Method and system for diagnosing faults on an aircraft

Abstract

A method and system for diagnosing faults on an aircraft are provided, the aircraft including aircraft systems configured to report faults to an on-board inference engine. The method includes receiving, at an on-board computer of the aircraft, a fault report from one of the aircraft systems in the aircraft system, the fault report indicating a failed test reported by the aircraft system. The on-board inference engine accesses an on-board diagnostic causal model, which is represented by a graph describing known causal relationships between possible failed tests reported by corresponding aircraft systems in the aircraft system and possible fault modes of the corresponding aircraft systems in the aircraft system. The on-board inference engine diagnoses a fault mode of one of the aircraft systems or another aircraft system in the aircraft system based on the failed test and using graph theory algorithms and the on-board diagnostic causal model. A maintenance action for the fault mode is determined and a maintenance message including the maintenance action is generated.

Description

Technical Field

[0001] The present disclosure generally relates to aircraft maintenance and, in particular, to on-board diagnosis of aircraft system failure modes and related aircraft maintenance. Background Art

[0002] Complex systems (such as machines including vehicles such as aircraft, spacecraft, ships, motor vehicles, and rail vehicles) typically include some type of performance monitoring system that records data regarding the performance of the machine, which includes the performance of various systems (and subsystems) of the machine. The data includes records of certain performance events that occur during operation of the machine. The performance monitoring system generally collects data and reports all the collected data to a user. The user can then utilize the data to determine the type of maintenance or repair (if any) that the machine may require. For example, if the data indicates that a particular mechanical or electromechanical system of the machine has failed or that the performance of one or more mechanical or electromechanical systems may lead to a future machine failure, the user can perform an appropriate repair on the machine at the next opportunity.

[0003] Although current systems for machine performance and fault monitoring provide the user with the necessary data to make appropriate repair decisions, it is still necessary for the user to sort through all the data to determine the most appropriate repair action to address the failure mode. Thus, the user must classify and interpret the data based on the user's knowledge of the particular machine. This can be time-consuming and does not always result in the repair action being performed first being the most appropriate repair action, especially for complex machines such as aircraft and other vehicles. For many types of machines, especially commercial vehicles, the amount of time the vehicle is out of service is expensive for the vehicle owner. Thus, the longer it takes to perform the most appropriate repair action for a given failure mode, the longer the vehicle will be out of service, which can be expensive for the vehicle owner if the vehicle would otherwise be in service.

[0004] Accordingly, there is a desire for a system and method that addresses at least some of the above problems and other possible problems. Summary of the Invention

[0005] Example implementations of the present disclosure relate to improved techniques for on-board diagnosis and correlation of fault data with maintenance actions and closed-loop diagnostic model shaping for complex systems. Example implementations provide an on-board inference engine on an aircraft and a process for diagnosing faults on the aircraft. Other example implementations provide a non-on-board inference engine and a process for closed-loop diagnostic model shaping to maintain the on-board inference engine for diagnosing faults on the aircraft.

[0006] Accordingly, the present disclosure includes, but is not limited to, the following example implementations.

[0007] Some example implementations provide a method for diagnosing faults on an aircraft that includes aircraft systems configured to report faults to an on-board inference engine. The method includes: receiving, at an on-board computer of the aircraft that includes the on-board inference engine, a fault report from one of the aircraft systems, the fault report indicating a failed test reported by the aircraft system; accessing, by the on-board inference engine, an on-board diagnostic causal model, the on-board diagnostic causal model being represented by a graph that describes known causal relationships between possible failed tests reported by corresponding ones of the aircraft systems and possible fault modes of corresponding ones of the aircraft systems; diagnosing, by the on-board inference engine, a fault mode of one of the aircraft systems or another aircraft system based on the failed test and using a graph theory algorithm and the on-board diagnostic causal model; determining a maintenance action for the fault mode; and generating, by the on-board computer, a maintenance message that includes at least the maintenance action.

[0008] In some example implementations of the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the on-board diagnostic causal model is represented by a graph that includes nodes connected by edges, the nodes representing possible failed tests and possible fault modes, and the edges indicating known causal relationships between the possible failed tests and the possible fault modes.

[0009] In some example implementations of the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the on-board diagnostic causal model is represented by a graph that is a collection of graphs for corresponding ones of the aircraft systems, the graphs reflecting fault propagation behavior within corresponding ones of the aircraft systems, and the collection reflecting fault propagation behavior across connected aircraft systems of the aircraft system.

[0010] In some example implementations of the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the graph theory algorithm is an Optimal Solution Set (OSS) algorithm.

[0011] In some example implementations of the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the method further includes sending the maintenance message to a display device on the aircraft or a display device on a maintenance component, the display device being configured to establish a connection with the on-board computer to receive the maintenance message.

[0012] In some example implementations of the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the display of the maintenance message references instructions for performing the maintenance action to resolve the fault mode diagnosed by the on-board inference engine.

[0013] In some example implementations of the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the fault report is one of a plurality of fault reports of failed tests reported by a corresponding aircraft system in an aircraft system, the fault mode is one of a plurality of diagnosed fault modes of at least some of the aircraft systems in the aircraft system that caused the failed test, and the method further includes: accessing diagnostic data received from an on-board computer, the diagnostic data including a plurality of fault reports and a plurality of diagnosed fault modes; accessing, by a non-on-board inference engine, a non-on-board diagnostic causal model that describes the causal relationship between the failed test and the plurality of diagnosed fault modes, the non-on-board diagnostic causal model being constructed using a graph theory machine learning algorithm trained using historical diagnostic data; comparing the diagnostic data with the non-on-board diagnostic causal model; and based on the comparison, determining a new causal relationship in the causal relationships described by the non-on-board diagnostic causal model, the new causal relationship being new relative to known causal relationships; and updating the on-board diagnostic causal model to further describe the new causal relationship, including generating an updated on-board diagnostic causal model and uploading the updated on-board diagnostic causal model to the on-board computer using a loadable software aircraft component upload toolset.

[0014] In some example implementations of the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the diagnostic data further includes a maintenance message having a maintenance action determined by the on-board inference engine for a corresponding diagnosed fault mode of the plurality of diagnosed fault modes, and the on-board diagnostic causal model further describes the relationship between possible fault modes and corresponding maintenance actions among the maintenance actions, and wherein, for a specific maintenance message having a specific maintenance action for a corresponding diagnosed fault mode, the method further includes: accessing a maintenance record having the performed maintenance action for the corresponding diagnosed fault mode and thus having a new maintenance action; and identifying a maintenance action difference between the specific maintenance action and the performed (new) maintenance action to determine a new relationship between the corresponding diagnosed fault mode and the performed maintenance action, wherein updating the on-board diagnostic causal model further includes updating the on-board diagnostic causal model to describe the new relationship between the corresponding diagnosed fault mode and the performed (new) maintenance action.

[0015] In some example implementations of the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the graph theory machine learning algorithm is a graph theory algorithm for finding maximal cliques.

[0016] Some example implementations provide an on-board computer for diagnosing faults on an aircraft, the aircraft including aircraft systems configured to report faults to an on-board inference engine, the on-board computer including: a memory configured to store computer-readable program code including the on-board inference engine; and a processing circuit configured to access the memory and execute the computer-readable program code to cause the device to at least perform the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations.

[0017] Some example implementations provide a system for diagnosing faults on an aircraft, the aircraft including aircraft systems configured to report faults to an on-board inference engine, the system including: an on-board computer including an on-board inference engine, the on-board computer being configured to perform the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations.

[0018] Some example implementations provide a method of maintaining an on-board inference engine to diagnose faults on an aircraft, the aircraft including aircraft systems configured to report faults to an on-board inference engine, the method including: accessing diagnostic data received from an on-board computer of the aircraft including the on-board inference engine, the diagnostic data including a plurality of fault reports of failed tests reported by respective aircraft systems in the aircraft systems and a plurality of diagnosed fault modes of at least some of the aircraft systems in the aircraft systems that caused the failed tests; constructing a non-on-board diagnostic causal model using a non-on-board inference engine, the non-on-board diagnostic causal model describing a causal relationship between the failed tests and the plurality of diagnosed fault modes, the non-on-board diagnostic causal model being constructed using a graph theory machine learning algorithm trained using historical diagnostic data; comparing the diagnostic data with the non-on-board diagnostic causal model; and based on the comparison, determining a new causal relationship in the causal relationships described by the non-on-board diagnostic causal model, the new causal relationship being new relative to known causal relationships; and updating the on-board diagnostic causal model to further describe the new causal relationship, including generating an updated on-board diagnostic causal model and uploading the updated on-board diagnostic causal model to the on-board computer using a loadable software aircraft component upload toolset.

