Method and system for identifying correlated events in an aircraft

Abstract

The present application discloses a method and system for identifying correlated events in an aircraft. In one example, a method for identifying correlated events in an aircraft is described. The method includes: obtaining sensor data; obtaining fault code data; generating a set of events, each of which occurs within a time interval at which (i) the sensor data indicates an abnormal measurement or (ii) a fault code associated with a specific aircraft subsystem of the aircraft is signaled; calculating a value of a statistical correlation between the group; based on the value exceeding a threshold, constructing a network representing the group as a sequence of correlated events and further representing the timing of the occurrence of the sequence; indexing the sequence and the value in a summary table stored in a memory and separately from the sensor data and the fault code data; and controlling a display device to display a visual representation of the summary table and the network.

Description

Technical Field

[0001] The present disclosure relates generally to identifying correlated events in an aircraft, and in particular to how to correlate events such as abnormal sensor readings and aircraft component failures. Background Art

[0002] Aircraft are often equipped with equipment configured to record data associated with the aircraft. This data may include sensor readings from various subsystems of the aircraft, such as a flight control subsystem, a propulsion subsystem, or a satellite communications subsystem. When the sensor readings meet certain criteria, the aircraft may responsively issue a fault message to alert a pilot or maintenance personnel to a potential problem, such as a failure of a particular aircraft subsystem. These fault messages may also be recorded.

[0003] In existing aircraft, when an event (such as an abnormal sensor reading or a fault message) occurs in one subsystem, it may follow a cascading chain of other events that can propagate through various other subsystems. When this happens, it is difficult for maintenance personnel to determine the root cause of the problem and, in turn, determine a solution to the problem.

[0004] Existing methods and systems build predictive and prognostic models that consider fault messages and other data sources as the basis for predicting when certain aircraft components may reach the end of their life and need to be removed or replaced. However, these existing methods and systems may be inefficient, inaccurate, and / or ineffective for the purpose of diagnosing the causes of hardware and software problems, especially problems that propagate across multiple subsystems as described above. As a result, additional effort may be expended during diagnostics and maintenance to resolve the problem.

[0005] What are needed are systems and methods for improving the accuracy and efficiency of fault diagnosis, which in turn can improve the accuracy and efficiency of efforts expended to resolve faults and reduce the costs associated therewith. Summary of the invention

[0006] In one example, a method for identifying a correlation event in an aircraft is described. The method includes: obtaining sensor data, the sensor data including voltage measurements of at least one electrical bus on the aircraft over time; obtaining a fault code (fault code The method comprises: generating a set of events from the sensor data and the fault code data, wherein the fault code data includes fault codes signaled over time by a plurality of aircraft subsystems on the flight, wherein each of the fault codes is associated with a corresponding aircraft subsystem of the plurality of aircraft subsystems, and upon detection of an error, causing the corresponding aircraft subsystem to output the fault code; calculating a voltage threshold value, and when the voltage measurement exceeds the voltage threshold value, indicating an abnormal voltage measurement in the sensor data; generating a set of events from the sensor data and the fault code data, wherein each event in the set of events occurs within a time interval at which (i) the sensor data indicates the abnormal voltage measurement or (ii) a fault code associated with a particular aircraft subsystem of the plurality of aircraft subsystems is signaled; calculating a value of a statistical correlation between the set of events, wherein the value of the statistical correlation between the set of events indicates a likelihood that the set of events are causally related events that occur in a time sequence; and constructing a temporal precursor network that represents the set of events as a sequence of related events and further represents the time sequence of occurrence of the sequence of related events based on the value of the statistical correlation between the set of events exceeding a predetermined threshold value. network); indexing said values ​​of statistical correlation between said sequence of related events and said set of events in a summary table stored in a memory and separate from said sensor data and said fault code data; and controlling a display device to display said summary table and a visual representation of said temporal precursor network on a graphical user interface.

[0007] In another example, a non-transitory computer-readable medium is described. The non-transitory computer-readable medium has instructions stored thereon, which, when executed by one or more processors of a computing device, cause the computing device to perform operations including: obtaining sensor data, the sensor data including voltage measurements of at least one electrical bus on an aircraft over time; obtaining fault code data, the fault code data including fault codes signaled over time by a plurality of aircraft subsystems on the aircraft, wherein each of the fault codes is associated with a corresponding aircraft subsystem in the plurality of aircraft subsystems, and once an error is detected, the corresponding aircraft subsystem is caused to output the fault code; calculating a voltage threshold, indicating an abnormal voltage measurement in the sensor data when the voltage measurement exceeds the voltage threshold; generating a set of events from the sensor data and the fault code data, wherein each event in the set of events occurs within a time interval, and ... At the time interval, (i) the sensor data indicates the abnormal voltage measurement or (ii) a fault code associated with a specific aircraft subsystem of the multiple aircraft subsystems is signaled; calculating a value of a statistical correlation between the set of events, wherein the value of the statistical correlation between the set of events indicates a likelihood that the set of events are causally related events occurring in a time sequence; constructing a temporal precursor network representing the set of events as a sequence of related events and further representing the time sequence of occurrence of the sequence of related events based on the value of the statistical correlation between the set of events exceeding a predetermined threshold; indexing the sequence of related events and the value of the statistical correlation between the set of events in a summary table stored in a memory and separate from the sensor data and the fault code data; and controlling a display device to display a visual representation of the summary table and the temporal precursor network on a graphical user interface.

[0008] In another example, a system is described. The system includes multiple aircraft subsystems on an aircraft. The system also includes a fault diagnosis computing device having a processor and a memory, the memory storing instructions executable by the processor to perform operations including the following: obtaining sensor data, the sensor data including voltage measurements of at least one electrical bus on the aircraft over time; obtaining fault code data, the fault code data including fault codes signaled by the multiple aircraft subsystems over time, wherein each of the fault codes is associated with a corresponding aircraft subsystem in the multiple aircraft subsystems, and once an error is detected, the corresponding aircraft subsystem is caused to output the fault code; calculating a voltage threshold, when the voltage measurement exceeds the voltage threshold, indicating an abnormal voltage measurement in the sensor data; generating a set of events from the sensor data and the fault code data, wherein each event in the set of events occurs within a time interval, and within the time interval (i) the sensor data indicates the abnormal voltage measurement or (ii) a fault code associated with a specific aircraft subsystem of the multiple aircraft subsystems is signaled; calculating a value of a statistical correlation between the set of events, wherein the value of the statistical correlation between the set of events indicates the likelihood that the set of events are causally related events that occur in a time sequence; constructing a time precursor network representing the set of events as a sequence of related events and further representing the time sequence of occurrence of the sequence of related events based on the value of the statistical correlation between the set of events exceeding a predetermined threshold; indexing the sequence of related events and the value of the statistical correlation between the set of events in a summary table stored in the memory and separately from the sensor data and the fault code data; and controlling a display device to display a visual representation of the summary table and the time precursor network on a graphical user interface.

[0009] The features, functions, and advantages that have been discussed can be achieved independently in various examples or may be combined in yet other examples. Further details of the examples can be seen with reference to the following description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The novel features believed to be characteristic of the illustrative examples are set forth in the appended claims. However, the illustrative examples together with the preferred mode of use, further objects and description thereof will be best understood by reference to the following detailed description of illustrative examples of the present disclosure when read in conjunction with the accompanying drawings, in which:

[0011] Figure 1 A block diagram of an exemplary aircraft is shown in accordance with an exemplary implementation.

[0012] Figure 2An example temporal precursor network is shown in accordance with an exemplary embodiment.

[0013] Figure 3 A flowchart of an example method for identifying correlated events in an aircraft is shown in accordance with an illustrative embodiment.

[0014] Figure 4 A method for performing Figure 3 A flow chart of an example method of calculating the functionality of the method.

[0015] Figure 5 According to an exemplary embodiment, Figure 3 The method is used in conjunction with a flowchart of an example method.

[0016] Figure 6 A method for performing Figure 3 A flowchart of an example method of the indexing functionality of the method.

[0017] Figure 7 A method for performing Figure 3 A flow chart of an example method of controlling functionality of a method.

[0018] Figure 8 A method for performing Figure 3 A flow chart of another example method of controlling functionality of a method.

[0019] Fig. 9 A method for performing Figure 3 A flow chart of another example method of controlling functionality of a method.

[0020] Fig.10 A method for performing Figure 3 A flow chart of another example method of controlling functionality of a method.

[0021] Fig.11 According to an exemplary embodiment, Figure 3 A flowchart of another example method for use with the method and also shows a method for performing Figure 3 A flow chart of another example method of controlling functionality of a method.

[0022] Fig.12 According to an exemplary embodiment, Figure 3 A flowchart of another example method for use in conjunction with the method.

