Determination of the risk of electrical arcing in an electrical circuit of an aeronautical system

Abstract

A method for determining the risk of arcing in an electrical circuit of an aeronautical system, comprising iterative steps (Pon) of: extracting (S3) a sample of electrical current corresponding to a first time window, said electrical current being continuously measured in said electrical circuit; evaluating (S4) at least one correlation measurement between said sample and a set of predetermined signals, each signal being representative of an arc; and predicting (S6), based on said at least one correlation measurement, the risk of arcing at a time after said first time window. Figure for the abstract: Fig. 3

Description

Title of the invention: Determination of the risk of electrical arcing in an electrical circuit of an aeronautical system. FIELD OF THE INVENTION

[0001] The present invention relates to determining the risk of an electric arc occurring in an electrical circuit. It is particularly applicable to the field of aeronautics and systems embedded in an aircraft.

[0002] Climate change is a major concern for many legislative and regulatory bodies worldwide. Indeed, various restrictions on carbon emissions have been, are being, or will be adopted by various states. In particular, an ambitious standard applies to both new types of aircraft and those already in service, requiring the implementation of technological solutions to bring them into compliance with current regulations. Civil aviation has been actively working for several years now to contribute to the fight against climate change.

[0003] Technological research efforts have already led to very significant improvements in the environmental performance of aircraft. The Applicant takes into account the factors impacting all phases of design and development in order to obtain aeronautical components and products that are less energy-intensive, more environmentally friendly, and whose integration and use in civil aviation have moderate environmental consequences, with the aim of improving the energy efficiency of aircraft.

[0004] Consequently, the Applicant is constantly working to reduce its negative climate impact by using methods and operating virtuous development and manufacturing processes that minimize greenhouse gas emissions to the minimum possible in order to reduce the environmental footprint of its activity.

[0005] It is within such a framework that the problem of determining the risk of the occurrence of an electric arc in an electrical circuit on board an aircraft arises, which can in particular cause premature degradation of the aircraft and impair its energy efficiency.

[0006] An electric arc is a self-sustaining discharge with a high current.

[0007] Two types of electrical arcs that may occur in an aeronautical environment can be distinguished: breaking arcs, present in contactors when a circuit is opened or closed, and fault arcs that may appear unexpectedly on all components of the electrical chain. These can cause severe damage to equipment, systems, and even the aircraft structure. This is therefore a crucial point to be aware of.

[0008] Furthermore, fault arcs can be of two types depending on their position on the electrical circuit: parallel arcs and series arcs. A parallel arc is located in parallel with the load (which causes an increase in current because a preferential path with low impedance is created) whereas a series arc is located in series with the load (which, on the contrary, causes a decrease in current because the impedance of the circuit is increased).

[0009] On an electrical distribution network, series arcs are always more complex to detect than parallel arcs.

[0010] Indeed, from an electrical signature perspective, parallel arcs generate a high inrush current, which allows overcurrent protection systems to detect them effectively and protect the line. In contrast, in the case of a series arc, the current is only slightly affected. This is because an electrical arc is a low-impedance fault. Consequently, when a series arc is present in the electrical system, the current will only experience a slight fluctuation, more precisely a small percentage reduction from its nominal value. This is the main reason why they are more complex to identify using currently available methods.

[0011] Furthermore, the waveform of the electrical signal in the network is also a factor that can complicate detection. Regarding alternating voltage networks, the 0V crossing is the point in the half-cycle where the electric arc can no longer sustain itself. When a series arc appears on this type of network, this point can result in an open circuit, leaving the line current at 0A for a short time (called a "current shoulder"), thus promoting the self-extinction of the arc at each half-cycle. Conversely, when the waveform is direct current (DC network), this phenomenon no longer occurs, and the arc can no longer take advantage of this shoulder to extinguish itself naturally.

[0012] In aircraft embedded systems, these problems are currently partly resolved, or mitigated, by several factors: - the distributed voltage levels are generally at a maximum of 230 V (at 400 Hz in fixed frequency or between 400 and 800 Hz in variable frequency), which reduces the impact of electric arcs; - the distribution of an alternating waveform (AC) also allows to promote the self-extinction of the art at each half period, when the voltage passes through 0 V. - voltage levels are generally lower, on the order of 28 VDC. - Passive protections can be used to limit the consequences of possible fault arcs: choice of materials resistant to these phenomena, use of segregation rules for optimal placement of these different elements, etc. Finally, active protection measures can also be implemented, such as arc detection and line opening mechanisms that activate when an arc is detected. Current active protection systems can detect most parallel arcs, but not series arcs, for which no active protection is currently deployed.

