FR2976560A1 - Method for maintenance of e.g. equipment, of rotary wing aircraft following flight, involves drawing list of maintenance activities based on localization and/or provisional rules, and allowing operator to act technically on aircraft unit - Google Patents

Abstract

The method involves using a diagnostic table (50) for associating an event (23) to an aircraft equipment (3) e.g. autopilot unit. An interim rule is created and recorded for each unidentified group for associating events and phase of flight of the unidentified group with corresponding set using the diagnostic table. A list of maintenance activities (40) is drawn based on localization rules and/or provisional rules, and an operator is allowed to act technically on the aircraft equipment to carry out maintenance (P7). An independent claim is also included for an aircraft maintenance automated diagnosis system.

Description

Maintenance process, system and aircraft. The invention relates to maintenance. It applies in particular to rotary wing aircraft, for example helicopters. Its application to other vehicles, including aircraft, is also quite possible. In particular, the invention aims at monitoring and automatic maintenance diagnostics. By maintenance diagnosis, we mean the use of information in order first of all to know if a malfunction has occurred and, secondly, to determine, if need be, the cause of this anomaly of functioning. act of equipment, flight event or flight environment eg meteorology. In an aircraft, various structures integrate an on-board electronic installation (hardware and / or software) called "embedded electronics" for simplification. For example, they may be avionics systems for automatic piloting, navigation, verification of functionalities (eg the so-called "HUMS" checks for the English "Health and Usage Monitoring Systems"), or systems Communication. These structures include various on-board electronic equipment, which are designated by "L.R.U. For "Line Replaceable Unit". These devices themselves comprise various components called equipment components (e.g., in the form of electronic modules such as cards and / or embedded software) which are called "S.R.U. "For" Shop-Replaceable Unit "in the aeronautical field.

Subsequently, and unless otherwise specified, the term "member" at least one embedded electronic equipment and / or at least one component of a device and / or at least one software.

Especially in aeronautics, the maintenance of on-board electronics is crucial. Maintenance includes operations including control, overhaul, replacement and repair. In terms of cost, a significant part of the operating costs of an aircraft is related to the maintenance of the on-board electronics, which influences the availability of an aircraft and therefore its profitability. For example, approximately one quarter of the direct maintenance costs of a current helicopter are dedicated to avionics systems alone (hardware and / or software combined). With this in mind, various techniques have been proposed to better plan maintenance. In particular, we try to make as automatic as possible prognoses and diagnostics maintenance.

But we understand that it is difficult (or even antagonistic) to optimize the scope, timing and duration of maintenance operations, while improving the security offered. In aeronautics, if an organ is suspected of being the source of a malfunction, this organ is called "offending organ" for convenience. It is understood that when several organs are incriminated, it may be difficult to produce a precise message indicating which organ must undergo a given maintenance operation.

Thus, it can lead to an unnecessary maintenance operation or even the anticipated replacement of a still usable body. Conversely, if due to an incorrect location a failing organ does not undergo a necessary maintenance operation, the risks are unacceptable. These difficulties are exposed in relation to previous aeronautical maintenance techniques. According to an ARINC624-1 specification, an integrated test called "B.I.T" (English: "Built-In Test") is required before each use of the aircraft, even during or after the flight. These integrated tests are used in particular to establish a fault diagnosis. An onboard maintenance computer may implement other known techniques to update operating anomalies in use called "events", potentially synonymous with a malfunction of an aircraft member. Therefore, according to a method, during operation of the aircraft, the maintenance computer lists all the events symptomatic anomalies of operation of this aircraft, associating an identifier of this event to a dating element of this event. For example, during a census step, the maintenance calculator uses a list of events associating an identifier with each predetermined event, this identifier being commonly referred to as "fault code" by the person skilled in the art. As a result, when one of said events occurs, the maintenance calculator informs a collection list listing the identifiers of the events occurring and the times at which these events occurred, in the form of a collection list. Reference will be made to the literature for details of such a census of events representative of operating anomalies. In order to carry out the maintenance of the aircraft, it is necessary to identify a potentially faulty organ at the origin of one or more detected events. 10 Therefore, a processing computer possibly embedded and confused with the maintenance computer, performs a temporal grouping of stored events according to given time criteria. Thus, the collection list is divided into at least one group of events. A list of location rules is then used to identify the organs likely to have generated the appearance of an event group. Such a list of rules then comprises a plurality of rules associating at least one event with at least one member, each member being associated with an occurrence coefficient. If an event group corresponds to a given rule, an operator is instructed to check the operation of the offending organs mentioned in said given rule. If an offending member is indeed down, an operator increments the occurrence coefficient of that member by one unit in the corresponding rule, and performs the required repair. On the other hand, if an event group does not correspond to any rule, we create a temporary rule. An operator then examines this provisional rule and validates it by adding it to the list of rules if the operator considers the provisional rule useful. An interim rule is therefore a rule proposal submitted to an operator for approval. In addition, various documents relating to the field of aircraft maintenance are known.