[0023] Fig.13 According to an exemplary embodiment, Figure 3A flowchart of another example method for use in conjunction with the method. DETAILED DESCRIPTION

[0024] The disclosed examples will now be described more fully below with reference to the accompanying drawings, in which some but not all disclosed examples are shown. In fact, several different examples may be described and should not be construed as being limited to the examples set forth herein. Instead, these examples are described so that the disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0025] In an example, methods and systems for identifying correlated events are described. More specifically, the example methods and systems relate to determining statistically significant correlations between events occurring on a plurality of different aircraft components, such as electrical buses and aircraft subsystems.

[0026] In the present disclosure, aircraft subsystems that may acquire sensor readings and electrical buses that carry those sensor readings may be considered separate components of the aircraft because, in some cases, it may not be readily apparent which aircraft subsystem a given sensor reading detected on an electrical bus is associated with since a given electrical bus may carry sensor readings from multiple different subsystems.

[0027] The data recorder on the aircraft is configured to record data over time, such as data recorded during the flight of the aircraft. The recorded data may include sensor data and fault code data, etc. The sensor data may take the form of voltage measurements or other measurements determined by sensors on the aircraft. Once acquired by the sensor, the sensor data may be transmitted to other systems on the aircraft, such as a data recorder, on one or more electrical buses of the aircraft. On the other hand, the fault code data includes fault codes that have been signaled over time by various aircraft subsystems on the aircraft, such as flight control subsystems, display subsystems, environmental control subsystems, propulsion control subsystems, communication subsystems, integrated pressurized air subsystems and / or integrated helmet and display sight system (IHADSS) subsystems and other possibilities. In particular, when an aircraft subsystem detects an error, the aircraft subsystem outputs a signal identifying a specific fault code. In an example, the fault code may take the form of a string of numbers and / or letters (e.g., "3456"). Additionally or alternatively, the signal may include data representing a textual description of the fault code, such as "bleed air pressure fault" or "flight management multiplexer (MUX) fault."

[0028] The methods and systems described herein may involve using recorded data from multiple sources (i.e., sensor data and fault code data) to generate events that occur in an aircraft and identify associations between events. An "event" is generally referred to as a time interval during which (i) a sensor reading from the sensor data meets criteria indicating an abnormality (e.g., a threshold high positive or negative spike in voltage) or (ii) a fault code is signaled. In this disclosure, abnormal sensor readings will be primarily described in the context of abnormal voltage measurements, but in other embodiments, other measurements may be used as a basis for identifying an event. For example, when sensor data indicates a measurement of current or power that exceeds a particular threshold, the measurement may be considered abnormal.

[0029] The methods and systems described herein may involve detecting that certain event criteria have been met (e.g., an abnormal sensor reading and / or signaling a fault code) and generating a set of events / event sets that occur with respect to an aircraft. As discussed above, the occurrence of one or more events may trigger causally related events in multiple components on an aircraft. For example, an error in a multiplexer in a signal processing subsystem may cause an error in a flight control subsystem, which may in turn cause an error in a multiplexer in a flight management subsystem. The data log may record a fault code indicating each error, but it may be difficult to discern which error caused the other errors, or even to determine if the errors are at all related. The problem may become complicated when many (e.g., dozens or hundreds) of errors occur during a single flight of the aircraft, on multiple different flights of the aircraft, and / or on flights of other aircraft.

[0030] Therefore, the methods and systems described herein may also involve calculating the value of a statistical dependence between a group of events. This value indicates the possibility that the group of events is a causal chain or "sequence" of events that occur in a temporal order. In other words, this value indicates whether the group of events occurs due to random coincidence or due to potential correlation. Associated (e.g., statistically correlated) paired or larger event groups are used to construct a temporal precursor network, which represents the sequence of related events and the temporal order in which they occur. In addition, the associated paired or larger event groups and the values ​​representing their statistical dependence are indexed in a summary table stored in a memory, separated from sensor data, fault code data, and other data recorded by a data recorder. Then, the summary table and the temporal precursor network are both displayed on a graphical user interface. Therefore, these methods and systems effectively provide accurate indications of associations (or lack of associations) between events, which are used as executable insights into problems on the aircraft.

[0031] In the manner discussed above, the method and system can facilitate the dynamic acquisition of recorded sensor data and fault code data and the identification of important events occurring with respect to the aircraft from the data, thereby reducing the amount of data considered when diagnosing and resolving aircraft problems. In addition, the method and system described herein can facilitate the calculation of statistical correlations between events (thereby narrowing down the potential causes of a sequence of related events) and indexing sequences of related events and indications of their statistical correlations in a summary table that is separate from the location where the sensor data and fault diagnostic data are stored. In addition, the method and system described herein can facilitate the display of sequences of related events in a simplified but useful manner in the form of a time precursor network. The time precursor network and summary table enable quick retrieval and efficient storage / display of valuable information that can be used to effectively diagnose and resolve aircraft problems caused by a sequence of events. The time precursor network and summary table can be dynamically updated as additional recorded data is analyzed and additional event sequences are identified.

[0032] Example methods and systems may involve indexing other useful information and performing other useful calculations to provide additional accuracy and efficiency in identifying associations between events. In one example, methods and systems may calculate and index data indicating consistency in occurrences of related events, such as data indicating whether a sequence of related events is exhibited in many different flights of an aircraft and / or on other aircraft, rather than if the sequence of related events is more isolated occurrences. Other examples are possible.

[0033] Embodiments of the present disclosure provide technical improvements specific to computer networks and computing systems, such as computing systems on an aircraft, that are integrated to facilitate safe and efficient operation of the aircraft.

[0034] Technical issues specific to computing systems, such as the management and use of large amounts of complex data originating from multiple sources and the inefficiencies associated therewith, can be fully or partially addressed by embodiments of the present disclosure. For example, embodiments of the present disclosure reduce the time spent analyzing data to reduce the causes of problems with aircraft subsystems, which in turn can reduce the time spent resolving these problems. Thus, embodiments of the present disclosure can reduce the cost and complexity of implementing less efficient methods and systems for diagnosing and resolving such problems. As another example, embodiments of the present disclosure improve the accuracy and reliability of diagnostic information.

[0035] Thus, embodiments of the present disclosure may introduce new and efficient ways in which a fault diagnostic system or other system on an aircraft identifies, stores and presents correlated events on the aircraft, and thereby facilitate new and efficient ways in which correlated events are used to diagnose and solve problems. Embodiments of the present disclosure may compress and perform calculations using large amounts of information that may not normally indicate a fault with a component of the aircraft in order to make the information interpretable for the purpose of identifying causal relationships of faults and other anomalies on the aircraft.

[0036] Now referring to the accompanying drawings, Figure 1 A block diagram of an exemplary aircraft 100 is shown, in accordance with an example implementation. Aircraft 100 includes system 102. In an example, system 102 may be configured to communicate with display device 104, and thus system 102 is shown in communication with display device 104.

[0037] The display device 104 can be in the form of a computing device with a touch screen, a computer monitor or other components capable of providing visual information such as a graphical user interface. The computing device can be a laptop computer, a tablet computer, a desktop computer, a smart phone or other types of computers, and the system 102 has a communication interface. Through the interface, the system 102 (or its components) can send instructions to the display device 104, and the instructions make the display device 104 display the information specified by the instructions when the display device 104 receives them. For example, the system 102 can control the display device 104 to provide the summary table and / or the time precursor network described in the present disclosure on the graphical user interface by providing instructions including data representing the summary table and / or the time precursor network to the display device 104. The display device 104 can also include other devices configured to receive input from a user (e.g., a maintenance worker of the aircraft 100 or a pilot of the aircraft 100) and / or provide output to the user, such as a keyboard, a mouse, etc.

[0038] exist Figure 1 , the display device 104 is shown as being separate from the aircraft 100. Additionally or alternatively, in some embodiments, the display device 104 may be a display device located on the aircraft 100, such as a computing device located in a cockpit of the aircraft 100.

[0039] In addition, the system 102 is shown to include a fault diagnosis computing device 106. The fault diagnosis computing device 106 is configured with program logic that enables the fault diagnosis computing device 106 to use the recorded data to perform the functions described in the present disclosure (e.g., functions related to identifying events that occurred in the aircraft 100) and to identify associations between events. In addition to or in place of the fault diagnosis computing device 106, other types of computing devices may be similarly configured and used to perform one or more functions. Such computing devices may be on the aircraft 100, or located remotely from the aircraft 100 but in communication with the aircraft 100.

[0040] The system 102 is also shown to include a plurality of aircraft subsystems 108 and data recorders 110, each of which is shown to be communicatively coupled to the fault diagnostic computing device 106 via a communication bus 112. The fault diagnostic computing device 106 may thus receive and transmit communications to the data recorders 110 and / or the aircraft subsystems 108 via the communication bus 112.