[0013] However, recent work on aircraft electrification proposes a distribution of direct current (DC) waveforms for high voltage levels, up to the kilovolt range. These are known as HVDC (High Voltage Direct Current) networks. This paradigm shift obviously calls into question the passive protection strategy mentioned earlier.

[0014] In addition, the series electric arc must also be taken into account, just like the parallel arc, because the damage associated with the generation of higher power faults subjected to a continuous voltage may damage the aircraft and its on-board systems, and could even impact the safety of passengers and crew.

[0015] There are some proposals for detecting an electric arc on a power line in order to isolate it (by opening the circuit). However, these proposals must comply with certain constraints: - Reliability: systematic detection in all circumstances of dangerous electrical phenomena, more often called "feared phenomena". In other words, it is about having a high rate of "true positives", close to 100%. - Robustness: the active protection mechanism must be immune to any event other than the one for which it was designed. In other words, the "false positive" rate must be as low as possible, i.e., close to 0%.

[0016] So far, these proposals show shortcomings either in terms of reliability or in terms of robustness.

[0017] In an aerial vehicle (airplane, helicopter, drone...), electrical distribution is carried out by cables and interconnection bars (or "busbars" according to the usual English terminology).

[0018] The connection between distinct means, or vectors, of distribution is made via connectors or bomies.

[0019] An example of a boomer is illustrated by figures IA and IB.

[0020] Fig. 1A represents a "top view" of a terminal block 10 in which a cable 20 is connected by means of lugs 11. The cable is held in contact with the lug by tightening a nut 12 on a stud attached to a base of the terminal block, or to a screw rod.

[0021] Figure 1B shows a cross-sectional view of a connection B1 of the same terminal block 10, in which are also shown an insulator 14 enclosing the connections in the form, for example, of a housing, and a support 15 which can be rigidly attached to the aerial vehicle. Figure 1B shows the stud 13 onto which the nut 12 is tightened to hold the cable 20 in place.

[0022] Vibrations generated by the aircraft in operation (in flight or taxiing) are transmitted to the terminal block 10 via the support 15. Due to these vibrations, it is possible that a connector or terminal block may loosen.

[0023] If, at any time, even a very short time, the physical connection between two distribution means (cables) is no longer ensured by the initial tightening, a series arc may occur, between the lug 11 and the nut 12.

[0024] The length and / or repetition of the series arcs may cause damage to the junction, or even to other elements of the aerial vehicle.

[0025] These connections via boomers are a major source of series arc generation within aerial vehicles.

[0026] It is therefore important to avoid series arcs, or at least substantially reduce their occurrence, within electrical circuits and in particular within terminal blocks and other types of connections present in the electrical distribution system. Summary of the invention

[0027] The invention aims to improve upon the state of the art. In particular, it makes it possible to avoid or reduce the occurrence of series arcs in the electrical circuits onboard an aircraft, thereby improving both the safety of the aircraft and its crew, and the lifespan of the various onboard circuits and devices. The invention thus has a beneficial effect of reducing the environmental impact of aircraft.

[0028] According to a first aspect, the present invention can be implemented by a method for determining the risk of an electric arc occurring in an electrical circuit of an aeronautical system, comprising iterative steps of: - extraction of a sample of an electric current corresponding to a first time window, said electric current being measured continuously in said electrical circuit; - evaluation of at least one correlation measure between said sample and a set of predetermined signals, each signal being representative of an electric arc; - prediction, based on said at least one correlation measure, of a risk of occurrence of an electric arc at a time after said first time window.

[0029] According to preferred embodiments, the invention comprises one or more of the following features which can be used separately or in partial combination with each other or in total combination with each other: - said at least one correlation measure includes a temporal correlation measure between values ​​of said sample and values ​​of said predetermined signal set. - said at least one correlation measure includes a frequency correlation measure between values ​​of the frequency transforms, respectively, of said sample and of said set of predetermined signals. - said at least one correlation measure is determined by a Pearson correlation measure and / or by a Spearman correlation measure. - the prediction of a risk of an electric arc occurring generates a break in said electrical circuit. - said set of predetermined signals is obtained beforehand by data acquisition steps including an identification of a voltage measurement corresponding to said electrical circuit exceeding a threshold value representative of the presence of an electric arc; and a memorization of a signal corresponding to intensity measurements in a second time window corresponding to said identification. - said data acquisition steps are carried out on a remote platform, and all the signals memorized by said data acquisition steps are stored within an aerial vehicle for the implementation of a method for determining the risk of occurrence of an electric arc as previously described.