[0019] In some example implementations of the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the non-on-board diagnostic causal model is represented by a graph including nodes connected by edges, the nodes representing the failed tests and the plurality of diagnosed fault modes, and the edges indicating known causal relationships between the failed tests and the plurality of diagnosed fault modes.

[0020] In some example implementations of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the comparison of diagnostic data with a non-airborne diagnostic causal model also includes a comparison with actual fault data, diagnostic data, and maintenance action data collected from an airborne aircraft system that reflects actual causal relationships.

[0021] In some example implementations of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the graph theory machine learning algorithm is a graph theory algorithm for finding maximum cliques.

[0022] In some example implementations of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the plurality of fault reports includes fault reports indicating those failed tests caused by a fault mode diagnosed by one of the plurality of diagnosed fault modes, and the method further includes: diagnosing, by a non-airborne inference engine, a corresponding fault mode of the aircraft system based on those failed tests caused by the diagnosed fault mode and using the graph theory machine learning algorithm and a graph; and reporting, based on diagnostic data received from an airborne computer, any differences between the corresponding fault mode diagnosed by the non-airborne inference engine and a fault mode diagnosed by one of the plurality of diagnosed fault modes.

[0023] In some example implementations of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the diagnostic data further includes maintenance messages having maintenance actions determined by an airborne inference engine for corresponding diagnosed fault modes of the plurality of diagnosed fault modes, and the airborne diagnostic causal model also describes the relationships between possible fault modes and corresponding maintenance actions among the maintenance actions, and wherein, for a particular maintenance message having a particular maintenance action for a corresponding diagnosed fault mode, the method further includes: accessing a maintenance record having the performed maintenance action for the corresponding diagnosed fault mode and thereby having a new maintenance action; and identifying a maintenance action difference between the particular maintenance action and the performed maintenance action to determine a new relationship between the corresponding diagnosed fault mode and the performed maintenance action, wherein updating the airborne diagnostic causal model further includes updating the airborne diagnostic causal model to describe the new relationship between the corresponding diagnosed fault mode and the performed maintenance action.

[0024] In some example implementations of the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the method further includes receiving, at an on-board computer of the aircraft that includes an on-board inference engine, a fault report from one of the aircraft systems in the aircraft system, the fault report indicating a failed test reported by the aircraft system; accessing, by the on-board inference engine, an updated on-board diagnostic causal model; diagnosing, by the on-board inference engine, a fault mode of one of the aircraft systems or another aircraft system in the aircraft system based on the failed test and using a graph theory algorithm and the updated on-board diagnostic causal model; determining a maintenance action for the fault mode; and generating, by the on-board computer, a maintenance message that includes at least the maintenance action.

[0025] In some example implementations of the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations, the graph theory algorithm is an optimal solution set (OSS) algorithm.

[0026] Some example implementations provide a non-on-board computer for maintaining an on-board inference engine to diagnose faults on an aircraft, the aircraft including an aircraft system configured to report faults to the on-board inference engine, the non-on-board computer including: a memory configured to store computer-readable program code including a non-on-board inference engine; and a processing circuit configured to access the memory and execute the computer-readable program code to cause the non-on-board computer to perform at least the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations.

[0027] Some example implementations provide a system for maintaining an on-board inference engine to diagnose faults on an aircraft, the aircraft including an aircraft system configured to report faults to the on-board inference engine, the system including an on-board computer and a non-on-board computer, the on-board computer including an on-board inference engine, the non-on-board computer including a non-on-board inference engine, and system components configured to perform the method of any of the foregoing example implementations or any combination of any of the foregoing example implementations.

[0028] These and other features, aspects, and advantages of the present disclosure will be apparent from the following detailed description and the accompanying drawings of the following brief description. The present disclosure includes any combination of two, three, four, or more features or elements set forth in the present disclosure, regardless of whether such features or elements are explicitly combined or otherwise recited in the specific example implementations described herein. The present disclosure is intended to be read as a whole such that any separable feature or element of the present disclosure should be considered combinable in any aspect and example implementation thereof, unless the context of the present disclosure clearly indicates otherwise.

[0029] Accordingly, it should be understood that the present disclosure of the invention is provided only for the purpose of summarizing some example implementations in order to provide a basic understanding of some aspects of the present disclosure. Accordingly, it should be understood that the above example implementations are merely examples and should not be construed as narrowing the scope or spirit of the present disclosure in any way. Other example implementations, aspects, and advantages will become apparent from the following detailed description in conjunction with the accompanying drawings, which illustrate by way of example the principles of some of the described example implementations. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Accordingly, having generally described example implementations of the present disclosure, reference will now be made to the accompanying drawings, which are not necessarily drawn to scale and in which:

[0031] Figure 1 An aircraft is shown in accordance with some example implementations of the present disclosure;

[0032] Figure 2A A system for aircraft maintenance is shown in accordance with example implementations of the present disclosure;

[0033] Figure 2B A block diagram of an on-board computer having an on-board inference engine is shown in accordance with example implementations of the present disclosure, the on-board inference engine being functionally connected to a non-on-board computer having a non-on-board inference engine;

[0034] Figure 3 A graphical representation of an on-board diagnostic causal model is shown in accordance with example implementations of the present disclosure;

[0035] Figure 4 A subgraph representing a maximum clique is shown in accordance with example implementations;

[0036] Figure 5A 、 Figure 5B and Figure 5C are flowcharts showing the respective steps in a method for diagnosing a fault on an aircraft in accordance with example implementations, the aircraft including aircraft systems configured to report faults to an on-board inference engine;

[0037] Figure 6A 、 Figure 6B and Figure 6C are flowcharts showing the respective steps in a method for maintaining an on-board inference engine to diagnose a fault on an aircraft in accordance with example implementations, the aircraft including aircraft systems configured to report faults to an on-board inference engine; and

[0038] Figure 7 An apparatus is shown in accordance with some example implementations. DETAILED DESCRIPTION

[0039] Some embodiments of the present disclosure will now be described more fully with reference to the accompanying drawings, in which some, but not all embodiments of the disclosure are shown. In fact, the various embodiments of the present disclosure may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these example implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. For example, unless otherwise specified, something referred to as first, second, etc. should not be construed as implying a particular order. Additionally, something described as being on top of something else (unless otherwise specified) may instead be below, and vice versa; and similarly, something described as being to the left of something else may instead be to the right, and vice versa. Throughout the text, the same reference numerals refer to the same elements.

[0040] Example implementations of the present disclosure relate to improved techniques for on-board diagnosis and correlation of fault data with maintenance actions and for forming closed-loop diagnostic models of complex systems. Example implementations provide an on-board inference engine on an aircraft and a process for diagnosing faults on the aircraft. Additionally or alternatively, some example implementations provide a non-on-board inference engine and a process for forming a closed-loop diagnostic model to maintain the on-board inference engine for diagnosing faults on the aircraft.

[0041] Figure 1 An example of an aircraft 100 that may benefit from the example implementations of the present disclosure is shown. As shown, the aircraft includes an airframe 102 having a fuselage 104, wings 106, and a tail 108. The aircraft also includes a plurality of high-level systems 110 such as a propulsion system. In Figure 1 the specific example shown, the propulsion system includes two wing-mounted engines 112. In other embodiments, the propulsion system may include other arrangements, such as engines carried by other parts of the aircraft including the fuselage and / or the tail. The high-level systems may also include an electrical system 114, a hydraulic system 116, and / or an environmental system 118. Any number of other systems may be included.

[0042] The above-mentioned advanced system may include multiple sensors and subsystems that provide fault and sensor data, which is transmitted to the Aircraft Condition Monitoring System (ACMS) via the aircraft data communication bus network and / or the Onboard Network System (ONS). The ACMS can collect, monitor, record, and report real-time aircraft system data, which may include error messages from the Flight Deck Effects (FDE) system, system test reports, fault reports, and other information. In addition to many other aircraft performance functions, the data collected by the ACMS is used, for example, to perform cabin pressure and temperature monitoring, hard landing detection, crew monitoring, and engine monitoring. Then, the received data is used to analyze aircraft performance, record significant flight events, report aircraft system test reports and fault reports, and troubleshoot faults.