The document FR2938676 describes a sequential pattern recognition for the processing of fault messages. For fault diagnosis of aeronautical system equipment, fault messages are analyzed and filtered. From sequential patterns, we construct a model of parasitic messages in the form of a deterministic finite automaton. This model is injected into an onboard processing computer in real time. Through a succession of algorithms, a posterior analysis is carried out on the ground to improve the long-loop diagnostic tool. As the outages appear, we get an adaptation to the new failures and an automatic improvement. According to this document, a segmentation is operated solely by simple coupling of a time criterion. Document FR2934397 describes the organization of aircraft maintenance. In this document, an algorithm concerns the calculation of the lifetime of mechanical parts exclusively. The PENTOM 2007 document titled "Diagnostics of avionics equipment by time correlation" provides that an anomaly is detected through the execution of integrated tests.

The identification of a time window makes it possible to determine a degree of correlation between failures, to distinguish the messages from faults. A flight phase is mentioned as information stored in an onboard computer at the appearance of fault codes, but without specifying how it can be used. We can refer to: http://hal.archives-ouvertes.fr/docs/00/18/54/92/PDF/Pentom ArnaudO7.pdf The invention aims in particular at an alternative method aimed at optimizing the maintenance of a aircraft. In summary, one or more aircraft in operation produce maintenance collection lists which, taken as they do not obviously indicate the causes that caused the memorized events. The invention makes it possible to determine said causes by judicious exploitation. The invention notably makes it possible to integrate the feedback from a fleet and / or a range of aircraft. From events recorded in operation, the invention thus makes it possible first to establish whether these events are symptomatic of actual failures, then the repair of these failures.

As a result, the subject of the invention is a technical method for repairing an aircraft using a highly automated maintenance diagnosis which, from the integration of a large number of symptoms, determines the origins and causes of failures with precision and precision. particularly fast and reliable way. With such a maintenance diagnosis, maintenance efficiency and performance are increased in terms of time and cost, but also by reducing aircraft downtime (increase in readiness) due to anomalies. unresolved operating conditions. According to the invention, during a technical maintenance process, one acts technically on at least one organ of an aircraft following a flight, according to a collection list listing events occurring in flight and a list of rules location, each location rule associating at least one organ to be checked as well as at least one event and a flight phase, an occurrence coefficient being associated with each organ of a location rule.

Therefore, during a census step, a maintenance calculator lists in flight events representing operating anomalies in a collection list associating an event as well as on the one hand a dating element of this event and on the other hand flight parameters determined using a flight recorder. When an anomaly-type event appears, the maintenance calculator saves an identifier representing this event in the collection list, and for example associates with this identifier the date and time when the event occurred or the date and the time at which the event disappeared. The maintenance calculator can use an event list to determine the identifier of a particular event. In addition, a usual flight recorder continuously records flight parameters that are for example to be chosen from a list including: a rotational speed of at least one engine of an aircraft, a rotational speed of a rotor of the aircraft, a torque generated by said engine, a position of at least one flight control, a geographical location of the aircraft, a pressure and a temperature outside the aircraft, a pressure and a temperature of a hydraulic fluid of a main gearbox driving said rotor, a temperature of the gases present in said engine, an altitude of the aircraft, at least one speed of movement in the space of the aircraft, at least one acceleration of the aircraft, a quantity and a flow of fuel, a mass of the aircraft.

This method also implements the following steps, one or more of these steps can be performed in the aircraft or outside the aircraft. Thus, during a grouping step, the collection list is partitioned into at least one group of events according to at least one predefined temporal criterion. Therefore, at least one group of events is formed, each group comprising at least one event. This time criterion can be a constant duration, of the order of one second for example, or be a variable duration. Such a variable duration may possibly have a first constant value before a given fact, such as the ignition of an engine possibly, and a second constant value after said fact.

In addition, during an association step, each event is associated with a flight phase, the flight phase being determined using the flight parameters, and for each group a flight phase during which the events of the group.

The association step can be carried out in parallel with the census step, in parallel with the regrouping step or following the census step or the regrouping step. In addition, the association step may comprise one or more association phases, for example a preliminary association phase associating each event with a flight phase, then a final association phase associating each group with a phase of association. flight according to the flight phases of events in this group. The flight phase of an event can be obtained using a list of flight phases associating each flight phase with values determined by the manufacturer of the flight parameters.