[0041] As described above, aircraft subsystems 108 may include various subsystems, each of which may be configured to perform certain functions, such as monitoring altitude, speed, temperature (e.g., temperature inside and outside aircraft 100), air pressure, fuel level, and orientation, monitoring whether various computing devices, sensors, or components thereof are operating as desired, and / or controlling physical components of aircraft 100 (e.g., rudders and ailerons), among other possibilities. Exemplary aircraft subsystems include flight control subsystems, display subsystems, and environmental control subsystems, among other possibilities. Exemplary aircraft subsystems themselves may each include sensors, computing devices, and / or other hardware necessary to perform the functions of the subsystem.

[0042] The data recorder 110 may take the form of a computing device having one or more processors, memory, and communication interfaces for receiving / transmitting communications from / to other components of the aircraft 100. The memory of the data recorder 110 may include one or more databases in which the data recorder 110 stores the recorded data. The data recorder 110 may store multiple types of data in the same database, or may store multiple different types of data, each in a corresponding database dedicated to storing that type of data.

[0043] As described above, the data recorder 110 is configured to acquire and store data from various sources, such as various sensors throughout the aircraft 100 or one or more other computing devices on the aircraft 100. The data recorded by the data recorder 110 may include sensor data and fault code data, among other possibilities. In the present disclosure, "sensor data" may include various types of measurements acquired by sensors throughout the aircraft 100, such as voltage, current, or power. In examples, these measurements may be directly associated with specific sensors or aircraft subsystems and may include information identifying the specific sensor or aircraft subsystem associated with the measurement. In other examples, it may be more difficult to determine the cause of the measurement, such as when the measurement is acquired from a sensor located along an electrical bus of the aircraft 100. For example, the electrical bus may carry signals associated with a condensing unit of the aircraft 100, a pneumatic system of the aircraft 100, and a compressor of a jet engine of the aircraft 100. When a sensor located at a certain location on the electrical bus measures a voltage spike at that location, the spike may be attributed to one or more of the condensing unit, the pneumatic system, or the compressor, but the measurement itself (being a voltage measurement) would not indicate so. There may be multiple electrical buses in aircraft 100, and at least one sensor is configured to acquire measurements on each such electrical bus. There are situations where a sensor reading from an electrical bus on aircraft 100 may differ from an expected reading, but the source of the difference may be unknown and not indicated by the sensor data.

[0044] As described above, the fault code data includes fault codes signaled by the aircraft subsystems 108 over time. Each fault code is associated with a corresponding aircraft subsystem of the aircraft subsystems 108. When the corresponding aircraft subsystem (or more specifically, a sensor associated with the corresponding aircraft subsystem) detects an error, the corresponding aircraft subsystem outputs a fault code signal. In the example, the error is a sensor reading outside a predetermined range, although the error can take other forms. A specific error corresponds to a specific fault code. Some aircraft subsystems may have multiple components that may fail or multiple ways in which an error may occur. Therefore, multiple fault codes may be associated with a single aircraft subsystem.

[0045] The data recorder 110 may receive signaled fault codes directly from the various aircraft subsystems 108, or may receive signaled fault codes from an intermediate device. For example, the aircraft 100 may include an intermediate device that, upon detecting a sensor reading that falls outside a predetermined range, triggers the intermediate device to signal a corresponding fault code to the data recorder 110 via the communication bus 112, and the fault code may be signaled as long as the sensor reading is outside the predetermined range.

[0046] When the data logger 110 receives a fault code, the data logger 110 may then store fault code data corresponding to the fault code, such as a start time when the fault code was first signaled and an end time when the fault code stopped signaling.

[0047] Each fault code of the fault codes is associated with a corresponding aircraft subsystem of the plurality of aircraft subsystems that, when an error is detected, causes the corresponding aircraft subsystem to output the fault code.

[0048] The data recorder 110 may also record other data. For example, when a fault code is signaled, the data recorder 110 may store data indicating the "tail number" (i.e., the identifier of the aircraft 100) associated with the fault code. The identifier of the aircraft 100 may uniquely identify the aircraft 100 relative to other aircraft.

[0049] The data recorder 110 may be configured to record data during a flight of the aircraft 100 and / or when the aircraft 100 is not in flight (e.g., in a hangar). For example, when the auxiliary power of the aircraft 100 is turned on, this may trigger the data recorder 110 to start recording data, and when the auxiliary power of the aircraft 100 is disconnected, this may trigger the data recorder 110 to stop recording data. For simplicity, the time period between the auxiliary power being turned on and off is referred to in this disclosure as a "flight" of the aircraft 100, even though the aircraft 100 may not be in the air during the entire period of time. However, it should be understood that other examples of depicting the start time and end time of a flight are also possible. In some embodiments, the data recorder 110 may be configured to store recorded data from one flight as one file (e.g., a delimited text file) and store recorded data from another flight in another separate file.

[0050] In some implementations, the fault diagnostic computing device 106 may be configured to perform some or all of the functions performed by the data logger 110. To facilitate this, for example, the data logger 110 may be a component of the fault diagnostic computing device 106 rather than a component of the data logger 110. Figure 1 Individual components shown.

[0051] Return to reference Figure 1 , the fault diagnosis computing device 106 is shown to include one or more processors 114, a communication interface 116, an output interface 118, and a data storage 120, all of which are connected to a communication bus 122. The fault diagnosis computing device 106 may also include hardware for implementing communication within the fault diagnosis computing device 106 and between the fault diagnosis computing device 106 and other devices (not shown). For example, the hardware may include a transmitter, a receiver, and an antenna.

[0052] The communication interface 116 can be a wireless interface and / or one or more wired interfaces, which allow short-range communication and long-range communication to one or more networks or one or more remote devices. Such a wireless interface can provide for communication under one or more wireless communication protocols, Bluetooth, WiFi (e.g., Institute of Electrical and Electronics Engineers (IEEE) 802.11 protocols), long term evolution (LTE), cellular communication, near field communication (NFC) and / or other wireless communication protocols. Such a wired interface can include an Ethernet interface, a universal serial bus (USB) interface or a similar interface to communicate to a wired network via a wire, twisted pair, coaxial cable, optical link, optical fiber link or other physical connection. Therefore, the communication interface 116 can be configured to receive input data from one or more devices, and can also be configured to send output data to other devices.

[0053] The output interface 118 outputs information, such as outputting information from the fault diagnosis computing device 106 to the data logger 110. For example, the output interface 118 may output a request to the data logger 110 to provide sensor data or fault code data to the fault diagnosis computing device 106. The output interface 118 may therefore be similar to the communication interface 116, and may also be a wireless interface (e.g., a transmitter) or a wired interface.

[0054] The data storage 120 may include or take the form of a memory, such as one or more computer-readable storage media that may be read or accessed by the processor(s) 114. The computer-readable storage media may include volatile and / or non-volatile storage components, such as optical, magnetic, organic, or other memory or disk storage, which may be integrated in whole or in part with the one or more processors 114. The data storage 120 is considered to be a non-transitory computer-readable medium. In some examples, the data storage 120 may be implemented using a single physical device (e.g., one optical, magnetic, organic, or other memory or disk storage unit), while in other examples, the data storage 120 may be implemented using two or more physical devices.

[0055] Thus, the data storage 120 is a non-transitory computer-readable storage medium, and executable instructions 124 are stored thereon. The executable instructions 124 include computer executable code. In addition, the data storage 120 also stores a summary table 126, in which the values ​​of the sequence and statistical correlation of related events can be placed. The data storage 120 also stores a time precursor network. The data storage 120 can also store events, such as a group of events (whether or not the group of events is determined to be a sequence of related events), and can store copies of sensor data and / or fault code data recorded by the data reader 110. Other data can also be stored in the data storage 120, examples of which will be described in more detail below.

[0056] The processor(s) 114 may be a general purpose processor or a special purpose processor (e.g., a digital signal processor, an application specific integrated circuit, etc.). The processor(s) 114 may be configured to execute executable instructions 124 (e.g., computer readable program instructions) stored in the data storage 120 and executable to provide the functionality of the fault diagnosis computing device 106 described herein. For example, the executable instructions 124 may define a software application that, when executed, causes the fault diagnosis computing device 106 to perform one or more of the functions described herein.