[0043] The ACMS can communicate with the Onboard Component / Computer 120, which may also be referred to as the Central Maintenance Computer (CMC), and the Aircraft Health Management or Maintenance Management System and Diagnostic Maintenance Computing Function (DMCF) may reside on the Central Maintenance Computer. The Onboard Computer 120 including the DMCF can receive aircraft system test reports and fault reports, and may also include an onboard diagnostic model. The DMCF can provide data acquisition for the onboard diagnostic model, which receives test report and fault report data.

[0044] Figure 2A A system 200 for maintaining an aircraft 202 that may correspond to an aircraft 100 is shown according to some example implementations. As described in more detail below, the system may include any one of a plurality of different subsystems for performing one or more functions or operations (each subsystem being a separate system). The subsystems may be co-located or directly coupled to each other, or in some examples, different subsystems among the subsystems may communicate with each other via one or more computer networks 204. Additionally, although shown as part of the system, it should be understood that any one or more of the subsystems may act or operate as a separate system, regardless of any of the other subsystems. It should also be understood that the system may include one or more additional or optional subsystems in addition to Figure 2A the subsystems shown.

[0045] As shown, in some examples, the system 200 includes an Onboard Computer 206 that may correspond to the Onboard Component / Computer 120. The Onboard Computer includes an Onboard Inference Engine 208, and in some examples, the Onboard Computer is configured to diagnose faults on the aircraft including aircraft systems configured to report faults to the Onboard Inference Engine. Examples of suitable aircraft systems include the propulsion system 110, the electrical system 114, the hydraulic system 116, and / or the environmental system 118 at least when the aircraft 202 corresponds to the aircraft 100. Any number of other systems may be included.

[0046] In some examples, the on-board computer 206 is configured to receive a fault report from one of the aircraft systems in the aircraft system, the fault report indicating a failed test reported by the aircraft system. The on-board computer is configured to access an on-board diagnostic causal model through the on-board inference engine 208, which can be represented by a graph describing the known causal relationships between possible failed tests reported by corresponding aircraft systems in the aircraft system and possible fault modes of the corresponding aircraft systems in the aircraft system. In some more specific examples, the on-board diagnostic causal model is represented by a graph including nodes connected by edges, where the nodes represent possible failed tests and possible fault modes, and the edges indicate the known causal relationships between the possible failed tests and the possible fault modes. Additionally or alternatively, in some examples, the on-board diagnostic causal model is represented by a graph that is a collection of graphs for corresponding aircraft systems in the aircraft system, where the graphs reflect the fault propagation behavior within the corresponding aircraft systems in the aircraft system, and the collection reflects the fault propagation behavior across the connected aircraft systems in the aircraft system.

[0047] Referring Figure 2B , the on-board computer includes an on-board inference engine 208, and the on-board computer is configured to use the on-board inference engine 208 to diagnose faults on the aircraft, which includes an aircraft system configured to report faults to the on-board inference engine. The on-board computer 206 is configured to have the on-board inference engine 208 diagnose the fault mode of one aircraft system or another in the aircraft system based on the failed test and using the on-board diagnostic causal model 210 and graph theory algorithms (such as the Optimal Solution Set (OSS) algorithm). The on-board computer is also configured to generate a fault report associated with the diagnosed fault mode, and based on this fault report, a maintenance message indicating a maintenance action can be generated. As Figure 2BAs shown, the non-airborne computer 218 is configured to maintain the on-board inference engine 208. In some examples, the non-airborne computer is configured to access diagnostic data received directly or indirectly from the on-board computer. The diagnostic data includes a plurality of fault reports of failed tests reported by corresponding aircraft systems in the aircraft system and a plurality of diagnosed fault modes of at least some of the aircraft systems in the aircraft system that caused the failed tests. The non-airborne computer 218 includes a non-airborne inference engine 224, which is configured to determine the causal relationship between the failed tests and the plurality of diagnosed fault modes. The non-airborne diagnostic causal model utilizes a machine learning algorithm trained using historical diagnostic data, which can be a machine learning algorithm, such as a graph theory algorithm for finding maximum cliques, such as the Bron-Kerbosch algorithm. The machine learning algorithm for finding maximum cliques uses diagnostic data to determine the known causal relationship between fault modes and failed tests. The diagnostic data includes historical data relating maintenance actions to maintenance messages or fault modes (obtained from a database of actual aircraft diagnostic data, which includes different fault modes; the relationships between test reports, fault reports, and components; maintenance action data, etc.). The non-airborne inference engine employing the machine learning algorithm is configured to use actual fault data, diagnostic data, and / or maintenance action data received from the on-board aircraft systems (from which the relationship between fault modes and failed tests is determined), and the actual fault data, diagnostic data, and / or maintenance action data are compared with the diagnostic data accessible from the aircraft on-board diagnostic system (on-board computer 206). The non-airborne inference engine can determine the differences based on this comparison, and the differences can then be used to update and improve the non-airborne diagnostic causal model. The identification of differences can include identifying new maintenance actions based on historical diagnostic data and maintenance action data collected from the aircraft on-board systems, where the difference identification identifies new maintenance actions associated with fault modes.

[0048] Figure 3FIG. 300 shows an example graph-based representation of an on-board diagnostic causal model suitable as the on-board diagnostic causal model 210 according to some example implementations. As shown, a fault report 302 from an anti-skid wheel speed transducer 304 of an aircraft braking system can indicate failed landing gear retraction brake tests 306A, 306B. Also shown, the on-board diagnostic causal model can include nodes 308A through 308C connected by edges 310. The nodes can include nodes 308A, 308B representing the failed landing gear retraction brake tests and a node 308C representing a fault mode having a known causal relationship with the failed tests, as indicated by the edges connecting the nodes. In this regard, the failed tests share a common fault mode of misoperation of the anti-skid wheel speed transducer of the aircraft braking system. Then, the fault mode can be diagnosed based on the failed tests and using the on-board causal model. Maintenance actions can be implemented for the fault mode, and a maintenance message 312 can be generated that references instructions 314 for performing the maintenance actions. Such maintenance action data can be collected from an aircraft condition monitoring system or other systems as historical maintenance action data and / or diagnostic data, where actual historical diagnostic data and maintenance action data can be used to determine actual causal relationships.

[0049] Return Figure 2A , the on-board computer 206 is configured to diagnose a fault mode in one or another aircraft system of the aircraft system by the on-board inference engine 208 based on the failed tests and using a graph theory algorithm and the on-board diagnostic causal model 210. An example of a suitable graph theory algorithm is the Optimal Solution Set (OSS) algorithm. The OSS algorithm can utilize submodular minimization of data to provide an efficient distributed optimization algorithm, where data regarding various failed test reports can be associated with elements or categories of the test reports or fault modes, and only data contained in subsets that include a specific failed test report or the fault mode itself is considered, while all other data is ignored. The optimization algorithm can also create a graph of nodes for graph theory analysis and can use adjacent node and / or edge probability analysis to determine the diagnosed fault mode, which will describe the diagnostic causal relationship between the failed tests reported by the aircraft and the possible fault modes of the reported aircraft system. The on-board computer is also configured to determine maintenance actions for the fault mode and generate a maintenance message 212 that includes at least the maintenance actions.

[0050] In some examples, the on-board computer 206 is also configured to send the maintenance message 212 to a display device 214 on the aircraft or a display device on a maintenance component, the display device being configured to establish a connection with the on-board computer to receive and display the maintenance message 212. In some of these examples, the display of the maintenance message references instructions 216 for performing maintenance actions to resolve the fault mode diagnosed by the on-board inference engine 208.

[0051] Also as Figure 2A shown, in some examples, system 200 includes a non-airborne computer 218 configured to maintain an on-board inference engine 208. In some examples, the non-airborne computer is configured to access diagnostic data received directly or indirectly from the on-board computer. The diagnostic data includes a plurality of fault reports of failed tests reported by corresponding aircraft systems in the aircraft system and a plurality of diagnosed fault modes of at least some of the aircraft systems in the aircraft system that caused the failed tests.

[0052] As shown, in some examples, system 200 includes at least one data source 220. In some examples, the source includes a memory that may be located in a single source or distributed across multiple sources. Data may be stored in a variety of different ways (such as in a database or flat file in any of a variety of different types or formats). In some of these examples, an aircraft condition monitoring system (ACMS) on aircraft 202 may collect, monitor, record, and report diagnostic data. At least some of the diagnostic data may be accessed from reports generated by the ACMS and may be wirelessly transmitted directly or via a satellite 222 or network 204 to a particular data source shown and sometimes referred to as an aircraft health management (AHM) system 220A. In other examples of these examples, the diagnostic data may be transmitted via a wired connection or a portable data storage device (e.g., flash memory, thumb drive).