For example, the flight phase list includes a level flight phase, a hover phase, a cornering flight phase, a flight phase when approaching a landing field, a flight phase of a flight. take off, or even a generic flight phase including all phases of flight or one or more phases of transition flight. Note that a particular group may include at least a first event associated with a dating element and a first flight phase and a second event associated with a dating element and a second flight phase. as a result, this group can be built by including a transition flight phase corresponding to the transition from the first phase of flight to the second phase of flight. According to one variant, during a preliminary association phase of the association step performed for example before the regrouping step, each event is associated with a flight phase by using the list of flight phases. Then, during a final association phase of the association step performed after the grouping phase, the flight phase of each group is defined. Each group then associates a flight phase with a plurality of events each linked to a dating element. Therefore: - if a group includes a plurality of events each associated with the same particular flight phase, this group includes said particular flight phase as well as said events and the corresponding timing elements, - if a group includes a plurality of events associated with distinct flight phases, this group includes a transition flight phase as well as said events and the corresponding timing elements During an updating step, it is then possible to add the new transition phase in the flight phase list. Consequently, during an analysis step, the events and the flight phase of each group are compared with the events and the flight phase of each location rule, an organ cited in each elected location rule corresponding to a group becoming an incriminated organ, each group not corresponding to a location rule becoming an unidentified group. Therefore, a group resulting from the grouping step is: - either an identified group corresponding to a location rule referred to by following location rule elected for convenience, when there is identity between the events and the flight phase of the group with events and the flight phase of a location rule, or an unidentified group not corresponding to any location rule.

It should be noted that at least one location rule can associate at least one organ to be verified as well as at least one event and a generic flight phase covering all phases of flight. During the analysis step, if a particular group associates a particular flight phase with particular events and if a particular location rule associates the particular events with a generic flight phase, then the group is considered to be an identified group. , the given location rule becoming an elected location rule.

If a group is an identified group, the organs mentioned in the corresponding localization rule are referred to as "incriminated organs" for convenience, these incriminated bodies being capable of causing a malfunction. On the other hand, during an identification step, by using a diagnostic table associating an event with at least one organ, a provisional rule is created and recorded for each unidentified group by associating the events and the flight phase of said non-identified group. identified to corresponding bodies established using the diagnostic table, the bodies associated with this provisional rule becoming incriminated organs. Therefore, for each event, the diagnostic table associates at least one organ likely to be at the origin of said event. Such a diagnostic table is established by the manufacturer of the aircraft. From each unidentified group, a temporary rule is established by associating the events of this unidentified group with the corresponding organs, namely the organs established using the diagnostic table possibly restricted by the manufacturer following a return. experience. These organs then become de facto offending organs. Moreover, it is understood that we associate an occurrence coefficient, for example zero, to each organ of a provisional rule.

It will be understood that the list of location rules includes location rules validated by an operator, but also location rules to be validated and provisional rules resulting from the last flight. The term "location rules" therefore refers to the location rules validated by an operator and the location rules to be validated established with provisional rules following other flights, and by "provisional rules" the rules established as a result. the study of the maintenance data of the last flight of an aircraft. During a maintenance step, a maintenance list is then established comprising at least one incriminated member cited in at least one elected location rule and / or at least one temporary rule, then an operator acts technically on each incriminated member to perform its maintenance. . It should be noted that a builder may deliberately decide to use only the validated location rules to establish the maintenance list. This technical process therefore makes it possible to repair an aircraft, by providing an operator with a list of the incriminated organs to be checked following a flight having given rise to operating anomalies. This operator then acts technically on the incriminated organs to check and repair them or replace them if necessary. Therefore, this method makes it possible to extract from a collection list combinations of events having logical connections to each other to improve the performance of the process. This method may further include one or more of the following features. For example, prior to the grouping step, a filtering step is performed. The filtering step may include a step of deleting the redundant events. An organ can send an abnormal signal to two other separate organs for example. The maintenance computer can then store two separate events relating to said signal. During the filtering step, one of the said events is then eliminated. In addition, prior to the grouping step, an event occurring momentarily during an appearance time separating a first instant corresponding to the occurrence of said event and a second instant corresponding to the disappearance of said event, it is possible to carry out a filtering step by removing from the collection list the events occurring momentarily during a duration of occurrence less than a threshold. As a result, we do not consider an event appearing and disappearing quickly. Likewise, prior to the grouping step, by using a list of undesignated events identified as not being symptomatic of failures, a filtering step of the collection list is carried out by suppressing said undesired events. a segmentation step, it is possible to segment each unidentified group into at least one alternative group, each alternative group comprising the flight phase of the unidentified group from which it is derived and at least two events associated with the same organ, and then creating a provisional rule by alternating group, each organ cited in an alternative group being an incriminated organ. Thus, we gather all the events of the unidentified group considered that are associated with the same organ in an alternative group. Thus, combinations are made no longer specific to a collection list sorted only temporally. As a result, the probability of finding these combinations in future flights increases.

