anomaly scores calculated by the model initial, depending on at least two (C1, C2) of the func- tion (C1….CN) forming two coordinates of the image (32), image (32) showing the operating point anomaly score (M), e) determination, by the user (18), of at least one correction desired (28) in the image (32) displayed on the screen (24), the correction desired (28) relating to at least one correction point (Y), or minus one correction zone (Z), and user input (18) of the desired correction (28) in the control system (16), f) from the initial model (F) and the desired correction (28), obtaining a corrected model (Fp) suitable for calculating, from the plurality of operating parameters (C1...CN), scores of anomalies corrected, g) display, on the screen (24), of a corrected image (36) showing the anomaly scores corrected according to the two coordinates, steps b) to g) being optionally iterated over time, the model corrected (Fp) of an iteration becoming the initial model (F) of an iteration next, h) obtaining a plurality of values ​​(38) of the func- operation (C1….CN) defining an operating point (M) of industrial equipment (12), and 1) calculation of the anomaly score (Fp({M)) associated with said operating point- tion (M) using the corrected model (Fp) obtained during the last of the iterations.

2. A method according to claim !, wherein step a) comprises a machine learning sub-step to obtain the initial model (F).

3. A method according to claim | or 2, wherein, in step h), and op- tionally in step b), measurements are carried out by a plurality of sensors (14) of industrial equipment (12) to obtain the values (30, 38) operating parameters (C1...CN).

4. A method according to any one of claims 1 to 3, comprising in in addition to a step j) of development by the user (18) or a machine, from said anomaly score (Fp(M)) obtained in step i), of a decision administration (40) of industrial equipment (12) intended to modify the operating condition of industrial equipment (12).

5. A method according to any one of claims 1 to 4, wherein: - in step e), the desired correction (28) relates to an area (21, 22) of the image (32) displayed, and - in step f), the calculation of the corrected model (Fp) includes a summation of the initial model (F) and a penalty function (Pen) having for variables at least said coordinates of the image (32), the function of penalty (Pen) being adapted to achieve in the corrected image (36) a predefined shape correction, e.g. spherical, elliptical, rec- tangular, or by band.

6. A method according to any one of claims 1 to 4, wherein: - in step e), the desired correction (28) relates to at least one point of correction (Ÿ) belonging to the displayed image (32), and - in step f), the calculation of the corrected model (Fp) includes a summation of the initial model (F) and a penalty function (Pen) having for variables at least said coordinates of the image (32), the function of penalty (Pen) including a term associated with the correction point (Y), the term being, in absolute value, maximum for said correction point (Y) and decreasing as a function of a distance (d(X,Y)) from said point of correction (Y).

7. The method of claim 6, wherein said term is pro- portional to a Gaussian function having said distance (d(X,Y)) as a variable.

8. A method according to claim 6 or 7, wherein said term comprises a preselected parameter (y) to define a propagation of the desired correction (28) in the corrected image (36), the term being form aK(5E where: X is a current point having as coordinates the parameters of operation (C1..CN), and where the penalty function is calculated, Y is the correction point, a is a coefficient, K is a decreasing function, with K(0) = 1, d(X,Y) is the distance between the correction point ŸY and the current point X, and y is the preselected parameter.

9. Method according to claim 8, comprising a step f1) of deter- mination of the preselected parameter (y) from a dataset (40) comprising a plurality of operating points (Y;) having respectively for coordinates of the values ​​of the func- parameters operation (C1...CN), the determination comprising the sub-steps following: - for increasing values ​​of a parameter of rank k, and for each of said operating points (ŸY;), calculation of a distance (D,(Y;)) between said operating point (M) and its nearest neighbor in the plurality of operating points (Y1), - for each of the values ​​of the parameter of rank k, calculation of a distance average (Dy) equal to an average of the distances (D,(Ÿ;)) obtained for the operating points (Y;), - calculation of a first derivative (D,') of the average distance with respect to to the parameter of rank k and, if the first derivative (D,') has at least a first peak (P1), choice of a value (k0) of the parameter of rank k located just before the first peak (P1), the preselected parameter (y) being then defined as the average distance (D,) associated with said value (k0) of the parameter of rank k, - if the first derivative (D,') does not have a first peak (P1), calculation of a second derivative (D,'') of the average distance (D,) with respect to the parameter of rank k and choice of a value (k1) of the parameter of rank k for which the second derivative (D,'”) becomes stable, the parameter pre- selected (y) being then defined as the average distance (D,) associated with said value (k1) of the parameter of rank k.

10. Installation (10) comprising: - industrial equipment (12), - sensors (14) adapted to provide a plurality of parameters of operation (C1..CN) representative of operating states (X) industrial equipment (12), and - a control system (16) adapted to detect anomalies of operation of industrial equipment (12), the control system (16) comprising at least one central unit (20), a memory (22), a screen (24) and a keyboard (26), the memory (22) comprising ins- instructions which when read by the central unit (20) lead the installation (10) for carrying out steps a) to d) and f) to i) of a method according to any one of claims 1 to 9, the screen (24) being adapted for display image (32) showing anomaly scores or image corrected (36).