[0053] In some examples, non-airborne computer 218 includes a non-airborne inference engine 224. The non-airborne computer is configured to use the non-airborne inference engine to construct a non-airborne diagnostic causal model 226, which may describe the causal relationship between a failed test and a plurality of diagnosed fault modes. Here, the non-airborne diagnostic causal model is constructed using a graph theory machine learning algorithm trained using historical diagnostic data. And in some examples, the non-airborne diagnostic causal model is represented by a graph including nodes connected by edges, where the nodes represent failed tests and a plurality of diagnosed fault modes, and the edges indicate known causal relationships between the failed tests and the plurality of diagnosed fault modes. The non-airborne diagnostic causal model 226 employs a machine learning algorithm that uses fault data, historical diagnostic data, and / or maintenance action data received from on-board aircraft systems, and the fault data, historical diagnostic data, and / or maintenance action data are compared with diagnostic data accessible from the on-board diagnostic system (on-board computer 206) of the aircraft. The non-airborne diagnostic causal model 226 may determine differences based on its comparison with the diagnostic data. In Figure 3In an example airborne diagnostic causal model, a graph-based representation of a fault report 302 from an aircraft braking system anti-skid wheel speed transducer 304 indicates failed landing gear retraction brake tests 306A, 306B. The non-airborne diagnostic causal model 226 can identify discrepancies based on the collected maintenance action data (such as a wheel speed transducer maintenance instruction (MM_32_30710)) to define new relationships in the diagnostic causal model related to the fault to be incorporated into the aircraft's airborne diagnostic inference engine model.

[0054] An example of a suitable graph theory machine learning algorithm is a graph theory algorithm for finding a maximum clique, such as the Bron-Kerbosch algorithm. The graph theory machine learning algorithm for finding a maximum clique can use diagnostic data to determine known causal relationships between fault patterns and failed tests. Figure 4 Diagnostic data is shown, and the relationships can be transformed into a graph of a maximum clique 400 with nodes 402, 404 and edges 406 as described below, each of which is called up in the accompanying drawings. Here, the nodes 402 represent possible failed tests T1, T2, T3 and T4, and the nodes 404 represent possible fault patterns F2, F4, F5, F8 and F10. The edges indicate known causal relationships between possible failed tests and possible fault patterns.

[0055] The maximum clique 400 can be determined from the graph based on the diagnostic data. In this example, the maximum clique 400 includes all the nodes 402, 404 and edges 406 in the graph. In other examples, the maximum clique can include fewer nodes and edges than all the nodes and edges. From the maximum clique, specific fault patterns associated with the failed test can be isolated.

[0056] For example, the edges 406 can be examined to find the active fault patterns 404 associated with the failed tests reported at the nodes 402. If the clique has more than one node for the failed test, then the non-airborne inference engine 224 can determine the clique as a maximum clique (e.g., the maximum clique 400). Additionally, if all the nodes for the failed test have one or more shared fault patterns determined according to the edges connecting the nodes, the maximum clique can be identified as a monochromatic clique. The shared fault pattern becomes the fault pattern of interest, and the monochromatic clique can then be used to identify the faulty component associated with the fault pattern of interest. In Figure 4 the maximum clique 400 is a monochromatic clique with the shared fault pattern F8, as identified from the edge 406 connecting the node 402 representing the failed test to the node 404 representing the shared fault pattern F8.

[0057] Returning again to Figure 2A, the non-airborne diagnostic causal model 226 can be shaped by discovering any new relationships between failure modes and failed tests that are the result of emergency behaviors not present in the airborne diagnostic causal model 210 or any aircraft system. This can be mined from historical data related to maintenance actions and maintenance messages 212 or failure modes, and then the maintenance messages 212 or failure modes can be used to update and improve the non-airborne diagnostic causal model.

[0058] The non-airborne diagnostic causal model 226 can have a specific signature that defines the expected propagation of failure modes. In operation, the non-airborne inference engine 224 can record the portions of the signature it encounters while performing diagnostics. The non-airborne diagnostic causal model can be improved from this historical data (using the historical signature), which can be compared to a previous non-airborne diagnostic causal model, such that any new relationships can be considered for addition to the updated non-airborne diagnostic causal model.

[0059] The non-airborne computer 218 is configured to compare diagnostic data from the airborne computer 206 with the non-airborne diagnostic causal model 226 to diagnose failure modes of aircraft systems. In this regard, the diagnostic data can describe known causal relationships between possible failed tests reported by corresponding aircraft systems in the aircraft system and possible failure modes of the corresponding aircraft systems in the aircraft system.

[0060] Based on the above, in these examples, the non-airborne computer 218 is configured to determine new causal relationships in the causal relationships described by the non-airborne diagnostic causal model 226, which are new relative to the known causal relationships. The non-airborne computer is also configured to update the airborne diagnostic causal model to further describe the new causal relationships. This includes the non-airborne computer being configured to generate an updated airborne diagnostic causal model and upload the updated airborne diagnostic causal model as a loadable software aircraft part (LSAP) to the airborne computer 206 via the LSAP upload toolset 228.

[0061] In some examples, multiple fault reports include fault reports indicating those failed tests caused by one of the multiple diagnosed failure modes. In some of these examples, the non-airborne computer 218 is also configured to diagnose the corresponding failure modes of the aircraft system by the non-airborne inference engine 224 based on those failed tests caused by the diagnosed failure modes and using graph theory machine learning algorithms and graphs. The non-airborne computer is then also configured to report any differences between the corresponding failure modes diagnosed by the non-airborne inference engine and one of the multiple diagnosed failure modes based on the diagnostic data received from the airborne computer 206.

[0062] In some examples, the diagnostic data accessed by the non-airborne computer 218 also includes maintenance messages 212 having maintenance actions for respective diagnosed fault modes determined by the on-board inference engine 208 for a plurality of diagnosed fault modes. In some of these examples, the on-board diagnostic causal model 210 also describes the relationship between possible fault modes and respective maintenance actions among the maintenance actions. Then, for a particular maintenance message having a particular maintenance action for a respective diagnosed fault mode, the non-airborne computer 218 is also configured to access a maintenance record having the performed maintenance action for the respective diagnosed fault mode and thus having a new maintenance action. The non-airborne computer is also configured to identify a maintenance action difference between the particular maintenance action and the performed maintenance action to determine a new relationship between the respective diagnosed fault mode and the performed maintenance action. The new relationship can be a new maintenance action identified from historical diagnostic data and maintenance action data collected from the aircraft, where the new maintenance action is associated with the fault mode. And the non-airborne computer is configured to update the on-board diagnostic causal model to describe the new relationship between the respective diagnosed fault mode and the performed maintenance action. For example, the updated on-board diagnostic causal model can also include a new maintenance action identified from the collected historical maintenance action data, which is associated with the fault mode.

[0063] Figure 5A , Figure 5B and Figure 5C are flowcharts showing the respective steps in a method 500 for diagnosing faults on an aircraft 202 according to an example implementation of the present disclosure, the aircraft including aircraft systems configured to report faults to an on-board inference engine 208.

[0064] As Figure 5A shown, at block 502, the method 500 includes receiving, at an on-board computer 206 of the aircraft 202 including the on-board inference engine 208, a fault report from one of the aircraft systems, the fault report indicating a failed test reported by the aircraft system. As shown at block 504, the method includes the on-board inference engine accessing an on-board diagnostic causal model 210, the on-board diagnostic causal model 210 being represented by a graph describing known causal relationships between possible failed tests reported by respective aircraft systems in the aircraft systems and possible fault modes of the respective aircraft systems in the aircraft systems.

[0065] As shown in block 506, the method includes diagnosing a fault mode of one or another aircraft system in an aircraft system by an on-board inference engine 208 based on a failed test and using a graph theory algorithm (e.g., the OSS algorithm) and an on-board diagnostic causal model 210. As shown in block 508, the method includes determining a maintenance action for the fault mode. And as shown in block 510, the method includes generating, by an on-board computer 206, a maintenance message 212 that includes at least the maintenance action.

[0066] In some examples, the method 500 further includes sending the maintenance message 212 to a display device 214 on the aircraft 202 or a display device on a maintenance component, the display device being configured to establish a connection with the on-board computer to receive the maintenance message, as shown in block 512.