[0057] In one example, in operation, when the executable instructions 124 are executed by the processor(s) 114 of the fault diagnosis computing device 106, the processor(s) 114 perform functions including: obtaining sensor data, the sensor data including voltage measurements of at least one electrical bus on the aircraft over time; obtaining fault code data, the fault code data including fault codes signaled by the plurality of aircraft subsystems over time, wherein each of the fault codes is associated with a corresponding aircraft subsystem of the plurality of aircraft subsystems, and upon detection of an error, causing the corresponding aircraft subsystem to output the fault code; calculating a voltage threshold, indicating an abnormal voltage measurement in the sensor data when the voltage measurement exceeds the voltage threshold; generating a set of events from the sensor data and the fault code data, wherein each event in the set of events occurs within a time interval. , at which time interval (i) the sensor data indicates the abnormal voltage measurement or (ii) a fault code associated with a particular aircraft subsystem of the plurality of aircraft subsystems is signaled; calculating a value of a statistical correlation between the set of events, wherein the value of the statistical correlation between the set of events indicates a likelihood that the set of events are causally related events occurring in a time sequence; constructing a temporal precursor network representing the set of events as a sequence of correlated events and further representing the time sequence of occurrence of the sequence of correlated events based on the value of the statistical correlation between the set of events exceeding a predetermined threshold; indexing the sequence of correlated events and the value of the statistical correlation between the set of events in a summary table stored in the memory and separate from the sensor data and the fault code data; and controlling a display device to display a visual representation of the summary table and the temporal precursor network on a graphical user interface.

[0058] More detailed example operations will now be described. Although some operations are described as being performed in a particular order, it should be understood that such operations may be performed in a different order.

[0059] In the example described herein, the fault diagnosis computing device 106 can obtain the sensor data and fault code data acquired over time. At this stage, or at a certain time before obtaining the sensor data and fault code data, the user can select and input a time threshold to the fault diagnosis computing device 106, which can be used to build a time precursor network and calculate the value of the statistical correlation between a group of times. The time threshold represents the maximum amount of time passed between two identified events, and wherein the two events can still be reasonably considered to be associated. A large time threshold (e.g., one hour or longer) can increase the complexity of the time precursor network by generating more event sequences, and some of these events may not accurately represent causally related events. Therefore, it is expected that the time threshold is not too large, for example, less than one hour or as low as one second or ten minutes. In some embodiments, the time threshold can be determined based on the rate of recording / sampling fault codes and / or voltage measurements or based on other factors.

[0060] There may be a large amount of sensor data and / or fault code data recorded by the data recorder 110, and not all of the sensor data and / or fault code data may be helpful in identifying associated events in the aircraft 100. Additionally or alternatively, it may be computationally inefficient to use an excessive or undesirable amount of data to identify associated events. For at least these reasons, in some embodiments, at some point before the fault diagnosis computing device 106 generates a set of events from the sensor data and fault code data, for example, after the fault diagnosis computing device 106 obtains the sensor data and fault code data, the fault diagnosis computing device 106 may filter the obtained sensor data and fault code data.

[0061] In an example, the fault diagnosis computing device 106 can remove a portion of the sensor data and / or fault code data obtained during a predetermined time period or during a portion of the flight process (e.g., before flight, during flight, takeoff, landing). For example, when the aircraft is started at the pre-flight startup procedure, such a procedure may trigger a false fault code and / or voltage spike to occur, so the predetermined time period can be a time period during which the aircraft is expected to be started at the pre-flight startup procedure. In other examples, the fault diagnosis computing device 106 can remove a portion of the sensor data and / or fault code data with a predefined scale. For example, the fault code data can take the form of a table, where each row corresponds to the same time interval (e.g., 0.16 seconds) and includes an indication that the fault code is valid (e.g., value 1) or invalid (e.g., value 0) during the time interval. In this way, the fault diagnosis computing device 106 can remove a predetermined number of rows from the table, such as a predetermined number of rows (e.g., one hundred rows) corresponding to the time period or another time period during which the aircraft is expected to be started at the pre-flight startup procedure. Sensor data may also be represented in a similarly arranged table, such as a table in which the value in a given row is the average voltage measurement present during the time interval defined by the row.

[0062] Table 1 below shows an exemplary fault code data table (where 1=valid, 0=invalid for status), and Table 2 below shows an exemplary sensor data table (where values ​​in the right column are expressed in volts (V)).

[0063] Fault Codes state 3564 1 3564 1 3564 0 765 1 765 0

[0064] Table 1

[0065] sensor Sensor value (V) Electrical bus 2 3 Electrical bus 2 2 Electrical bus 2 7 Electrical bus 2 7 Electrical bus 2 3

[0066] Table 2

[0067] In addition, in other examples, data recorded during certain parts of the flight of the aircraft may be more helpful or more desirable in some cases than other parts. Thus, in some embodiments, the fault diagnosis computing device 106 can remove voltage measurements obtained during a time period when the rotational speed of the rotor of the aircraft exceeds a threshold number of revolutions per minute from the sensor data. Additionally or alternatively, the fault diagnosis computing device 106 can remove fault codes that are signaled during a time period when the rotational speed of the rotor of the aircraft exceeds a threshold number of revolutions per minute from the fault code data. For example, the threshold number of revolutions per minute can be a value selected from a range of 50 to 110 revolutions per minute. Other exemplary filtering criteria and associated thresholds are also possible, which can additionally or alternatively consider the rotational speed of the rotor of the aircraft 100. For example, one such criterion can involve the landing gear of the aircraft 100. For example, there may be a value associated with the position of the landing gear, where the value is zero when the landing gear is not lowered. Thus, the fault diagnosis computing device 106 can remove the data portion obtained / signaled when the landing gear value is zero. Other examples are also possible. In some cases, it may be useful to select a threshold related to the flight of aircraft 100 because it may be desirable to remove any data for which aircraft 100 was not in flight, but in other cases, data not obtained during flight may also be considered.

[0068] After obtaining the sensor data, the fault diagnosis computing device 106 can calculate a voltage threshold value, which indicates an abnormal voltage measurement value in the sensor data when exceeded by the voltage measurement value. In an example, in order to calculate the voltage threshold value, the fault diagnosis computing device 106 can calculate the mean value and standard deviation of the voltage measurement values ​​obtained over time relative to the aircraft 100. Specifically, the mean value and standard deviation can be calculated using the voltage measurement values ​​from a selected period of time (e.g., the past day, week) or perhaps since the first flight of the aircraft. The voltage threshold value can be a certain number of standard deviations away from the average value, and the voltage measurement values ​​falling outside this number can be considered abnormal. For example, if the voltage measurement value is more than one standard deviation away from the average value, it is considered abnormal. Other examples are also possible. In other examples, the voltage threshold value can be an actual voltage value (e.g., 5 volts) instead of a certain number of standard deviations.

[0069] Along these lines, the fault diagnosis computing device 106 may use the mean and standard deviation as a way to encode the sensor data so that the fault diagnosis computing device 106 may more easily identify events from the sensor data. For example, for each value P(i), the voltage measurements are taken at a predetermined interval (e.g., for each value in each row of the table), the fault diagnosis computing device 106 may calculate the encoded sensor value according to the following equation 1: where μT is the average value of aircraft 100 (i.e., the tail number of aircraft 100), and σ T is the standard deviation of aircraft 100.

[0070]

[0071] Using Equation 1, subtract the mean from the values ​​in the interval and divide the result by the standard deviation to find how much the value deviates positively or negatively from the mean. Remove any decimals from the result to convert the number of deviations to an integer value. For example, using Equation 1 on Table 2 would result in Table 3. The encoded sensor values ​​in Table 3 are not in volts.

[0072] sensor Encoded sensor value Electrical bus 2 0 Electrical bus 2 0 Electrical bus 2 1 Electrical bus 2 1 Electrical bus 2 0

[0073] Table 3

[0074] The encoded sensor value can then be compared to a voltage threshold. For example, when the voltage threshold is zero, the encoded sensor value must be 1 or greater to be considered abnormal. In some embodiments, the fault diagnosis computing device 106 may not calculate a voltage threshold, but is configured to have 1.

[0075] Then, the fault diagnosis computing device 106 generates a set of events from the sensor data and the fault code data. As described above, each event in the set of events is an event that occurs within a time interval, at which (i) the sensor data indicates an abnormal voltage measurement or (ii) a fault code associated with a particular aircraft subsystem of the plurality of aircraft subsystems is signaled. For example, with respect to Table 3, the fault diagnosis computing device 106 can generate events for time intervals in which the encoded sensor value is 1. And for example, with respect to Table 1, the fault diagnosis computing device 106 can generate events for any time interval in which a given fault code is valid.

[0076] For each event generated, the fault diagnosis computing device 106 can mark the event with a start time and an end time, such as a start time and an end time relative to the start of the flight, or a start time and an end time relative to a different time reference point. In addition, each event generated by the fault diagnosis computing device 106 can take the form of a value and the associated time interval in which the value occurs. For events involving abnormal voltage measurements, the value can take the form of a sensor value encoded during the time interval, as described above. For events involving fault codes, the value can take the form of a fault code signaled during the time interval.