[0030] Another aspect of the invention relates to a computer program comprising instructions for implementing a process as previously described when executed on an information processing platform.

[0031] Another aspect of the invention relates to a device for determining the risk of an electric arc occurring in an electrical circuit of an aeronautical system, adapted for - extract a sample of an electric current corresponding to a first time window, said electric current being measured continuously in said electrical circuit; - to evaluate at least one correlation measure between said sample and a set of predetermined signals, each signal being representative of an electric arc; and - predict, based on said at least one correlation measure, a risk of the occurrence of an electric arc at a time after said first time window.

[0032] Another aspect of the invention relates to an aerial vehicle comprising at least one device as previously defined.

[0033] The invention thus makes it possible to improve the safety of on-board electrical circuits, in particular by avoiding the occurrence of faulty electrical arcs, or at least by substantially reducing their occurrence.

[0034] In particular, it makes it possible to predict the occurrence of electric arcs (that is, to determine the risk of such an occurrence) in order to anticipate, if necessary, an action intended to prevent their actual occurrence. In this way, the various inconveniences linked to the occurrence and recurrence of electric arcs can be eliminated or considerably reduced.

[0035] Other features and advantages of the invention will become apparent from the following description of a preferred embodiment of the invention, given by way of example and with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE FIGURES

[0036] The attached drawings illustrate the invention: Figures IA and IB schematically illustrate a terminal block allowing the connection of electrical distribution vectors within an aerial vehicle.

[0037] Figures 2A, 2B, 2C, 2D and 2E illustrate an example of an electrical circuit and current and voltage measurements on this circuit.

[0038] Fig. 3 represents an illustrative flowchart of a process according to a possible implementation.

[0039] Figure 4 illustrates an example of application of the process, according to one embodiment.

[0040] Figure 5 illustrates an example of a device for determining the risk of an electric arc in an electrical circuit, according to one embodiment.

[0041] DETAILED DESCRIPTION OF EMBODIMENT METHODS OF THE INVENTION

[0042] Studies undertaken by the inventors have revealed the appearance of pre-disruptive phenomena, or "precursors", prior to the appearance of a Series arcing, at least in the case of series arcs caused by a loose connector. These phenomena take the form of short sparks, much less powerful than electric arcs, so they are not perceptible to the naked eye or with conventional instruments.

[0043] This phenomenon was discovered using a high-speed camera. This camera can be designed to capture 3,500 to 500,000 frames per second, depending on the resolution, to film a time window around an electrical arc caused by a loose connector. The resulting video makes it possible to identify artifacts of these pre-disruptive phenomena.

[0044] One initial idea is to use this phenomenon to predict the occurrence of an electric arc. A statistical study has finally demonstrated that the appearance of these pre-disruptive phenomena leads to the subsequent occurrence of an electric arc, with a high percentage, on the order of 90% based on the tests carried out. Furthermore, even without the subsequent occurrence of an electric arc, the detection of these pre-disruptive phenomena is valuable because they can eventually damage connectors.

[0045] Fig. 2A illustrates an electrical circuit 50 that can be installed on a platform located away from an aerial vehicle (i.e., outside of it), but reproducing a set of main characteristics of the electrical circuit that one seeks to protect within the aerial vehicle.

[0046] This electrical circuit 50 comprises a current source, typically a direct current source, 40, a terminal block 10' similar to the terminal block 10 present in the aerial vehicle, and a load 30 similar to the load connected to the terminal block 10 of the aerial vehicle. This circuit must be sufficiently similar so that measurements taken on it are representative of the behavior of the electrical circuit on board the aerial vehicle.

[0047] At least one connection of the 10' bogie is improperly screwed or improperly adjusted, so as to cause electric arcs.

[0048] Fig. 2C represents the behavior of the voltage U across the terminal block 10'.

[0049] The terminal block having a very low impedance, the voltage is nominally zero, However, significant voltage spikes (20-30 V) appear during the electric arcs. Figure 2C shows the appearance of a succession of electric arcs over time T.