[0067] As Figure 5B shown, in some examples, the fault report is one of a plurality of fault reports of failed tests reported by a corresponding aircraft system in the aircraft system, and the fault mode is one of a plurality of diagnosed fault modes in at least some of the aircraft systems that caused the failed test. As shown in block 514, the method 500 further includes accessing diagnostic data received from the on-board computer 206, and the diagnostic data includes a plurality of fault reports and a plurality of diagnosed fault modes.

[0068] As shown in block 516, the method 500 further includes accessing, by a non-on-board inference engine 224, a non-on-board diagnostic causal model 226 that describes a causal relationship between the failed test and the plurality of diagnosed fault modes. The non-on-board diagnostic causal model is constructed using a graph theory machine learning algorithm trained using historical diagnostic data. The method further includes comparing the diagnostic data with the non-on-board diagnostic causal model, as shown in block 518.

[0069] Based on the comparison at block 518, the method 500 further includes identifying a difference between the diagnostic data and the non-on-board diagnostic causal model 226 to determine a new causal relationship in the causal relationships described by the non-on-board diagnostic causal model 226, the new causal relationship being new relative to known causal relationships, as shown in block 520. As shown in block 522, the method further includes updating the on-board diagnostic causal model to further describe the new causal relationship, including generating an updated on-board diagnostic causal model and uploading the updated on-board diagnostic causal model to an on-board component using the LSAP upload toolset 228.

[0070] As Figure 5CAs shown, in some examples, the diagnostic data further includes a maintenance message 212 having a maintenance action for a respective diagnosed fault mode of a plurality of diagnosed fault modes determined by the on-board inference engine 208, and the on-board diagnostic causal model 210 also describes the relationship between the possible fault modes and the respective maintenance actions among the maintenance actions. For a particular maintenance message having a particular maintenance action for a respective diagnosed fault mode, the method 500 further includes accessing a maintenance record having the performed maintenance action for the respective diagnosed fault mode and thus having a new maintenance action, as shown in block 524. As shown in block 526, the method further includes identifying a maintenance action difference between the particular maintenance action and the performed maintenance action to determine a new relationship between the respective diagnosed fault mode and the performed maintenance action. The step of identifying the maintenance action difference may further include identifying the new maintenance action from historical diagnostic data and maintenance action data collected from the aircraft condition monitoring system, wherein the difference identification identifies the new maintenance action associated with the fault mode. Then, updating the on-board diagnostic causal model further includes updating the on-board diagnostic causal model to describe the new relationship between the respective diagnosed fault mode and the performed maintenance action, as shown in block 522A.

[0071] Figure 6A 、 Figure 6B and Figure 6C are flowcharts showing the respective steps in a method 600 of maintaining an on-board inference engine 208 to diagnose faults on an aircraft 202, the aircraft including aircraft systems configured to report faults to the on-board inference engine. As Figure 6A shown, at block 602, the method includes accessing diagnostic data received from an on-board computer 206 of the aircraft that includes the on-board inference engine, the diagnostic data including a plurality of fault reports of failed tests reported by respective aircraft systems in the aircraft system and a plurality of diagnosed fault modes of at least some of the aircraft systems in the aircraft system that caused the failed tests.

[0072] As shown in block 604, the method 600 includes constructing a non-on-board diagnostic causal model 226 using a non-on-board inference engine 224, the non-on-board diagnostic causal model describing the causal relationship between the failed tests and the plurality of diagnosed fault modes. The non-on-board diagnostic causal model is constructed using a graph theory machine learning algorithm trained using historical diagnostic data.

[0073] As shown in block 606, method 600 includes comparing diagnostic data with a non-airborne diagnostic causal model 226. Based on the comparison at block 606, method 600 includes determining a new causal relationship in the causal relationships described by the non-airborne diagnostic causal model 226, the new causal relationship being new relative to known causal relationships, as shown in block 608. As shown in block 610, the method includes updating the airborne diagnostic causal model to further describe the new causal relationship, including generating an updated airborne diagnostic causal model, and uploading the updated airborne diagnostic causal model to the airborne computer using the LSAP upload toolset 228.

[0074] In some examples, the plurality of failure reports includes failure reports indicating those failed tests diagnosed by one of the plurality of diagnosed failure modes, and as Figure 6B shown, at block 612, method 600 further includes diagnosing, by the non-airborne inference engine 224, corresponding failure modes of the aircraft system based on those failed tests caused by the diagnosed failure modes and using graph theory machine learning algorithms and diagrams. As shown in block 614, the method further includes reporting any differences between the corresponding failure modes diagnosed by the non-airborne inference engine and one of the plurality of diagnosed failure modes based on the diagnostic data received from the airborne computer 206.

[0075] In some examples, the diagnostic data further includes maintenance messages 212 having maintenance actions determined by the airborne inference engine 208 for corresponding diagnosed failure modes of the plurality of diagnosed failure modes, and the airborne diagnostic causal model 210 further describes the relationships between possible failure modes and corresponding maintenance actions among the maintenance actions. For a particular maintenance message having a particular maintenance action for a corresponding diagnosed failure mode, as Figure 6C shown, method 600 further includes accessing a maintenance record having the performed maintenance actions for the corresponding diagnosed failure modes and thus having new maintenance actions, as shown in block 616. The method further includes identifying a maintenance action difference between the particular maintenance action and the performed maintenance actions to determine a new relationship between the corresponding diagnosed failure mode and the performed maintenance actions, as shown in block 618. Then, updating the airborne diagnostic causal model further includes updating the airborne diagnostic causal model to describe the new relationship between the corresponding diagnosed failure mode and the performed maintenance actions, as shown in block 610A.

[0076] In some examples, method 600 may be combined with method 500, where an updated on-board diagnostic causal model may be used to diagnose a fault mode of an aircraft system. This may include receiving a fault report at an on-board computer 206 of the aircraft 202 that includes an on-board inference engine 208 (see block 502). The on-board inference engine may access the updated on-board diagnostic causal model 210 and diagnose a fault mode of one or another aircraft system in the aircraft system based on a failed test and using graph theory algorithms and the updated on-board diagnostic causal model (see blocks 504 and 506). A maintenance action for the fault mode may be determined, and a maintenance message 212 that includes at least the maintenance action may be generated (see blocks 508 and 510).

[0077] According to example implementations of the present disclosure, system 200 and its subsystems may be implemented by various means. The means for implementing the system and its subsystems may include hardware, either alone or under the direction of one or more computer programs from a computer-readable storage medium. In some examples, one or more devices may be configured to act as or otherwise implement the system and its subsystems shown and described herein. In examples involving more than one device, the corresponding devices may be connected to each other or otherwise communicate with each other in a variety of different ways, such as directly or indirectly via a wired or wireless network, etc.

[0078] Figure 7 Illustrated is a device 700 that may be configured to implement Figure 2A the on-board computer 206 and / or the non-on-board computer 218 shown. Generally, the devices of example implementations of the present disclosure may be included in, include, or be implemented in one or more fixed or portable electronic devices. Examples of suitable electronic devices include smart phones, tablet computers, laptop computers, desktop computers, workstation computers, server computers, etc. The device may include one or more of each of a plurality of components, such as, for example, processing circuitry 702 (e.g., a processor unit) connected to a memory 704 (e.g., a storage device).

[0079] The processing circuitry 702 may be composed of one or more processors alone or in combination with one or more memories. Processing circuitry is generally any computer hardware capable of processing information such as, for example, data, computer programs, and / or other suitable electronic information. The processing circuitry consists of a collection of electronic circuits, some of which may be encapsulated as an integrated circuit or multiple interconnected integrated circuits (integrated circuits are sometimes more commonly referred to as "chips"). The processing circuitry may be configured to execute a computer program that may be stored on the processing circuitry or otherwise stored in the memory 704 (of the same or another device).

[0080] According to a particular implementation, the processing circuitry 702 can be multiple processors, multi-core processors, or some other type of processor. Additionally, the processing circuitry can be implemented using multiple heterogeneous processor systems, in which a main processor coexists with one or more auxiliary processors on a single chip. As another illustrative example, the processing circuitry can be a symmetric multi-processor system that includes multiple processors of the same type. In yet another example, the processing circuitry can be implemented as or otherwise include one or more ASICs, FPGAs, etc. Thus, although the processing circuitry is capable of executing a computer program to perform one or more functions, the processing circuitry of the various examples is capable of performing one or more functions without the aid of a computer program. In any instance, the processing circuitry can be appropriately programmed to perform the functions or operations according to the example implementations of the present disclosure.