[0077] In addition, the fault diagnostic computing device 106 may store in the data storage 120 or access from a remote data storage reference data that associates the fault codes with corresponding aircraft subsystems on the aircraft 100, and may reference the reference data to determine the aircraft subsystem associated with the fault code for each event involving the fault code, and tag the event with the appropriate aircraft subsystem. This may assist in constructing and displaying the temporal precursor network and summary table 126, as one or both may include a label for each event so that a user can see the aircraft subsystem associated with the event.

[0078] Then, the fault diagnosis computing device 106 calculates the value of the statistical correlation between the group of events. The value of the statistical correlation (which is a form of "mutual information" mentioned in information theory) indicates the possibility that the group of events is a causal chain of events that occur in time order. In other words, the value of the statistical correlation indicates the strength of association between events. The value of the statistical correlation can be used to construct a temporal precursor network and show the causal relationship between events, such as which fault code precedes and may cause another fault code and / or which fault code precedes and may cause abnormal voltage measurement. Therefore, this can indicate which aircraft subsystems may be interacting or otherwise affecting each other.

[0079] In this example, the value of the statistical correlation between k events M(A1, A2, ..., A k ) can be expressed by Equation 2, where A1, A2, …, A k is each different event, m is the total number of events that occurred with respect to the aircraft 100 in all flights of the aircraft 100, N(A1, A2, ..., A k ) is a sequence in which event A1 has a start time that is more than a predetermined threshold period of time (e.g., the time threshold discussed above) before the start time of event A2 and event A2 has a start time that is before the start time of the subsequent event, and so on (until event A k ), N(A1) is the total number of times event A1 occurs, N(A2) is the total number of times event A2 occurs, and so on.

[0080]

[0081] For example, if the value of the statistical correlation is calculated between a pair of events (ie, a first event and a second event), where k is equal to 2, then Equation 3 can be used, where the first event is represented as A and the second event is represented by B (rather than A2).

[0082]

[0083] The fault diagnosis computing device 106 then compares the value of the statistical correlation of a set of events with a predetermined threshold, and the comparison indicates the likelihood that the set of events is a causally related event that occurs in temporal order. For example, M(A, B) being less than a threshold of 1 may indicate that event A is unlikely to precede event B, and that the two events are unlikely to be associated. Therefore, this may indicate that two events occurring within a predetermined time period of each other are random chance and not statistically significant. On the other hand, M(A, B) being greater than 1 may indicate the likelihood that the two events have a stronger association, but, depending on the value, the association may not be strong. For example, a value between 1 and 10 may be a weaker association, while a value greater than 10 (or much higher, such as 10,000) may indicate a stronger association.

[0084] Then, the fault diagnosis computing device 106 uses the comparison of the value of statistical correlation with a predetermined threshold value (e.g., 1) as the basis for constructing a time precursor network. When the fault diagnosis computing device 106 determines that the value of the statistical correlation between the group of events is less than a predetermined threshold value, the group of events may not be added to the time precursor network. On the other hand, when the fault diagnosis computing device 106 determines that the value of the statistical correlation between the group of events exceeds a predetermined threshold value, the group of events is added to the time precursor network and is represented as a sequence of related events that occur in chronological order. Because the value of statistical correlation is related to the possibility of event correlation, the time sequence represented by the time precursor network may not be the time sequence of confirmation, but the time sequence that may have occurred, so it is important enough to diagnose the cause of the associated event. The user (such as a maintenance personnel or operator of the aircraft 100) can be more confident that the information in the time precursor network indicates a causal event, rather than a random, irrelevant event without causality.

[0085] In an example, a set of events may be added to a temporal precursor network as a sequence of related events in the form of connected nodes. For example, each event of the sequence of related events may be represented as a separate node and may be connected by links indicating the chronological order in which the sequence of related events occurred. The temporal precursor network may be used as a visualization tool to indicate which aircraft subsystems interact with each other. Additionally, in some cases, the temporal precursor network may be used to assist in fault diagnosis.

[0086] Figure 2 An exemplary temporal precursor network 128 is shown in accordance with an exemplary embodiment. Figure 2The temporal precursor network 128 shown in FIG. 1 only includes pairs of related events, but in other embodiments, the temporal precursor network can indicate a sequence of two or more related events. An example set of events represented as a sequence of related events includes nodes 130 and 132, which are connected by arrows 134, which show that the event represented by node 130 may have statistically caused the event represented by node 132. For example, the thickness, color, or other characteristics of the displayed arrows can be changed according to the value of the statistical correlation of a given set of events. For example, when the value of the statistical correlation is high, the arrow can be thicker, or when the value of the statistical correlation is low, the arrow can be thinner.

[0087] In some embodiments, values ​​of statistical correlations between two or more events may be included in the temporal precursor network 128. Also shown in the temporal precursor network 128 is a subsystem table 136, which may be included as part of the temporal precursor network in some embodiments. The subsystem table 136 may include a list of aircraft subsystems and / or electrical buses and, for each aircraft subsystem or electrical bus, an indication of how many events on the aircraft 100 are associated with that aircraft subsystem or electrical bus (e.g., a percentage (%), as shown). The temporal precursor network 128 may also include other information.

[0088] At some point before, during, or after constructing the temporal precursor network, the fault diagnosis computing device 106 indexes both the sequence of related events and the value of the statistical correlation between the set of events in the summary table 126 and separately from the database and / or table storing the sensor data and the fault code data. The fault diagnosis computing device 106 then controls the display device 104 to display a visual representation of the summary table 126 and the temporal precursor network 128 on a graphical user interface, such as Figure 2 The visual representation shown in .

[0089] Both the time precursor network 128 and the summary table 126 are provided to maintenance personnel, operators of the aircraft 100, or other users in different ways so that potential event causal chains can be viewed and reliably evaluated. For example, the time precursor network 128 visually depicts events that are statistically likely to cause other events, and visually depicts which aircraft subsystems are experiencing more events than other events. Using the time precursor network 128, a user can track events between multiple different aircraft subsystems and / or electrical buses. For example, a few seconds after a flight control subsystem fault code is signaled, fault codes in two other subsystems may be signaled. However, only one of these two subsequent fault codes may be statistically dependent on the flight control subsystem fault code, and the time precursor network 128 can depict the causal chain, allowing the user to confidently eliminate the possibility that another of the latter two fault codes is caused by the flight control subsystem fault code. Other examples are also possible.

[0090] In addition, the summary table 126 provides a brief summary of events with a threshold-high statistical relevance. Using the summary table 126, a user can view and browse the summary table 126 to quickly search and identify interesting event chains, and reliably eliminate event chains that are shown to be statistically related but may not actually be causally related. The user's ability to discern which event chains are in fact causal chains with root causes can even be further assisted by the fault diagnosis computing device 106 indexing additional measures in the summary table 126, examples of which will be discussed in more detail below.

[0091] The action of indexing may involve organizing the sequence of related events and the values ​​of statistical correlation into rows and columns of summary table 126. In addition, once summary table 126 is created, fault diagnosis computing device 106 may maintain summary table 126 in data storage 120, and the action of indexing may involve updating summary table 126 to add each row with the newly identified sequence of events and the corresponding values ​​of statistical correlation. In some cases, the sequence of events may occur multiple times, but may change in various ways. For example, in one case, event A may occur and event B may occur five seconds later, and at a later time (e.g., during the next flight of aircraft 100) event B may follow event A for thirty seconds, which may affect the value of statistical correlation between the two events. In this case, the action of indexing may involve updating the value of statistical correlation of the existing sequence of events in summary table 126. Other examples are also possible.

[0092] The sequence of related events may be represented in summary table 126 using identifiers. For example, if the event is a sensor measurement anomaly, fault diagnostic computing device 106 may include in summary table 126 the identifier of the electrical bus and / or sensor that acquired the anomaly measurement. If fault diagnostic computing device 106 has stored in data storage 120 the aircraft subsystems that feed into one or more electrical buses (in question), fault diagnostic computing device 106 may also include the identifier(s) of the one or more aircraft subsystems. Additionally, in some implementations, fault diagnostic computing device 106 may include in summary table 126 how many standard deviations of the voltage measurements are below or above the mean.

[0093] As another example, if the event is a signaled fault code, the fault diagnostic computing device 106 may include an identifier of the fault code and / or an identifier of the aircraft subsystem associated with the fault code in the summary table 126. As described above, the fault diagnostic computing device 106 may reference the reference data to determine which aircraft subsystem the fault code corresponds to.

[0094] In some embodiments, the fault diagnosis computing device 106 can rank the sequence of related events based on one or more factors (e.g., the value of statistical correlation). For example, if the first sequence of related events has a higher value of statistical correlation than the second sequence of related events, the first sequence of related events can be ranked higher than the second sequence of related events. The summary table 126 may include an indication of the ranking, such as by a number (e.g., 1, 2, 3, etc.) or other text, or may include a sequence of the highest-ranked related events at the first top row of the summary table 126 and a sequence of related events of lower ranking in lower rows. Other examples are also possible.