In addition, each unidentified group is retained through a temporary rule. Therefore, we do not rule out the fact that the "random" part that constitutes it may have an interest in learning.

In another aspect, the steps of grouping, analysis, identification or even segmentation and maintenance can involve a processing computer. The treatment computer and the maintenance computer can be confused or distinct, the processing computer can be embedded or not embedded. Moreover, during the segmentation step, if at least two given alternative groups consist of the same events, these two alternative groups are concatenated to form a single alternating group associating said same events with the members of the at least two alternative groups. According to another aspect, during an updating step, a maintenance list comprising at least one incriminated organ cited in at least one elected location rule and / or at least one temporary rule updated during said identification steps and for segmentation, an operator for example increments by one unit the occurrence coefficient of each incriminated organ judged to be out of order in each elected location rule and each provisional rule citing that incriminated organ judged to be out of order. For this purpose, these elected and provisional location rules can be stored in the maintenance list or in a dedicated file for example. On the other hand, if an operator notes that for the same set of events, the list of location rules comprises several rules associated with different flight phases but with the same members, these members having substantially similar occurrence coefficients, this operator can concatenate said rules to obtain a single rule associated with a generic flight phase. Indeed, in this configuration, the single rule seems independent of the flight phase. In addition, during an update step, an operator determines whether a temporary rule should be deleted, or retained in the list of location rules possibly by validating it to transform it into a validated rule.

Note that the list of location rules can be established taking into account the experience accumulated on all aircraft of the same type. According to the invention, the diagnostic table being a generic table adaptable to a plurality of aircraft, it is possible to modify the diagnostic table to adapt it to the aircraft in question. In addition to a method, the invention relates to an automated aircraft maintenance diagnostic system comprising an on-board electronic electronics comprising at least one member, this system having at least one maintenance computer listing in flight events representing malfunctions in a collection list. This system implements the method according to the invention according to a collection list identifying events occurring in flight, a datum element of this event and flight parameters determined using a flight recorder, and a list of location rules, each location rule associating at least one organ to be checked as well as at least one event and a flight phase, an occurrence coefficient being associated with each organ of a location rule .