[0081] The memory 704 is generally any computer hardware capable of storing information, such as data, computer programs (e.g., computer-readable program code 706), and / or other suitable information, on a temporary basis and / or a permanent basis. The memory can include volatile and / or non-volatile memory and can be fixed or removable. Examples of suitable memories include random access memory (RAM), read-only memory (ROM), hard disk drives, flash memory, thumb drives, removable computer disks, optical disks, magnetic tapes, or some combination of the foregoing. Optical disks can include compact disk-read only memory (CD-ROM), compact disk-read / write (CD-R / W), digital versatile disk (DVD), etc. In various instances, the memory can be referred to as a computer-readable storage medium. A computer-readable storage medium is a non-transitory device capable of storing information and can be distinguished from a computer-readable transmission medium, such as an electronic transitory signal capable of carrying information from one location to another. As described herein, a computer-readable medium generally can refer to a computer-readable storage medium or a computer-readable transmission medium.

[0082] In addition to the memory 704, the processing circuitry 702 can also be connected to one or more interfaces for displaying, transmitting, and / or receiving information. The interfaces can include a communication interface 708 (e.g., a communication unit) and / or one or more user interfaces. The communication interface can be configured to, for example, transmit information to other devices, networks, etc. and / or receive information from other devices, networks, etc. The communication interface can be configured to transmit and / or receive information via physical (wired) and / or wireless communication links. Examples of suitable communication interfaces include network interface controllers (NICs), wireless NICs (WNICs), etc.

[0083] The user interface may include a display device 710 and / or one or more user input interfaces 712 (e.g., input / output unit). The display device may be configured to present or otherwise display information to the user. Suitable examples of the display device include a liquid crystal display (LCD), a light emitting diode display (LED), a plasma display panel (PDP), etc. The user input interface may be wired or wireless and may be configured to receive information from the user into the device, such as for processing, storing, and / or displaying. Suitable examples of the user input interface include a microphone, an image or video capture device, a keyboard or keypad, a joystick, a touch-sensitive surface (separate from or integrated into a touch screen), a biometric sensor, etc. The user interface may also include one or more interfaces for communicating with peripheral devices such as printers, scanners, etc.

[0084] As indicated above, the program code instructions may be stored in the memory and executed by the processing circuit thus programmed to implement the functions of the systems, subsystems, tools, and their corresponding elements described herein. As will be appreciated, any suitable program code instructions may be loaded from a computer-readable storage medium into a computer or other programmable device to produce a particular machine, such that the particular machine becomes a means for implementing the functions specified herein. These program code instructions may also be stored in a computer-readable storage medium, which may direct a computer, a processing circuit, or other programmable device to operate in a particular manner, thereby generating a particular machine or a particular article of manufacture. The instructions stored in the computer-readable storage medium may produce an article of manufacture, wherein the article of manufacture becomes a means for implementing the functions described herein. The program code instructions may be retrieved from the computer-readable storage medium and loaded into a computer, a processing circuit, or other programmable device to configure the computer, the processing circuit, or other programmable device to perform operations that are performed on or by the computer, the processing circuit, or other programmable device.

[0085] The retrieval, loading, and execution of the program code instructions may be performed sequentially such that one instruction is retrieved, loaded, and executed at a time. In some example implementations, the retrieval, loading, and / or execution may be performed in parallel such that multiple instructions are retrieved, loaded, and / or executed together. The execution of the program code instructions may produce a computer-implemented process such that the instructions executed by the computer, the processing circuit, or other programmable device provide operations for implementing the functions described herein.

[0086] The execution of instructions by a processing circuit or the storage of instructions in a computer-readable storage medium supports a combination of operations for performing a specified function. In this way, device 700 may include processing circuit 702 and a computer-readable storage medium or memory 704 coupled to the processing circuit, where the processing circuit is configured to execute computer-readable program code 706 stored in the memory. It will also be understood that one or more functions and combinations of functions may be implemented by a dedicated hardware-based computer system and / or processing circuit that performs the specified function and / or a combination of dedicated hardware and program code instructions.

[0087] In addition, the present disclosure includes embodiments in accordance with the following:

[0088] Item 1. An on-board computer (206, 700) for diagnosing faults on an aircraft (202), the aircraft including an aircraft system configured to report faults to an on-board inference engine (208), the on-board computer including:

[0089] A memory (704) configured to store computer-readable program code (706) including an on-board inference engine; and

[0090] A processing circuit (702) configured to access the memory and execute the computer-readable program code to cause the on-board computer to at least:

[0091] Receive (502) a fault report (302) from one of the aircraft systems in the aircraft system, the fault report indicating a failed test (306A, 306B) reported by the aircraft system;

[0092] Have the on-board inference engine access (504) an on-board diagnostic causal model (210, 300), the on-board diagnostic causal model being a graphical representation describing known causal relationships between possible failed tests (308A, 308B) reported by corresponding aircraft systems in the aircraft system and possible fault modes (308C) of the corresponding aircraft systems in the aircraft system;

[0093] Have the on-board inference engine diagnose (506) a fault mode of one of the aircraft systems or another aircraft system in the aircraft system based on the failed test and using graph theory algorithms and the on-board diagnostic causal model;

[0094] Determine (508) a maintenance action for the fault mode; and

[0095] Generate (510) a maintenance message (212, 312) by the on-board computer including at least the maintenance action.

[0096] Item 2. The airborne computer (206, 700) according to Item 1, wherein the airborne diagnostic causal model (210, 300) is represented by a graph including nodes connected by edges (310), the nodes representing possible failed tests (308A, 308B) and possible fault modes (308C), and the edges indicating known causal relationships between the possible failed tests and the possible fault modes.

[0097] Item 3. The airborne computer (206, 700) according to Item 1, wherein the airborne diagnostic causal model (210, 300) is represented by a graph that is a set of graphs for respective aircraft systems in the aircraft system, the graphs reflecting fault propagation behavior within the respective aircraft systems in the aircraft system, and the set reflecting fault propagation behavior across connected aircraft systems in the aircraft system.

[0098] Item 4. The airborne computer (206, 700) according to Item 1, wherein the graph theory algorithm is an optimal solution set (OSS) algorithm.

[0099] Item 5. The airborne computer (206, 700) according to Item 1, wherein the processing circuit (702) is configured to execute computer-readable program code (706) to further cause the airborne computer to:

[0100] Send maintenance messages (212, 312) to a display device (214, 710) on the aircraft (202) or a display device on a maintenance component, the display device being configured to establish a connection with the airborne computer to receive and display the maintenance messages.

[0101] Item 6. The airborne computer (206, 700) according to Item 5, wherein the display of the maintenance messages (212, 312) references instructions (216, 314) for performing maintenance actions to resolve the fault modes (308C) diagnosed by the airborne inference engine (208).

[0102] Item 7. A method (500) for diagnosing faults on an aircraft (202), the aircraft including an aircraft system configured to report faults to an airborne inference engine (208), the method including:

[0103] Receiving (502) at an airborne computer (206) of the aircraft (202) including the airborne inference engine from one of the aircraft systems in the aircraft system a fault report (302) indicating a failed test (306A, 306B) reported by the aircraft system;

[0104] An on-board inference engine accesses (504) an on-board diagnostic causal model (210, 300), which is a graphical representation describing known causal relationships between possible failure tests (308A, 308B) reported by corresponding aircraft systems in an aircraft system and possible fault modes (308C) of the corresponding aircraft systems in the aircraft system;

[0105] The on-board inference engine diagnoses (506) a fault mode of one aircraft system or another in the aircraft system based on the failure test and using a graph theory algorithm and the on-board diagnostic causal model;

[0106] Determine (508) a maintenance action for the fault mode; and

[0107] An on-board computer generates (510) a maintenance message (212, 312) that at least includes the maintenance action.

[0108] Item 8. The method (500) according to item 7, wherein the on-board diagnostic causal model (210, 300) is represented by a graph including nodes connected by edges (310), the nodes represent possible failure tests (308A, 308B) and possible fault modes (308C), and the edges indicate known causal relationships between the possible failure tests and the possible fault modes.