[0095] In some embodiments, a user may select a row of the summary table 126, which may trigger the fault diagnosis computing device 106 to display supplemental information associated with the sequence of related events corresponding to the selected row. For example, the fault diagnosis computing device 106 may be triggered to display the temporal precursor network 128 and highlight or otherwise emphasize the nodes in the temporal precursor network 128 corresponding to the selected sequence of related events. Additionally or alternatively, the fault diagnosis computing device 106 may be triggered to retrieve and display additional information about the aircraft subsystems and / or electrical buses involved in the sequence of related events, such as visual indications (e.g., graphs or maps) and / or other information of the locations of the aircraft subsystems and / or electrical buses within the aircraft 100. Other examples are also possible.

[0096] The fault diagnosis computing device 106 may also index and display other useful information in the summary table 126, such as the total number of flights of the aircraft 100 and other aircraft (i.e., other tail numbers) during the sequence of related events (e.g., "# of flights", as shown in Tables 4 and 5) and / or the total number of aircraft for which the sequence of related events occurred (e.g., "# of tails", as shown in Tables 4 and 5). Thus, the act of indexing may also involve updating the summary table 126 to include a new column indicating these two totals, or adding these two totals to the existing columns corresponding to the two types of totals. For convenience, the fault diagnosis computing device 106 may refer to the reference data stored in the data storage 120 to determine the total number of flights of the aircraft and other aircraft during the sequence of related events and / or the total number of aircraft for which the sequence of related events occurred, and then update the summary table 126 to include those numbers. This information and other information are shown in Table 4 by way of example.

[0097]

[0098] Table 4

[0099] Fault diagnosis computing device 106 may also index and display in summary table 126 the total number of flights of aircraft 100 and other aircraft during which the sequence of related events occurred when the environmental conditions (e.g., high winds, ice, thunderstorms) existed, and / or the total number of aircraft during which the sequence of related events occurred when the environmental conditions existed. This information is shown in Table 5 by way of example.

[0100]

[0101] Table 5

[0102] In the manner discussed above and in other ways, summary table 126 can be kept up to date and can indicate to a user, such as a maintenance personnel or pilot, the consistency of occurrence of a sequence of related events. In other words, if the correlated events are threshold-likely to be statistically correlated and exhibited over many tail numbers and flights of each tail number, this may indicate to the maintenance personnel, aircraft operator, or other user that the correlated events may be inherent to the underlying hardware / software of aircraft 100, rather than tail-specific or environment-specific events. On the other hand, the correlated events may be tail-specific or environment-specific and not inherent to the design of the underlying hardware / software, in which case the number of different tail numbers in which the correlated events occur may be small.

[0103] Another important statistic that helps users assess causal relationships between events is the time period (or "time lag") that elapses between any two consecutive events in a sequence of related events. This statistic advantageously characterizes the ability to predict the occurrence of a sequence of events.

[0104] For convenience, given two events of a sequence of events, the fault diagnosis computing device 106 may calculate the average (i.e., mean) of the time period between the first event and the second event over the past occurrences of the sequence of related events, store the average (i.e., mean) in the summary table 126, and display it in the summary table 126. The fault diagnosis computing device 106 may also calculate the standard deviation of the time period between the first event and the second event over the past occurrences of the sequence of related events, store the standard deviation in the summary table 126, and display it in the summary table 126. Comparing the average and the standard deviation here may yield valuable insights into the predictability of the two events.

[0105] Consider an example in which there are three events, A, B, and C. The fault diagnosis computing device 106 may calculate the mean μ1 and standard deviation σ1 of the time lag between the start times of events A and B for past occurrences of A and B, and (ii) the mean μ2 and standard deviation σ2 of the time lag between the start times of events B and C for past occurrences of B and C. If the standard deviation for any given pair of events is, for example, three times or more of the mean, this indicates that there is a large amount of deviation in the time lag between these events, and therefore the event is not inherent to the aircraft 100. On the other hand, if the standard deviation is much smaller (e.g., one-third of the mean), this may indicate that the deviation in the time lag between these events is much smaller, and therefore it may be predicted that one of the two events may cause the other event. However, in the case where both the mean and the standard deviation are very small (e.g., a mean of 3 and a standard deviation of 6), the deviation from the mean may not be significant, but rather may be attributed to a defect in the aircraft 100. Table 6 includes example means and standard deviations for two different sequences of three events.

[0106] Event A Event B Event C Statistical correlation <![CDATA[μ1]]> <![CDATA[σ1]]> <![CDATA[μ2]]> <![CDATA[σ2]]> 7049 6521 1554 199.2 57.2 10.5 60 20 Electrical bus 3 1771 6480 275.8 3.13 6.01 28 218.6

[0107] Table 6

[0108] In the preceding examples and other possible examples, event correlation may be prioritized and used for diagnostic and troubleshooting purposes.

[0109] In some embodiments, for example, given a set of events, the fault diagnosis computing device 106 may perform additional calculations to obtain additional information detailing the likelihood of a causal relationship between the events. For example, consider a set of three events: A, B, and C, where A occurs before B and B occurs before C. Given this set of events, there may be three main causal possibilities: (i) A causes B and B causes C; (ii) A causes B and A causes C; and (iii) A causes B, B causes C, and A causes C. For each of these possibilities, the fault diagnosis computing device 106 may add the value of the statistical correlation of the pair and divide by the number of pairs to obtain an additional value of the statistical correlation. For example, for the first possibility, the additional value of the statistical correlation is equal to (M(A,B)+M(B,C)) / 2. For the second possibility, the additional value of the statistical correlation is equal to (M(A,B)+M(A,C)) / 2. And for the third possibility, the additional value of the statistical correlation is equal to (M(A,B)+M(A,C)+M(B,C)) / 3. Then, the fault diagnosis computing device 106 determines which of the three additional values ​​of the statistical correlation is the highest and can be stored in the summary table 126 and / or display the highest value of the three. The highest of the three values ​​is attributed to the most likely causal possibility, which can further help identify the root cause in the group of events. Other examples are also possible, such as examples where there are longer chains of events, such as four or more events.

[0110] Figure 3 A flow chart is shown of an example method 200 of identifying correlated events in an aircraft in accordance with an illustrative embodiment. Figure 3 The method 200 shown in FIG. 1 presents a method that can be used with Figure 1 An example of a method for use with the aircraft 100 shown in FIG. 1 or an example of a method for use with a component of the aircraft 100. In addition, a device or system such as the system 102 or the fault diagnosis computing device 106 may be used or configured to perform Figure 3 . In some instances, the components of the device and / or system may be configured to perform functions so that the components are actually configured and structured (with hardware and / or software) to achieve such performance. In other examples, the components of the device and / or system may be arranged to be suitable for, capable of, or adapted to perform functions, such as when operated in a particular manner. Method 200 may include one or more operations, functions, or actions, as shown in one or more of blocks 202-216. Although blocks are shown in a continuous order, these blocks may also be executed in parallel, and / or in an order different from the order described herein. Moreover, various blocks may be combined into fewer blocks, divided into additional blocks, and / or removed based on the desired implementation.

[0111] It should be understood that for this and other processes and methods disclosed herein, the flow chart illustrates the functions and operations of a possible implementation of this example. In this regard, each box or a portion of each box can represent a module, a segment or a portion of a program code, and the program code includes one or more instructions executable by a processor for implementing a specific logical function or step in the process. The program code can be stored on any type of computer-readable medium or data storage, such as a storage device including a disk or a hard drive. In addition, the program code can be encoded in a machine-readable format on a computer-readable storage medium, or encoded on other non-transient media or products. For example, a computer-readable medium may include a non-transient computer-readable medium or memory, such as a computer-readable medium for storing data in a short period of time, such as a register memory, a processor cache, and a random access memory (RAM). For example, a computer-readable medium may also include a non-transient medium, such as a secondary or persistent long-term storage, such as a read-only memory (ROM), an optical disk or a disk, a compact disk read-only memory (CD-ROM). The computer-readable medium may also be any other volatile or non-volatile storage system. For example, a computer-readable medium may be considered to be a tangible computer-readable storage medium.

[0112] in addition, Figure 3 Each box or part of each box in and other processes and methods disclosed herein may represent circuits that are wired to perform specific logical functions in the process. Alternative implementations are included within the scope of the examples of the present disclosure, where functions can be performed from the order shown or discussed, including substantially simultaneously or in reverse order, depending on the functions involved, as reasonably understood by those skilled in the art.

[0113] At block 202 , method 200 includes obtaining sensor data including voltage measurements over time of at least one electrical bus on an aircraft.