This system comprises a maintenance tool provided with a memory and a processing computer including a calculation unit, said calculation unit executing stored instructions: during a grouping step, for partitioning said collection list into at least one group of events according to at least one predefined temporal criterion, during an association step, for associating each event with a flight phase, said flight phase being determined using said flight parameters, each group comprising a flight phase and at least one event associated with a dating element .,. during an analysis step, for comparing the events and the flight phase of each group with the events and the flight phase of each location rule, an organ cited in an elected location rule corresponding to a group becoming an offending organ stored in a maintenance list, each group not corresponding to a location rule becoming an unidentified group,. during an identification step, using a stored diagnostic table associating an event with at least one organ, to create and record a provisional rule for each unidentified group by associating the events and the flight phase of said unidentified group with the organs correspondents established using said diagnostic table, the bodies associated with this provisional rule becoming incriminated organs. during a maintenance step, to establish a maintenance list comprising at least one incriminated organ cited in at least one elected location rule and / or at least one temporary rule so that an operator acts technically on each incriminated organ to carry out its maintenance . The computing unit may also require storage of the elected location rules and provisional rules for subsequent adjustment purposes, so that an operator can adjust the coefficient of occurrence of an organ of said rules in the event of a confirmed failure. of said body. It should be noted that the maintenance diagnostic system may comprise an operator-operable control means for, in particular, updating the provisional rules and the location rules, such as a means comprising a keyboard and a screen, for example. Finally, the invention relates to an aircraft of the type provided for implementing the method according to the invention, this aircraft comprising a rotary wing. The invention and its advantages appear in more detail in the description which follows. This description is in no way limiting, and illustrates examples of embodiments of the invention, with reference to the appended figures. FIG. 1, a diagram of an aircraft according to the invention, FIG. 2, a diagram illustrating the method according to the invention, FIG. 3, a diagram illustrating a collection list, and FIG. 4. a diagram illustrating a segmentation step of an unidentified group FIG. 1 shows an aircraft 1 with a rotary wing. This aircraft 1 comprises a plurality of members 3, and in particular members 3 of an avionics system 2. Consequently, the aircraft comprises members 3 including on-board electronic equipment 4, embedded electronic equipment components 4, integrated in hardware form 5 and / or in software form 6. A common hardware form of such embedded electronic equipment components 4 is an integrated electronic card. A typical software form of such embedded electronic equipment components 4 is a computer program, a database or the like. The aircraft 1 has, for example, a man-machine interface 8 with at least one on-board electronic equipment 4 in the form of a display screen 9. In addition, the aircraft 1 comprises an automated maintenance diagnostic system 7. This system 7 includes an onboard maintenance computer 10 capable of recording events 20 symptomatic of a malfunction. In addition, this system 7 includes a maintenance tool 16 communicating with the maintenance computer 10. This maintenance tool 16 is provided with a memory 17, and a processing unit 18 including a calculation unit 19. maintenance tool 16 can be embedded in the aircraft 1, being confused with the maintenance computer for example, or be located outside this aircraft 1. The maintenance tool 16 further includes a flight recorder 102 for record flight parameters 101 determined using standard measuring means. The flight parameters are for example to be chosen from a list including: a rotation speed of at least one engine of an aircraft, a rotational speed of a rotor of the aircraft, a torque generated by said engine, a position of at least one flight control, a geographical location of the aircraft, a pressure and a temperature outside the aircraft, a pressure and a temperature of a hydraulic fluid of a main power transmission gearbox causing said rotor, a temperature of the gases present in said engine, an altitude of the aircraft, at least one speed of movement in the space of the aircraft, at least one acceleration of the aircraft, a quantity and a flow of fuel, a mass of the aircraft. This system 7 aims to detect operating anomalies and locate the origin of its anomalies to allow the intervention of an operator. The maintenance diagnostic process performed facilitates the integration of experience feedback specific to the aircraft 1 and / or to other aircraft, and aims to avoid the ambiguities encountered with the known techniques, for the localization of an organ 3 failed. With reference to FIG. 2, the invention comprises a technical method, that is to say a sequence of actions, possibly reiterative, and typically executable on a computer such as the maintenance tool. operation of the aircraft 1 and the operating conditions of the aircraft 1, maintenance information is updated on board, in a collection list 22. In fact, during a census step P1, according to In known techniques, the maintenance computer 10 lists in flight events that are symptomatic of operating anomalies. Thus, the maintenance calculator establishes in flight a collection list listing, on the one hand, the events 23 occurring in flight and, on the other hand, a dating element of each event including, for example, a first instant 25 at which an event occurred and a second instant 26 at which said event has disappeared. It should be noted that it may be easier to identify an event by an identifier, rather than by a word or an expression. Therefore, the maintenance calculator 10 can use a list of events 21 associating each event 23 with an identifier 24, the collection list then listing each event by means of its identifier 24. The list of events 21 is classically from a design phase of the aircraft 1, but may carry information from other aircraft. In addition, the maintenance computer associates each event with flight parameters 101 determined by the flight recorder. According to one option, during a preliminary association phase, the maintenance calculator can deduce therefrom a flight phase during which a given event has occurred, using a flight phase list established. by the manufacturer. The flight phase list may include a level flight phase, a hover phase, a cornering flight phase, a flight phase approaching a landing field, a take-off flight phase, even a generic flight phase including all phases of flight. Therefore, according to the variant, the collection list includes for each event either an identifier of the flight phase of the aircraft at the time of occurrence of the event, or flight parameters 101 making it possible to determine said phase of flight. flight later. The collection list is transmitted to the treatment calculator 18 for operation in the aircraft or outside the aircraft.

At this point, it is not desirable in all embodiments to group the events 23 occurring using time criteria. Thus, during an optional filtering step P9, the processing calculator 18 filters the collection list 22 to suppress the redundant events 23, the events 23 appearing momentarily and the events 23 reproduced at regular intervals in particular. Consequently, during a grouping step P2, the processing computer 18 partitions the collection list, possibly filtered, into at least one group according to a time criterion. With reference to FIG. 3, the processing computer 18 can therefore partition the collection list into a plurality of groups 30, four groups 30 according to the example shown, each event 23 appearing through its identifier 24. It should be noted that FIG. 3 shows neither the phase of flight associated with an event nor the flight parameters associated with an event so as not to burden this FIG.