[0050] Fig. 2E illustrates the behavior of the voltage U on a smaller scale, in the Z zone.

[0051] It is noted that the voltage is constant before the appearance of the voltage peak corresponding to the electric arc, at t=0.028 seconds.

[0052] It therefore appears that the measurement of the voltage does not allow the pre-disruptive phenomena to appear.

[0053] The inventors then became interested in the behavior of the intensity I of the electric current flowing in the electrical circuit, over time t.

[0054] [Fig.2D] illustrates a smaller scale view of the area Z shown in [Fig.2B], these two figures representing the behavior of the current intensity I over time.

[0055] We can then note that the intensity I has a strong variability between 39 and 39.5 A prior to the current drop corresponding to the electric arc at t=0.028 s.

[0056] In other words, it is possible to analyze the intensity I of the electric current in order to detect pre-disruptive phenomena that foreshadow the subsequent appearance of an electric arc.

[0057] Furthermore, a more detailed analysis reveals a correlation between the intensity signal I during these pre-disruptive discharges and during an electric arc, both in temporal behavior (rapid drop followed by a rapid rise) and frequency behavior.

[0058] If a correlation measure is determined from the behavior of the intensity during a pre-disruptive phenomenon, this will always be higher with the behavior during an electric arc than with the nominal behavior (i.e. without the presence of an electric arc or pre-disruptive phenomena).

[0059] These different observations and analysis results can be used to define a method for determining the risk of the occurrence of an electric arc in an electrical circuit of an aeronautical system.

[0060] Determining the risk of occurrence consists, in other words, in predicting or forecasting, with a high level of probability, the occurrence of an arc before it actually occurs, so that mechanisms can be implemented to prevent it from starting.

[0061] Fig. 3 represents an illustrative flowchart of a process according to one possible implementation of this process.

[0062] According to one embodiment, this process consists of two phases. - an "offline", or ground-based, Poff, data acquisition phase, - an "online" phase, or operational phase of the aerial vehicle, Pon, of determination of the risk of an electric arc occurring in an electrical circuit of this aerial vehicle, based on data acquired during the offline phase.

[0063] According to other embodiments, the data can be acquired by other means. In particular, they can be stored in a reference database which can be used without the need for a second data acquisition phase for the aircraft operator. This database can be shared by different operators.

[0064] The offline phase Poff can be implemented on a platform designed to generate current behavior similar to that of the electrical circuit considered in the aircraft. It is therefore suitable so that the acquired data can be used for the online phase Pon. In particular, the platform has a load and terminal block similar to those of the onboard electrical circuit.

[0065] According to one embodiment, this platform makes it possible to constitute a set of signals, or patterns, representative of pre-disruptive phenomena.

[0066] To achieve this, a terminal block with loosely tightened terminals can be installed on a vibrating structure adapted to simulate the vibrations of an aircraft. When the system is powered, the vibrations cause the generation of electric arcs.

[0067] Preferably, the simulation conditions on the platform should reproduce as closely as possible the conditions of the aircraft in flight: in particular aeronautical booms, cables, pressure variation, etc.

[0068] This offline phase Poff may include, in particular, an identification step, SI, of a voltage measurement corresponding to the electrical circuit, exceeding a threshold value representative of the presence of an electric arc.

[0069] Indeed, as previously mentioned, electric arcs are associated with voltage peaks. These are easily detectable by continuously measuring the voltage level and comparing it to a threshold value.

[0070] Since the voltage across an electrical connection (with zero or very low impedance) is zero, this threshold value can be a few volts. However, to avoid false alarms due to accidental voltage fluctuations, this threshold should not be too close to zero. An optimal value can be determined experimentally, depending on the electrical circuit under consideration, and in particular on the DC current source used. For example, a value close to half the peak voltage can be used (here, 15 V).

[0071] The identification of an electric arc can trigger an S2 step of memorizing a signal corresponding to intensity measurements in a second time window.

[0072] This second time window should preferably contain the current drop. It can be on the order of 1 ms. Since we are looking at a mathematical correlation with the current drop, the latter must be entirely represented within the window. Its exact position is not critical.

[0073] The storage may involve sampling a continuous measurement of the intensity to provide a set of discrete values ​​which can then be stored. The signal sampling frequency can be 500 kHz (current sensor with 1 MHz bandwidth).