[0114] At block 204 , method 200 includes obtaining fault code data including fault codes signaled over time by a plurality of aircraft subsystems on the aircraft, wherein each of the fault codes is associated with a corresponding aircraft subsystem of the plurality of aircraft subsystems that causes the corresponding aircraft subsystem to output the fault code upon detection of an error.

[0115] At block 206 , method 200 includes calculating a voltage threshold that, when exceeded by the voltage measurement, indicates an abnormal voltage measurement in the sensor data.

[0116] Figure 4 A flow chart of an example method for performing the calculations shown in block 206 is shown in accordance with an example embodiment. In particular, Figure 4 An example method of performing the calculations shown in block 206, wherein the sequence of related events includes a first event and a second event. At block 218, the functionality includes: calculating the value of the statistical correlation between the set of events based on (i) a total number of events in the set of events, (ii) a total number of past occurrences of the sequence of related events in which the first event preceded the second event by more than a predetermined threshold period of time, (iii) a total number of times the first event has occurred, and (iv) a total number of times the second event has occurred.

[0117] refer to Figure 3 , at box 208, method 200 includes: generating a set of events from the sensor data and the fault code data, wherein each event in the set of events occurs within a time interval at which (i) the sensor data indicates the abnormal voltage measurement or (ii) a fault code associated with a particular aircraft subsystem of the plurality of aircraft subsystems is signaled.

[0118] In some implementations, some functions may be performed prior to generating the set of data from the sensor data and the fault code data. Figure 5 A flow chart is shown of an example method for use with method 200 in accordance with an exemplary embodiment. Prior to generating the set of data from sensor data and fault code data, functions include removing from the sensor data voltage measurements obtained during a time period when the rotational speed of the rotor of the aircraft exceeds a threshold rpm; and removing from the fault code data fault codes signaled during a time period when the rotational speed of the rotor of the aircraft exceeds the threshold rpm, as shown in blocks 220 and 222.

[0119] refer to Figure 3 At block 210 , the method 200 includes calculating a value of a statistical correlation between the set of events, wherein the value of the statistical correlation between the set of events indicates a likelihood that the set of events are causally related events that occur in temporal sequence.

[0120] At block 212 , method 200 includes constructing a temporal precursor network representing the set of events as a sequence of related events and further representing the timing of occurrence of the sequence of related events based on the value of the statistical correlation between the set of events exceeding a predetermined threshold.

[0121] At block 214 , the method 200 includes indexing the sequence of related events and the value of the statistical correlation between the set of events in a summary table stored in memory and separate from the sensor data and the fault code data.

[0122] Figure 6 A flow chart is shown of an example method for performing indexing as shown in block 214 in accordance with an exemplary embodiment. At block 224, the functionality includes updating the summary table to indicate the total number of flights of the aircraft and the total number of flights of other aircraft during the period in which the sequence of related events occurred and the total number of aircraft in which the sequence of related events occurred.

[0123] refer to Figure 3 At block 216 , the method 200 includes controlling a display device to display the summary table and a visual representation of the temporal precursor network on a graphical user interface.

[0124] Figure 7 A flow chart is shown of an example method for performing control as shown in block 216 according to an example embodiment. At block 226, the function includes controlling the display device to depict, in the visual representation, (i) each event of the sequence of related events as a node and (ii) the sequence of related events as two or more nodes connected by a link, the link indicating a timing in which the sequence of related events occurs.

[0125] Figure 8 A flow chart of an example method for performing control as shown in block 216 according to an exemplary embodiment is shown. Specifically, Figure 8 An example method for performing control as shown in block 216, wherein the sequence of related events includes a first event and a second event, wherein the sensor data indicates the abnormal voltage measurement, and wherein the fault code associated with the particular aircraft subsystem of the plurality of aircraft subsystems is signaled. At block 228, the function includes controlling the display device to display, on the graphical user interface, in the summary table: (i) for the first event, a first identifier of an electrical bus of the at least one electrical bus on which the abnormal voltage measurement exists; (ii) for the second event, a second identifier of the fault code and a third identifier of the particular aircraft subsystem; and (iii) displaying a value of a statistical correlation between the first event and the second event.

[0126] Fig. 9 A flow chart is shown of an example method according to an exemplary implementation for performing control as shown in block 216. At block 230, the function includes controlling the display device to display, on the graphical user interface, in the summary table, a fourth identifier of an aircraft subsystem of the plurality of aircraft subsystems associated with the abnormal voltage measurement.

[0127] Fig.10A flow chart of an example method for performing control as shown in block 216 according to an exemplary embodiment is shown. Specifically, Fig.10 An example method of performing control as shown in block 216, wherein the sequence of related events includes at least three events. At block 232, the function includes controlling the display device to display on the graphical user interface, in the summary table, (i) a corresponding identifier for each of the at least three events and (ii) the value of the statistical correlation between the at least three events.

[0128] Fig.11 A flowchart of another example method for use with method 200 according to an exemplary embodiment is shown. Specifically, Fig.11 An example method is provided for use with method 200, wherein a sequence of related events includes a first event and a second event. At block 234, functionality includes calculating an average of a time period between the first event and the second event at past occurrences of the sequence of related events. And at block 236, functionality includes calculating a standard deviation of the time period between the first event and the second event at the past occurrences of the sequence of related events.

[0129] Further, Fig.11 A flow chart is shown of another example method for performing control as shown in block 216 according to an exemplary embodiment. At block 238, the function includes controlling the display device to display, on the graphical user interface, in the summary table (i) an identifier of the first event, (ii) an identifier of the second event, (iii) a value of a statistical correlation between the first event and the second event, (iv) the average value of the time period between the first event and the second event at the past occurrences of the sequence of related events, and (v) the standard deviation of the time period between the first event and the second event at the past occurrences of the sequence of related events.

[0130] Fig.12 A flow chart of another example method for use with method 200 according to an exemplary embodiment is shown. At box 240, the function includes referencing reference data stored in the memory to determine the total number of flights of the aircraft during the period when the sequence of related events occurred and the total number of flights of other aircraft and the total number of aircraft in which the sequence of related events occurred. And at box 242, the function includes controlling the display device to display the total number of flights of the aircraft during the period when the sequence of related events occurred and the total number of flights of the other aircraft and the total number of aircraft in which the sequence of related events occurred.

[0131] Fig.13A flow chart of another example method for use with method 200 is shown in accordance with an exemplary embodiment. At block 244, the functionality includes referencing reference data stored in the memory to determine a total number of flights of the aircraft and a total number of flights of other aircraft during the period when the sequence of related events occurs when the environmental condition exists, and a total number of aircraft for which the sequence of related events occurs when the environmental condition exists. And at block 246, the functionality includes controlling the display device to display the total number of flights of the aircraft and the total number of flights of the other aircraft during the period when the sequence of related events occurs when the environmental condition exists, and the total number of aircraft for which the sequence of related events occurs when the environmental condition exists.

[0132] The terms "substantially" and "approximately" as used herein mean that the features, parameters or values ​​described need not be achieved precisely, but deviations or changes (including, for example, tolerances, measurement errors, measurement precision limitations and other factors known to those skilled in the art) may occur by amounts that do not negate the effects that the features are intended to provide.

[0133] Different examples of the (one or more) systems, (one or more) devices, and (one or more) methods disclosed herein include various components, features, and functions. It should be understood that the various examples of the (one or more) systems, (one or more) devices, and (one or more) methods disclosed herein may include any components, features, and functions of any other examples of the (one or more) systems, (one or more) devices, and (one or more) methods disclosed herein in any combination or any subcombination, and all of these possibilities are within the scope of the present disclosure.

[0134] Furthermore, the present disclosure includes embodiments according to the following clauses:

[0135] Clause 1. A method for identifying correlated events in an aircraft, the method comprising:

[0136] obtaining sensor data comprising voltage measurements over time of at least one electrical bus on the aircraft;

[0137] obtaining fault code data, the fault code data comprising fault codes signaled over time by a plurality of aircraft subsystems on the aircraft, wherein each of the fault codes is associated with a corresponding aircraft subsystem of the plurality of aircraft subsystems, causing the corresponding aircraft subsystem to output the fault code upon detection of an error;

[0138] calculating a voltage threshold, and indicating an abnormal voltage measurement in the sensor data when the voltage measurement exceeds the voltage threshold;

[0139] generating a set of events from the sensor data and the fault code data, wherein each event in the set of events occurs within a time interval over which (i) the sensor data indicates the abnormal voltage measurement or (ii) a fault code associated with a particular aircraft subsystem of the plurality of aircraft subsystems is signaled;

[0140] calculating a value of the statistical correlation between the set of events, wherein the value of the statistical correlation between the set of events indicates a likelihood that the set of events are causally related events occurring in time sequence;

[0141] constructing a temporal precursor network representing the set of events as a sequence of related events and further representing the temporal order of occurrence of the sequence of related events based on the value of the statistical correlation between the set of events exceeding a predetermined threshold;

[0142] indexing said values ​​of statistical correlation between said sequence of related events and said set of events in a summary table stored in memory and separate from said sensor data and said fault code data; and

[0143] A display device is controlled to display the summary table and a visual representation of the temporal precursor network on a graphical user interface.