[0109] Item 9. The method (500) according to item 7, wherein the on-board diagnostic causal model (210, 300) is represented by a graph that is a set of graphs for corresponding aircraft systems in the aircraft system, the graphs reflect fault propagation behavior within the corresponding aircraft systems in the aircraft system, and the set reflects fault propagation behavior across connected aircraft systems in the aircraft system.

[0110] Item 10. The method (500) according to item 7, wherein the graph theory algorithm is an optimal solution set (OSS) algorithm.

[0111] Item 11. The method (500) according to item 7 further includes:

[0112] Sending (512) the maintenance message (212, 312) to a display device (214) on the aircraft (202) or a display device on a maintenance component, the display device being configured to establish a connection with the on-board computer to receive the maintenance message.

[0113] Item 12. The method (500) according to item 7, wherein the display of the maintenance message (212, 312) references instructions (216, 314) for performing the maintenance action to resolve the fault mode (308C) diagnosed by the on-board inference engine (208).

[0114] Item 13. The method (500) according to item 7, wherein the fault report (302) is one of a plurality of fault reports of failed tests (306A, 306B) reported by a corresponding aircraft system in an aircraft system, the fault mode (308C) is one of a plurality of diagnosed fault modes of at least some of the aircraft systems in the aircraft system that caused the failed test, and the method further includes:

[0115] Accessing (514) diagnostic data received from an on-board computer (206, 700), the diagnostic data including a plurality of fault reports and a plurality of diagnosed fault modes;

[0116] Accessing (516) by a non-on-board inference engine (224) a non-on-board diagnostic causal model (226) that describes the causal relationship between a failed test (402) and a plurality of diagnosed fault modes (404), the non-on-board diagnostic causal model being constructed using a graph theory machine learning algorithm trained using historical diagnostic data;

[0117] Comparing (518) the diagnostic data with the non-on-board diagnostic causal model; and based on the comparison,

[0118] Determining a new causal relationship in the causal relationships described by the non-on-board diagnostic causal model, the new causal relationship being new relative to known causal relationships; and

[0119] Updating (522) the on-board diagnostic causal model to further describe the new causal relationship, including generating an updated on-board diagnostic causal model and uploading the updated on-board diagnostic causal model to the on-board computer using a loadable software aircraft component upload toolset (228).

[0120] Item 14. The method (500) according to item 13, wherein the diagnostic data further includes a maintenance message (212) having a maintenance action for a corresponding diagnosed fault mode in a plurality of diagnosed fault modes (308C) determined by an on-board inference engine (208), and the on-board diagnostic causal model (210, 300) further describes the relationship between possible fault modes and corresponding maintenance actions among the maintenance actions, and

[0121] wherein, for a specific maintenance message (212, 312) having a specific maintenance action for a corresponding diagnosed fault mode, the method further includes:

[0122] Accessing (524) a maintenance record having the performed maintenance action for the corresponding diagnosed fault mode; and

[0123] Identify (526) the maintenance action differences between the specific maintenance actions and the performed maintenance actions to determine a new relationship between the corresponding diagnosed fault modes and the performed maintenance actions.

[0124] Wherein, updating the on-board diagnostic causal model (210, 300) further includes updating (522A) the on-board diagnostic causal model to describe the new relationship between the corresponding diagnosed fault modes and the performed maintenance actions.

[0125] Item 15. The method (500) according to item 13, wherein the graph theory machine learning algorithm is a graph theory algorithm for finding a maximum clique.

[0126] Item 16. A system (200) for diagnosing faults on an aircraft (202), the aircraft including aircraft systems configured to report faults to an on-board inference engine (208), the system including:

[0127] An on-board computer (206, 700) including an on-board inference engine, the on-board computer being configured to receive a fault report (302) of the aircraft from one of the aircraft systems in the aircraft system, the fault report indicating a failed test (306A, 306B) reported by the aircraft system, the on-board inference engine being configured to:

[0128] Access (504) an on-board diagnostic causal model (210, 300) represented by a graph, the graph describing known causal relationships between possible failed tests (308A, 308B) reported by the corresponding aircraft systems in the aircraft system and possible fault modes (308C) of the corresponding aircraft systems in the aircraft system; and

[0129] Diagnose (506) a fault mode of one of the aircraft systems or another aircraft system in the aircraft system according to the failed test and using the graph theory algorithm and the on-board diagnostic causal model.

[0130] Wherein, the on-board computer is configured to determine (508) a maintenance action for the fault mode and generate (510) a maintenance message (212, 312) at least including the maintenance action.

[0131] Item 17. The system (200) according to item 16, wherein the on-board diagnostic causal model (210, 300) is represented by a graph including nodes connected by edges (310), the nodes representing possible failed tests (308A, 308B) and possible fault modes (308C), and the edges indicating the known causal relationships between the possible failed tests and the possible fault modes.

[0132] Item 18. The system (200) according to item 16, wherein the on-board diagnostic causal model (210, 300) is represented by a graph which is a set of graphs for respective aircraft systems in the aircraft system, the graph reflecting the fault propagation behavior within the respective aircraft systems in the aircraft system, and the set reflecting the fault propagation behavior across the connected aircraft systems in the aircraft system.

[0133] Item 19. The system (200) according to item 16, wherein the graph theory algorithm is an optimal solution set (OSS) algorithm.

[0134] Item 20. The system (200) according to item 16, wherein the on-board computers (206, 700) are further configured to send maintenance messages (212, 312) to a display device (214) on the aircraft (202) or a display device on a maintenance component, the display device being configured to establish a connection with the on-board computer to receive the maintenance message.

[0135] Item 21. The system (200) according to item 20, further comprising: a display device (214) on the aircraft (202) or a display device on a maintenance component, wherein the display of the maintenance message (212, 312) references instructions (216, 314) for performing maintenance actions to resolve the fault mode (308C) diagnosed by the on-board inference engine (208).

[0136] Item 22. The system (200) according to item 16, wherein the fault report (302) is one of a plurality of fault reports of failed tests (306A, 306B) reported by respective aircraft systems in the aircraft system, the fault mode (308C) is one of a plurality of diagnosed fault modes in at least some of the aircraft systems in the aircraft system that caused the failed test,

[0137] wherein the on-board computers (206, 700) are further configured to send diagnostic data including the plurality of fault reports and the plurality of diagnosed fault modes, and

[0138] wherein the system further comprises a non-on-board computer (218, 700) which includes a non-on-board inference engine (224), the non-on-board computer being configured to:

[0139] access (514) the diagnostic data received from the on-board computer;

[0140] access (516) by the non-on-board inference engine a non-on-board diagnostic causal model (226) which describes the causal relationship between the failed test (402) and the plurality of diagnosed fault modes (404), the non-on-board diagnostic causal model being constructed using a graph theory machine learning algorithm trained using historical diagnostic data;

[0141] Compare the diagnostic data with a non-airborne diagnostic causal model (518); and based on this comparison,

[0142] Determine (520) new causal relationships in the causal relationships described by the non-airborne diagnostic causal model, where the new causal relationships are new relative to the known causal relationships; and

[0143] Update (522) the airborne diagnostic causal model to further describe the new causal relationships, including that the non-airborne computer is configured to generate an updated airborne diagnostic causal model, and upload the updated airborne diagnostic causal model to the airborne computer using a loadable software aircraft component upload toolset (228).

[0144] Item 23. The system (200) according to item 22, wherein the diagnostic data further includes maintenance messages (212, 312) having maintenance actions for corresponding diagnosed fault modes (308C) determined by an airborne inference engine (208), and the airborne diagnostic causal model (210, 300) also describes the relationship between possible fault modes and corresponding maintenance actions among the maintenance actions,

[0145] wherein, for a specific maintenance message (212) having a specific maintenance action for a corresponding diagnosed fault mode, the non-airborne computer (218, 700) is further configured to:

[0146] Access (524) a maintenance record having the performed maintenance actions for the corresponding diagnosed fault mode; and

[0147] Identify (526) the maintenance action difference between the specific maintenance action and the performed maintenance action to determine a new relationship between the corresponding diagnosed fault mode and the performed maintenance action, and

[0148] wherein the non-airborne computer configured to update the airborne diagnostic causal model includes a non-airborne computer further configured to update (522A) the airborne diagnostic causal model to describe the new relationship between the corresponding diagnosed fault mode and the performed maintenance action.

[0149] Item 24. The system (200) according to item 22, wherein the graph theory machine learning algorithm is a graph theory algorithm for finding maximum cliques.