[0074] Since pre-disruptive phenomena are difficult to detect directly, the offline phase includes a step S1 for identifying electric arcs (or more precisely, for identifying a voltage measurement exceeding a threshold value representative of the presence of an electric arc) and a step S2 for storing a signal in a (second) time window corresponding to the identification, i.e., the window in which this electric arc is identified. As previously discussed, there is a correlation between the electrical behavior during the occurrence of an electric arc and a pre-disruptive phenomenon. Therefore, the signals, or patterns, corresponding to these detected arcs can then be used to determine the presence of pre-disruptive phenomena during the online phase P1.

[0075] This allows us to bypass the difficulty of directly detecting disruptive phenomena.

[0076] It is possible to store a set of signals, or patterns, each corresponding to the behavior of the electric current when an electric arc appears for the electrical circuit considered.

[0077] These stored signal sets can be transferred to a memory associated with a monitoring device for an electrical circuit onboard the aircraft. This transfer can be carried out by any means, conventional or otherwise.

[0078] This monitoring device is adapted to allow the determination of a risk of occurrence of an electric arc in an electrical circuit of an aeronautical system of the aircraft.

[0079] Figure 5 illustrates one embodiment of a determination device 60 of a risk of the appearance of an electric arc in an electrical circuit 50a. In this example, the device 60 includes a sensor 61 allowing continuous measurement of the electrical current and voltage within the monitored electrical circuit 50a, a computer 62 and a memory 63.

[0080] The computer 62 has computing power enabling it to perform correlation measurement calculations, including frequency correlation (involving a Fourier transform) on the fly.

[0081] The memory 63 is adapted to contain predetermined signals, or patterns, which may originate from a remote platform (outside the aerial vehicle) 70. Data transfer can be carried out by any means of communication, typically during an initialization or update phase of the various electronic circuits of the vehicle.

[0082] This embedded device 60 for determining the risk of an electric arc occurring in the associated electronic circuit is configured to implement the The online phase, Pon, previously mentioned and illustrated in the embodiment of [Fig. 3], comprises iterative steps enabling continuous monitoring of the electric current and the detection at any time of a risk of arcing, thus allowing, if necessary, proactive action to prevent its occurrence. In the embodiment illustrated in [Fig. 3], this iterative aspect is represented by the loop connecting step S6 to step S3.

[0083] Step S3 includes the extraction of a sample of an electric current corresponding to a (first) time window, said electric current being measured continuously in said electrical circuit.

[0084] This extraction is preferably carried out with a current sensor having the same characteristics as that used, where appropriate, for the offline data acquisition phase Poff.

[0085] The first time window corresponding to a current measurement corresponding to the sample preferably has the same size (i.e., the same number of sampled discrete values) as the second time window, i.e., the one corresponding to the signals, or patterns, stored during the offline phase of Poff data acquisition. This makes subsequent correlation calculations easier and faster.

[0086] This sample consists of, or comprises, measurements of the intensity I of the electric current. Thus, according to one embodiment, the extracted sample comprises a series of intensity measurements, the number of which depends on the size of the time window and the sampling frequency.

[0087] For each extracted sample, a step S4 is triggered, consisting of evaluating at least one correlation measure between that sample and the set of predetermined signals, or patterns, stored in memory. These stored signals can typically be those provided during the data acquisition phase. As seen previously, these signals, or patterns, are representative of an electric arc, due to the data acquisition process corresponding to the offline phase Poff.

[0088] The extracted sample can also be stored to allow for correlation calculation. It can then be deleted, or overwritten by storing a subsequent sample.

[0089] One possible correlation measure is a temporal correlation measure between values ​​in the sample and values ​​in the stored signal set. In particular, a correlation measure can be determined between the sample and each of the signals available in the stored set. This temporal correlation measure corresponds to step S41 in the example in [Fig. 3].

[0090] According to one embodiment, this temporal correlation measure r^y is a Pearson correlation measure.

[0091] In the preferential case where the sample and the stored patterns have the same size, that is, they contain the same number n of discrete values, this Pearson correlation measure can be written:

[0092] i\~'^ 2 1 x^L^-yy-

[0093] x represents the sample. x; is the ith intensity value of this sample. Similarly, y represents a predetermined signal, or pattern, and y; is the ith intensity value of this pattern. * and represent the average intensities of, respectively, the sample x and the pattern y.

[0094] According to another embodiment, this temporal correlation measure is a Spearman correlation measure.