With reference to FIG. 2, in one embodiment during which the flight phases associated with the events have not yet been determined, during a preliminary association phase of an association step P10, each event 23 is associated with a flight phase 100, the flight phase 100 being determined using said flight parameters 101 and a list of flight phases 103. Then, during a final association phase, a phase is determined for each given group. flight during which the events of this particular group occurred. Each group 30 then comprises a flight phase 100 and at least one event 23 associated with a dating element 25, 26. If the events of a given group have occurred during the same flight phase, this group 30 includes the information following: said same phase of flight, said events and the elements of dating of these events. By cons, a group 30 may comprise at least a first event 23 associated with a dating element 25, 26 and a first phase of flight and a second event 23 associated with a dating element 25, 26 and a second flight phase. It can then be considered that the first event and the second event occurred during a so-called transition flight phase, for example, a new transition phase created for this purpose or an existing transition phase.

In this configuration, the group 30 includes the following information: said transition flight phase, the first and second events, and the first and second dating elements. Therefore, it is understood that during a preliminary association phase, each event is associated with a so-called "individual flight phase" flight phase for convenience, insofar as this individual flight phase is related to a given event. Moreover, during a final association phase, all the events of a given group are associated with a so-called "general flight phase" flight phase for convenience. If the events of a group occurred during the same individual flight phase, the general flight phase of the group is equal to that particular individual flight phase. On the other hand, if the events of a group occurred during separate individual flight phases, the general flight phase is equal to a transition flight phase. Therefore, during an analysis step P3, the processing calculator 18 implements a list of location rules 60 comprising a plurality of location rules 61. Each location rule 61 associates at least one event 23 and a phase For example, the first rule 61 'associates three events 23', represented by their identifier 24, with two members 3 'and a flight phase "A". According to this first rule 61 ', if a group 30 comprises only the three events 23' and this general flight phase "A", it is necessary to examine the two members 3 '. In addition, it will be noted that each member 3 is associated with an occurrence coefficient 62. This occurrence coefficient 62 indicates the number of occurrences of failures of a given organ 3 that led to the appearance of the rule concerned. . Thus, if a first member of a rule has a high rate of occurrence compared to the other members of this rule, this first member should first be examined.

Furthermore, it is noted that the list of location rules 60 is updated according to the maintenance actions developed on all aircraft of a given type. In addition, it is noted that the list of location rules 60 may include validated rules and proposed rules to validate subject to the approval of an expert based on feedback. Therefore, the processing calculator 18 compares each group 30 with the location rules 61.

If the events and the general flight phase of a group are identical to the events and the flight phase of a given location rule, this group is called "identified group" 31 for convenience, the location rule being called " location rule elected for convenience.

Each member of an elected locating rule is then an organ referred to as "incriminated organ" for convenience, to be used during a subsequent maintenance step P6. Conversely, groups not corresponding to any rule of the location rule list 60 are referred to as "unidentified groups" for convenience. For each unidentified group, the processing computer 18 implements an identification step P4 and a segmentation step P5. During an identification step P4, the processing computer 18 implements a diagnostic table 50 associating with each event 23 at least one member 3. This diagnostic table is established at the design of the aircraft by a manufacturer, this table can be updated if necessary based on feedback. The diagnostic table may be a generic table modified to fit a particular aircraft. For example, the maintenance tool includes a control means 15 for indicating the type of aircraft, the diagnostic table being modified according to this information. When the identification step is implemented inside the aircraft, a computer can determine the organs present on board and restrict the diagnostic table to these organs. Regardless of the variant, the diagnostic table 50 is intended to indicate the organs likely to be at the origin of an event, these organs being represented in the form of an identifier sometimes called "incrimination value". Therefore, for each unidentified group, the processing means 18 determines a rule proposal in the form of a provisional rule associating the events of the unidentified group, the general flight phase during which said events occurred, and the organs likely to induce said events. These organs become incriminated organs to be exploited during the maintenance step P6. In addition, the interim rule is exploited during an update step P8. During a segmentation step P5, the treatment computer 18 can segment each unidentified group into 25 alternative groups leading to the development of provisional rules and the determination of incriminated organs, each alternative group being associated with the flight phase of the flight. general flight phase of the unidentified group from which it originates.

With reference to FIG. 4, the organs 3 corresponding to each event 23 are determined using the diagnostic table, namely the organs known under the identifiers "132, 140, 144, 131" for the event having for identifier "124019", the organs known under the identifiers "135, 140, 144, 131" for the event having for identifier "125019", the organs known under the identifiers "134, 140, 144, 133" for the event having for identifier "126019" and the bodies known under the identifiers "133, 140, 144" for the event having for identifier "127019". Each alternative group is obtained by grouping the events by common organ, each alternative group comprising at least two events associated with the same organ and the flight phase corresponding to the unidentified group.