[0150] Benefiting from the teachings presented in the foregoing description and the associated drawings, those skilled in the art to which this disclosure pertains will envision many modifications and other implementations of the disclosure set forth herein. Accordingly, it is to be understood that the disclosure is not limited to the specific implementations disclosed, and that modifications and other implementations are intended to be included within the scope of the appended claims. Additionally, although the foregoing description and the associated drawings describe example implementations in the context of certain example combinations of elements and / or functions, it is to be understood that different combinations of elements and / or functions may be provided by alternative implementations without departing from the scope of the appended claims. In this regard, for example, combinations of elements and / or functions different from those explicitly described above are also contemplated, as may be set forth in some of the appended claims. Although specific terms are employed herein, they are used only in a general and descriptive sense and not for purposes of limitation.

Claims

1. A method for diagnosing faults on an aircraft, the aircraft including aircraft systems configured to report faults to an on-board inference engine, the method including: Receiving a fault report at an on-board computer of the aircraft that includes the on-board inference engine from one of the aircraft systems in the aircraft systems, wherein the fault report indicates a failed test reported by the one aircraft system; Accessing, by the on-board inference engine, an on-board diagnostic causal model, the on-board diagnostic causal model being represented by a graph that describes known causal relationships between possible failed tests reported by corresponding aircraft systems in the aircraft systems and possible fault modes of the corresponding aircraft systems in the aircraft systems; Diagnosing, by the on-board inference engine, a fault mode of the one aircraft system or another aircraft system in the aircraft systems based on the failed test and using a graph theory algorithm and the on-board diagnostic causal model; Determining a maintenance action for the fault mode; and Generating, by the on-board computer, a maintenance message that includes at least the maintenance action; wherein the fault report is one of a plurality of fault reports of failed tests reported by corresponding aircraft systems in the aircraft systems, the fault mode is one of a plurality of diagnosed fault modes of at least some of the aircraft systems in the aircraft systems that caused the failed test, and the method further includes: Accessing diagnostic data received from the on-board computer, the diagnostic data including the plurality of fault reports and the plurality of diagnosed fault modes; Accessing, by a non-on-board inference engine, a non-on-board diagnostic causal model that describes causal relationships between the failed test and the plurality of diagnosed fault modes, the non-on-board diagnostic causal model being constructed using a graph theory machine learning algorithm trained using historical diagnostic data; Comparing the diagnostic data with the non-on-board diagnostic causal model; and based on the comparison; Determining new causal relationships in the causal relationships described by the non-on-board diagnostic causal model, the new causal relationships being new relative to the known causal relationships; and Updating the on-board diagnostic causal model to further describe the new causal relationships, including generating an updated on-board diagnostic causal model and uploading the updated on-board diagnostic causal model to the on-board computer using a loadable software aircraft component upload toolset.

2. The method according to claim 1, wherein, The on-board diagnostic causal model is represented by the graph that includes nodes connected by edges, the nodes representing the possible failed tests and the possible fault modes, and the edges indicating the known causal relationships between the possible failed tests and the possible fault modes.

3. The method according to claim 1, wherein, The on-board diagnostic causal model is represented by the graph, the graph being a set of graphs for corresponding aircraft systems in the aircraft systems, the graphs reflecting fault propagation behavior within the corresponding aircraft systems in the aircraft systems, and the set reflecting fault propagation behavior across connected aircraft systems in the aircraft systems.

4. The method according to claim 1, further including: Sending the maintenance message to a display device on the aircraft or a display device on a maintenance component, the display device being configured to establish a connection with the on-board computer to receive the maintenance message.

5. The method according to claim 4, wherein, The display reference of the maintenance message is used to execute the instructions for the maintenance action to address the fault mode diagnosed by the on-board inference engine.

6. The method according to claim 1, wherein, The diagnostic data further includes a maintenance message having a maintenance action for a respective diagnosed fault mode among the plurality of diagnosed fault modes determined by the on-board inference engine, and the on-board diagnostic causal model further describes the relationship between the possible fault modes and the respective maintenance actions among the maintenance actions, and wherein, for a particular maintenance message having a particular maintenance action for a respective diagnosed fault mode, the method further includes: accessing a maintenance record having the executed maintenance action for the respective diagnosed fault mode; and identifying a maintenance action difference between the particular maintenance action and the executed maintenance action to determine a new relationship between the respective diagnosed fault mode and the executed maintenance action, wherein updating the on-board diagnostic causal model further includes updating the on-board diagnostic causal model to describe the new relationship between the respective diagnosed fault mode and the executed maintenance action.

7. The method according to claim 1, wherein, The graph theory machine learning algorithm is a graph theory algorithm for finding a maximum clique.

8. A system for diagnosing faults on an aircraft, the aircraft including aircraft systems configured to report faults to an on-board inference engine, the system including: An on-board computer including the on-board inference engine, the on-board computer being configured to receive a fault report of the aircraft from one of the aircraft systems in the aircraft system, wherein the fault report indicates a failed test reported by the one aircraft system, and the on-board inference engine is configured to: access an on-board diagnostic causal model represented by a graph, the graph describing the known causal relationship between possible failed tests reported by respective aircraft systems in the aircraft system and possible fault modes of the respective aircraft systems in the aircraft system; and diagnose a fault mode of the one aircraft system or another aircraft system in the aircraft system according to the failed test and using a graph theory algorithm and the on-board diagnostic causal model, wherein the on-board computer is configured to determine a maintenance action for the fault mode and generate a maintenance message including at least the maintenance action; wherein the fault report is one of a plurality of fault reports of failed tests reported by respective aircraft systems in the aircraft system, and the fault mode is one of a plurality of diagnosed fault modes of at least some of the aircraft systems in the aircraft system that cause the failed test; wherein the on-board computer is further configured to send diagnostic data including the plurality of fault reports and the plurality of diagnosed fault modes; and wherein the system further includes a non-on-board computer, the non-on-board computer including a non-on-board inference engine, and the non-on-board computer is configured to: access the diagnostic data received from the on-board computer; access a non-on-board diagnostic causal model by the non-on-board inference engine, the non-on-board diagnostic causal model describing the causal relationship between the failed test and the plurality of diagnosed fault modes, and the non-on-board diagnostic causal model is constructed using a graph theory machine learning algorithm trained with historical diagnostic data; Compare the diagnostic data with the off-board diagnostic causal model; and based on the comparison; Determine new causal relationships in the causal relationships described by the off-board diagnostic causal model, the new causal relationships being new relative to the known causal relationships; and Update the on-board diagnostic causal model to further describe the new causal relationships, including that the off-board computer is configured to generate an updated on-board diagnostic causal model and upload the updated on-board diagnostic causal model to the on-board computer using a loadable software aircraft component upload toolset.

9. The system according to claim 8, wherein, The on-board diagnostic causal model is represented by the graph including nodes connected by edges, the nodes representing the possible failed tests and the possible fault modes, and the edges indicating the known causal relationships between the possible failed tests and the possible fault modes.

10. The system according to claim 8, wherein, The on-board diagnostic causal model is represented by the graph, which is a set of graphs for respective aircraft systems in the aircraft system, the graphs reflecting the fault propagation behavior within the respective aircraft systems in the aircraft system, and the set reflecting the fault propagation behavior across the connected aircraft systems in the aircraft system.

11. The system according to claim 8, wherein, The on-board computer is further configured to send the maintenance message to a display device on the aircraft or a display device on a maintenance component, the display device being configured to establish a connection with the on-board computer to receive the maintenance message.

12. The system according to claim 11, wherein, The display of the maintenance message refers to instructions for performing the maintenance action to resolve the fault mode diagnosed by the on-board inference engine.

13. The system according to claim 8, wherein, The diagnostic data further includes a maintenance message having a maintenance action determined by the on-board inference engine for a respective diagnosed fault mode among the plurality of diagnosed fault modes, and the on-board diagnostic causal model further describes the relationship between the possible fault mode and the respective maintenance action among the maintenance actions, wherein, for a specific maintenance message having a specific maintenance action for a respective diagnosed fault mode, the off-board computer is further configured to: Access a maintenance record having the performed maintenance action for the respective diagnosed fault mode; and Identify a maintenance action difference between the specific maintenance action and the performed maintenance action to determine a new relationship between the respective diagnosed fault mode and the performed maintenance action, and wherein the off-board computer configured to update the on-board diagnostic causal model includes the off-board computer further configured to update the on-board diagnostic causal model to describe the new relationship between the respective diagnosed fault mode and the performed maintenance action.

Images