[0095] Spearman's correlation is the non-parametric equivalent of Pearson's correlation. It also assesses the relationship between two sets of values, but it does not use the values ​​of the data but their rank, Rx, Ry respectively, for the sample and a pattern.

[0096] Before calculating Spearman's rank correlation, the data must be transformed into ranks. To do this, the data are sorted in ascending order, and the values ​​are replaced by their ranks. When values ​​are identical, the average of their ranks is used. Once the data have been transformed into ranks, the Spearman correlation measure can be calculated using the same formula as that used to calculate Pearson's correlation measure, rx > y, but using the ranks.

[0097] Thus, we can write:

[0098] t _

[0099] Another possible correlation measure is a frequency correlation measure based on the values ​​of the frequency transforms of the sample and the predetermined signal set, respectively. In particular, a correlation measure can be determined between the sample and each of the signals available in the stored set. This frequency correlation measure corresponds to step S42 in the example in [Fig. 3].

[0100] This frequency transform is typically a Fourier transform. According to this embodiment, one can therefore calculate the magnitude of the Fourier transform of the sample x, denoted x, and of the patterns stored in memory, denoted y, and then calculate the Pearson or Spearman correlation, r^, between these two quantities.

[0101] For Pearson's correlation, we can then write, with p being the number of frequencies:

[0102]

[0103] And for Spearman's correlation, we can write, with p being the number of frequencies:

[0104] f r ' y ~

[0105] According to one embodiment, in a step S5, an aggregate correlation measure can be determined from a plurality of correlation measures, for example from a temporal correlation measure and a frequency correlation measure.

[0106] For example, one can write:

[0107]

[0108] These correlation measures can be determined for each stored pattern. Each electric arc can indeed possess several different characteristics, and the constitution of a set of patterns of substantial size makes it possible to maximize the probability of detecting pre-disruptive phenomena.

[0109] From a set of correlation measures rx^' determined for a set of predetermined signals, or patterns, a unique correlation measure can be derived, for example by calculating the maximum value obtained, or the average value, etc.

[0110] The process then includes a step S6 of predicting the risk of an electric arc occurring at a time after the time window corresponding to the extracted sample (on which the correlation measure was calculated) as a function of this correlation measure.

[0111] To achieve this, one possible implementation consists of predicting the appearance of an electric arc as soon as a correlation measurement exceeds a given threshold.

[0112] Another possible implementation consists of making the prediction of the occurrence of an electric arc subject to a plurality of exceedances of a given threshold. For example, if the value of the correlation measurement exceeds this threshold, for example, 3 times, then the occurrence of a fault arc can be predicted.

[0113] The threshold may, for example, correspond to the median value of a correlation rate, i.e. 0.5.

[0114] The number of overshoots taken into account must not be too high, since this mechanically delays the time at which the prediction can be made and therefore the proximity of the time when the electric arc may actually occur. The figure of 3 given previously seems a good compromise between optimizing the risk estimate and the need to estimate this risk early enough and, at a minimum, before the electric arc occurs.

[0115] According to one embodiment, step S6 makes it possible to predict a risk of occurrence that is not binary, but a probability value. This probability can be related to the value, or values, of the correlation measure, without using thresholding. In the case where a single correlation measure is taken into account, the risk of occurrence can be equal to the correlation measure. If several correlation measures are taken into account (for example, 3), the risk of occurrence can be the average of these measures.

[0116] When no risk is predicted at step S6, the process loops back to step S3 to extract a new sample, to which the same steps will be applied. Thus, continuous monitoring of the electric current can be carried out.

[0117] When at step S6 a (non-zero) risk is predicted, then a step S7 triggering a preventive action can be carried out.

[0118] This step S7 may include the automatic sending of an instruction to cut off the electrical current to a cutting system associated with the monitored electrical circuit.

[0119] Thus, the predicted imminent occurrence of an electric arc can be prevented from actually appearing in the circuit. The occurrence of electric arcs, and in particular series arcs which are difficult to detect as previously explained, can therefore be eliminated.

[0120] Also, an alert message can be relayed to an interface allowing an operator to be informed of an anomaly. A human operator can then intervene to remedy this fault (activation of a possible redundant system following the shutdown of the faulty system, interventions to troubleshoot the fault, etc.).