It should be noted that this flight phase does not appear in FIG. 4 so as not to weigh down this FIG. 4. Moreover, if two alternative groups 33 ', 33 "comprise the same events, these two alternative groups are merged. the treatment calculator creates a provisional rule by alternating group according to our example: - a first provisional rule composed of four events having for identifying the codes "124019, 125019, 126019, 127019" and being associated with two equipments having for identifying the values "140, 144" and at a particular phase of flight, - a second provisional rule composed of two events having for identifying the codes "124019, 125019" and being associated with equipment having for identifier the value "131" and at the phase particular flight, - a third provisional rule composed of two events having for identifying the codes "126019, 127019" and being associated with an equipment t having for identifier the value "133" and the particular flight phase.

Each piece of equipment of a provisional rule is also associated with an occurrence coefficient, equal to zero for example. Note that each tentative rule can be saved automatically for future flights in the list of location rules 60 as a rule to be validated.

With reference to FIG. 2, during a maintenance step, the treatment computer 18 generates a maintenance list comprising all of the organs incriminated during the analysis P3, identification P4 and segmentation P5 steps. During a control and repair step P7, an operator acts technically on these incriminated organs to inspect them. If an offending member proves to be defective, the operator repairs or replaces it, and memorizes the fault of the offending member for its operation during a subsequent updating step. Indeed, during an update step P8, the operator can increment by one unit the occurrence coefficient associated with a particular organ incriminated having proved defective in the set of elected location rules and provisional rules referring to this particular body. Optionally, the maintenance list may include elected location rules and provisional rules resulting from the flight. Alternatively, these elected and provisional location rules can be stored in a memory so that the operator can easily determine the rules to be amended.

In addition, it is noted that an operator can determine whether a provisional rule should be deleted or kept in the list of location rules, possibly by validating it to transform it as a validated rule.

This operator can also delete or modify validated localization rules. Moreover, this operator can modify the diagnostic table according to the experience feedback. In another aspect, it is possible to update the list of flight phases 10 to incorporate the new flight phases appeared following the consolidation of two existing flight phases. This human processing loop involves the personal knowledge of the operator, but also the own knowledge available to this operator such as documentation 15. An operator can either validate each rule to validate that is submitted to him, or keep or modify the rule to validate, or revoke this rule to validate. The same is true of the provisional rules.

Claims (10)

REVENDICATIONS1. Technical maintenance method characterized in that one acts technically on at least one member (3) of an aircraft (1) following a flight, according to a collection list (22) listing events (23) occurring in flight and a list of location rules (60), each location rule (61) associating at least one member (3) to be checked and at least one event (23) and a flight phase (100), an occurrence coefficient (62) being associated with each member (3) of a location rule (61), during which:. during a census step (P1), a maintenance calculator (10) lists in flight events representing operating anomalies in a collection list (22) associating an event (23) as well as a dating element (25, 26) of this event and on the other hand flight parameters (101) determined using a flight recorder (102) during a grouping step (P2), said collection list is partitioned. (22) at least one group (30) of events according to at least one predefined temporal criterion, during an association step (P10), each event (23) being associated with a flight phase (100) determined using said flight parameters (101), a flight phase is defined for each group (30) (100) during which the events of the group occurred. . during an analysis step (P3), comparing the events and the flight phase of each group (30) with the events and the flight phase of each location rule (61), an organ that is cited in each rule locating unit (61) elected corresponding to a group (30) becoming an incriminated organ, each group (30) not corresponding to a location rule becoming an unidentified group (32),. during an identification step (P4), using a diagnostic table (50) associating an event (23) with at least one organ (3), a provisional rule is created and recorded for each unidentified group (32) associating the events and the flight phase of said unidentified group (32) with the corresponding members (3) established with the aid of said diagnostic table (50), the organs associated with this provisional rule becoming incriminated organs,. during a maintenance step (P6), a maintenance list (40) comprising at least one incriminated organ cited in at least one elected location rule and / or at least one temporary rule is established, then an operator acts technically on each organ incriminated to perform maintenance (P7). 2. Technical process according to claim 1, characterized in that, prior to the grouping step (P3), a filtering step (P9) is performed. 3. Technical process according to any one of claims 1 to 2, characterized in that during a segmentation step (P5), each unidentified group (32) is segmented into at least one alternative group (33), each alternative group (33) comprising the flight phase of the unidentified group from which it is derived and at least two events (23) associated with one and the same organ (4), then creating a provisional rule by alternating group (33), each organ cited in an alternating group (33) being an offending member. 4. Technical process according to any one of claims 1 to 3, characterized in that, said grouping steps (P2), association (P10), analysis (P3), identification (P4) and maintenance (P6) involve a processing calculator. 5. Technical process according to claim 4, characterized in that, during the segmentation step (P5), if at least two alternative groups (33 ', 33 ") given are constituted by the same events, these two alternative groups are concatenated. to form a single alternating group (231) associating said same events with the members of said at least two alternative groups. 6. Technical method according to any one of claims 1 to 5, characterized in that, during an update step (P8), a maintenance list comprising at least one offending member cited in at least one location rule elected and / or at least one provisional rule updated during said identification and segmentation steps, an operator increments the occurrence coefficient of each incriminated organ judged to be out of order in each elected location rule and each provisional rule citing this incriminated organ judged out of order. 7. Technical method according to any one of claims 1 to 6, characterized in that, during an update step (P8), an operator determines whether a provisional rule must be deleted or kept in the list of location rules as a rule to validate later or even as a validated rule. 8. Technical process according to any one of claims 1 to 7, characterized in that, said diagnostic table being a generic table, modifying said diagnostic table to adapt to the aircraft in question. 9. Technical method according to any one of claims 1 to 8, characterized in that, if a group (30) comprises at least a first event (23) associated with a dating element (25, 26) and a first flight phase and a second event (23) associated with a dating element (25, 26) and a second flight phase, said group includes a transition flight phase corresponding to the transition from the first flight phase to the second phase of flight. 10. Technical process according to any one of claims 1 to 9, characterized in that at least one localization rule (61) associates at least one member (3) to verify and at least one event (23) and a generic flight phase covering all flight phases (100). 1 1. Technical process according to any one of claims 1 to 10, characterized in that said flight parameters are to be selected from a list including: a rotational speed of at least one engine of an aircraft, a speed of rotation of a rotor of the aircraft, a torque generated by said engine, a position of at least one flight control, a geographical location of the aircraft, a pressure and a temperature outside the aircraft, a pressure and a temperature of a hydraulic fluid of a main gearbox driving said rotor, a temperature of the gases present in said engine, an altitude of the aircraft, at least one speed of movement in the space of the aircraft, at least one acceleration of the aircraft, a quantity and a flow of fuel, a mass of the aircraft. 12. System (7) for Automated Aircraft Maintenance Diagnostics (1), this automated maintenance diagnostic system (7) comprising an on-board electronic unit comprising an element (3), this system having at least one maintenance calculator (10) in flight listing events representing operating anomalies in a collection list (22), characterized in that said system implementing a method according to claim 1 as a function of a collection list ( 22) identifying events (23) occurring in flight, a dating element (25, 26) of this event and flight parameters (101) determined using a flight recorder (102), and a list of location rules (60), each location rule (61) associating at least one member (3) to be checked and at least one event (23) and a flight phase (100), a coefficient of occurrence (62) being associated with each member (3) of a e location (61), this system comprises a maintenance tool (16) provided with a memory (17) of a processing computer (18) comprising a calculation unit (19), said calculation unit (19) executing stored instructions for:. during a grouping step (P2), partitioning said collection list (22) into at least one group (30) of events according to at least one predefined temporal criterion, during an association step (P10), associating each event (23) with a flight phase (100), said flight phase (100) being determined by means of said flight parameters (101), eachgroup (30) having a flight phase (100) and at least one event (23) associated with a dating element (25, 26),. during an analysis step (P3), comparing the events and the flight phase of each group (30) with the events and the flight phase of each location rule (61), an organ cited in an elected location rule corresponding to a group (31) becoming an offending member stored in a maintenance list (40), each group (30) not corresponding to a location rule (61) becoming an unidentified group (32),. during an identification step (P4), using a stored diagnostic table (50) associating an event (23) with at least one organ (3), creating and registering a provisional rule for each unidentified group (32) in associating the events (23) and the flight phase of said unidentified group (32) with the corresponding members (3) established using said diagnostic table (60), the organs associated with this provisional rule becoming incriminated organs, . during a maintenance step (P6), to establish a maintenance list (40) comprising at least one incriminated organ cited in at least one elected location rule and / or at least one provisional rule so that an operator acts technically on each organ incriminated to carry out its maintenance (P7). 13. System according to claim 12, characterized in that said maintenance diagnostic system comprises an operator operable control means for updating the provisional rules and the localization rules. 14.Aeronef (1) of the type provided for carrying out the method according to one of claims 1 to 11, characterized in that this aircraft (1) is rotary wing.