[0121] It can also be noted that the proposed method for determining the risk of occurrence, based on a correlation measure, also makes it possible to detect the presence of an electric arc. Indeed, if an electric arc were to occur (for example, because it was not preceded by pre-disruptive phenomena), it would be detected according to the same mechanism as for a pre-disruptive phenomenon because it exhibits a high correlation with a stored pattern.

[0122] Now, since an electric arc denotes abnormal behavior, the appearance of an electric arc is associated with a high probability of the appearance of new arcs in the future. The described process therefore makes it possible to anticipate the appearance of future arcs as soon as one or more "first" electric arcs appear. Thus, according to the same mechanism as for a pre-disruptive phenomenon, detections of electric arc(s) will generate the determination of a risk of the appearance of an electric arc.

[0123] Fig. 4 illustrates an example that highlights certain aspects of the proposed process and its advantages.

[0124] This figure illustrates from top to bottom the current I and voltage U measured for a monitored electrical circuit, typically comprising a connection (terminal block...), then a temporal correlation measure r', a frequency correlation measure rf, and an aggregate correlation measure r.

[0125] In zone Zl, no arc is present because the voltage U does not exceed the threshold (e.g., 15V). However, it is observed that the time correlation r' and frequency correlation rf are higher than their nominal values ​​(close to zero). Consequently, the aggregate correlation r exceeds the threshold rs three times. According to one implementation, this triple exceedance generates the prediction of a risk of electric arcing in the future.

[0126] Normally, according to the proposed process, in the online step Pon, the prediction of such a risk should trigger the interruption of the electrical current. In the illustrated example, for explanatory purposes, no interruption is carried out, and the electrical circuit remains energized.

[0127] This shows that a first electric arc appears in zone Z2, thus confirming the previously made prediction. This example clearly demonstrates the effectiveness of the proposed method, which makes it possible to predict the occurrence of electric arcs and prevent their occurrence by anticipating their onset through the interruption of the electrical circuit.

[0128] Of course, the present invention is not limited to the examples and embodiment described and illustrated, but is defined by the claims. In particular, it is susceptible of numerous variations accessible to those skilled in the art.

Claims

Demands

1. A method for determining the risk of an electric arc occurring in an electrical circuit of an aeronautical system, comprising iterative steps (Pon) of: extracting (S3) a sample of an electric current corresponding to a first time window, said electric current being measured continuously in said electrical circuit; evaluating (S4) at least one correlation measure between said sample and a set of predetermined signals, each signal being representative of an electric arc; predicting (S6), based on said at least one correlation measure, the risk of an electric arc occurring at a time after said first time window.

2. A method according to the preceding claim, wherein said at least one correlation measure comprises a temporal correlation measure between values ​​of said sample and values ​​of said predetermined signal set.

3. A method according to any one of the preceding claims, wherein said at least one correlation measure comprises a frequency correlation measure between values ​​of the frequency transforms, respectively, of said sample and of said set of predetermined signals.

4. A method according to any one of claims 2 or 3, wherein said at least one correlation measure is determined by a Pearson correlation measure and / or by a Spearman correlation measure.

5. A method according to any one of the preceding claims, wherein the prediction of a risk of the occurrence of an electric arc generates a break (S7) of said electrical circuit.

6. A method according to any one of the preceding claims, wherein said set of predetermined signals is previously obtained by data acquisition steps (Poff) comprising: - identification (SI) of a voltage measurement corresponding to said electrical circuit exceeding a threshold value representative of the presence of an electric arc; - memorization (S2) of a signal corresponding to intensity measurements in a second time window corresponding to said identification.

7. A method according to the preceding claim, wherein said data acquisition steps are carried out on a remote platform, and all the signals memorized by said data acquisition steps are stored within an aerial vehicle for the implementation of a method for determining the risk of occurrence of an electric arc according to any one of claims 1 to 5.

8. A computer program comprising instructions for implementing a method according to any one of claims 1 to 5, when executed on an information processing platform.

9. Device for determining the risk of an electric arc occurring in an electrical circuit of an aeronautical system, adapted to extract a sample of an electric current corresponding to a first time window, said electric current being measured continuously in said electrical circuit; to evaluate at least one correlation measure between said sample and a set of predetermined signals, each signal being representative of an electric arc; and to predict, as a function of said at least one correlation measure, a risk of an electric arc occurring at a time after said first time window.

10. Aerial vehicle comprising at least one device according to the preceding claim.

Images