[0144] Clause 2. The method of clause 1, wherein the sequence of related events comprises a first event and a second event, and

[0145] Calculating the value of the statistical correlation between the set of events includes: calculating the value of the statistical correlation between the set of events based on (i) the total number of events in the set of events, (ii) the total number of past occurrences of the sequence of related events in which the first event preceded the second event by more than a predetermined threshold time period, (iii) the total number of times the first event has occurred, and (iv) the total number of times the second event has occurred.

[0146] Clause 3. The method according to clause 1, further comprising:

[0147] Prior to generating the set of events from the sensor data and the fault code data:

[0148] removing from the sensor data voltage measurements obtained during time periods when a rotational speed of a rotor of the aircraft exceeds a threshold rpm; and

[0149] Fault codes signaled during a time period when the rotational speed of the rotor of the aircraft exceeds the threshold rpm are removed from the fault code data.

[0150] Clause 4. A method according to clause 1, wherein in the summary table stored in the memory and separate from the sensor data and the fault code data, the value indexing the sequence of related events and the statistical correlation between the set of events includes: updating the summary table to indicate the total number of flights of the aircraft and the total number of flights of other aircraft during the period when the sequence of related events occurred and the total number of aircraft in which the sequence of related events occurred.

[0151] Clause 5. A method according to clause 1, wherein controlling the display device to display the summary table and the visual representation of the temporal precursor network on the graphical user interface includes: controlling the display device to depict in the visual representation (i) each event of the sequence of related events as a node and (ii) depicting the sequence of related events as two or more nodes connected by links, which indicate the timing of occurrence of the sequence of related events.

[0152] Clause 6. The method of clause 1, wherein the sequence of related events includes a first event in which the sensor data indicates the abnormal voltage measurement and a second event in which the fault code associated with the particular aircraft subsystem of the plurality of aircraft subsystems is signaled, and

[0153] Wherein controlling the display device to display the summary table and the visual representation of the time precursor network on the graphical user interface includes: controlling the display device to display on the graphical user interface, in the summary table: (i) for the first event, a first identifier of an electrical bus of the at least one electrical bus on which the abnormal voltage measurement exists; (ii) for the second event, a second identifier of the fault code and a third identifier of the specific aircraft subsystem; and (iii) displaying a value of a statistical correlation between the first event and the second event.

[0154] Clause 7. A method according to Clause 6, wherein controlling the display device to display the summary table and the visual representation of the time precursor network on the graphical user interface includes: controlling the display device to display a fourth identifier of an aircraft subsystem of the multiple aircraft subsystems associated with the abnormal voltage measurement in the summary table on the graphical user interface.

[0155] Clause 8. The method of clause 1, wherein the sequence of related events comprises at least three events, and

[0156] Wherein controlling the display device to display the summary table and the visual representation of the temporal precursor network on the graphical user interface includes: controlling the display device to display on the graphical user interface, in the summary table (i) a corresponding identifier for each of the at least three events and (ii) the value of the statistical correlation between the at least three events.

[0157] Clause 9. The method of clause 1, wherein the sequence of related events includes a first event and a second event, the method further comprising:

[0158] calculating an average of the time period between the first event and the second event for past occurrences of the sequence of related events; and

[0159] calculating a standard deviation of the time period between the first event and the second event of the past occurrences of the sequence of related events,

[0160] Wherein controlling the display device to display the summary table and the visual representation of the temporal precursor network on the graphical user interface includes: controlling the display device to display on the graphical user interface, in the summary table (i) an identifier of the first event, (ii) an identifier of the second event, (iii) a value of the statistical correlation between the first event and the second event, (iv) the average value of the time period between the first event and the second event when the sequence of related events occurred in the past, and (v) the standard deviation of the time period between the first event and the second event when the sequence of related events occurred in the past.

[0161] Clause 10. The method according to clause 1, further comprising:

[0162] referencing the reference data stored in the memory to determine the total number of flights of the aircraft during the period in which the sequence of related events occurred and the total number of flights of other aircraft and the total number of aircraft in which the sequence of related events occurred; and

[0163] The display device is controlled to display the total number of flights of the aircraft and the total number of flights of the other aircraft during the period in which the sequence of related events occurs, and the total number of aircraft in which the sequence of related events occurs.

[0164] Clause 11. The method according to clause 1, further comprising:

[0165] referencing reference data stored in the memory to determine a total number of flights of the aircraft and a total number of flights of other aircraft during which the sequence of related events occurred when the environmental condition existed and a total number of aircraft during which the sequence of related events occurred when the environmental condition existed; and

[0166] The display device is controlled to display the total number of flights of the aircraft and the total number of flights of the other aircraft during which the sequence of related events occurs when the environmental condition exists, and the total number of aircraft in which the sequence of related events occurs when the environmental condition exists.

[0167] Clause 12. A non-transitory computer-readable medium having instructions stored thereon that, when executed by one or more processors of a computing device, cause the computing device to perform operations comprising:

[0168] obtaining sensor data comprising voltage measurements over time of at least one electrical bus on the aircraft;

[0169] obtaining fault code data, the fault code data comprising fault codes signaled over time by a plurality of aircraft subsystems on the aircraft, wherein each of the fault codes is associated with a corresponding aircraft subsystem of the plurality of aircraft subsystems, causing the corresponding aircraft subsystem to output the fault code upon detection of an error;

[0170] calculating a voltage threshold, and indicating an abnormal voltage measurement in the sensor data when the voltage measurement exceeds the voltage threshold;

[0171] generating a set of events from the sensor data and the fault code data, wherein each event in the set of events occurs within a time interval over which (i) the sensor data indicates the abnormal voltage measurement or (ii) a fault code associated with a particular aircraft subsystem of the plurality of aircraft subsystems is signaled;

[0172] calculating a value of the statistical correlation between the set of events, wherein the value of the statistical correlation between the set of events indicates a likelihood that the set of events are causally related events occurring in time sequence;

[0173] constructing a temporal precursor network representing the set of events as a sequence of related events and further representing the temporal order of occurrence of the sequence of related events based on the value of the statistical correlation between the set of events exceeding a predetermined threshold;

[0174] indexing said values ​​of statistical correlation between said sequence of related events and said set of events in a summary table stored in memory and separate from said sensor data and said fault code data; and

[0175] A display device is controlled to display the summary table and a visual representation of the temporal precursor network on a graphical user interface.

[0176] Clause 13. The non-transitory computer-readable medium of Clause 12, wherein the operations further comprise:

[0177] Prior to generating the set of events from the sensor data and the fault code data:

[0178] removing from the sensor data voltage measurements obtained during time periods when a rotational speed of a rotor of the aircraft exceeds a threshold rpm; and

[0179] Fault codes signaled during a time period when the rotational speed of the rotor of the aircraft exceeds the threshold rpm are removed from the fault code data.

[0180] Clause 14. The non-transitory computer-readable medium of Clause 12, wherein the sequence of related events comprises a first event and a second event, and

[0181] wherein calculating the value of the statistical correlation between the set of events comprises: calculating the value of the statistical correlation between the set of events based on (i) the total number of events in the set of events, (ii) the total number of past occurrences of the sequence of related events where the first event preceded the second event by more than a predetermined threshold period of time, (iii) the total number of times the first event has occurred, and (iv) the total number of times the second event has occurred.

[0182] Clause 15. A non-transitory computer-readable medium according to clause 12, wherein in the summary table stored in the memory and separate from the sensor data and the fault code data, the value indexing the sequence of related events and the statistical correlation between the set of events includes: updating the summary table to indicate the total number of flights of the aircraft and the total number of flights of other aircraft during the period when the sequence of related events occurred and the total number of aircraft in which the sequence of related events occurred.

[0183] Clause 16. A non-transitory computer-readable medium according to clause 12, wherein controlling the display device to display the summary table and the visual representation of the temporal precursor network on the graphical user interface comprises: controlling the display device to depict each event of the sequence of related events as a node in the visual representation and (ii) depicting the sequence of related events as two or more nodes connected by links, which indicate the timing of occurrence of the sequence of related events.

[0184] Clause 17. A system comprising:

[0185] a plurality of aircraft subsystems on the aircraft; and

[0186] A fault diagnosis computing device having a processor and a memory, wherein the memory stores instructions executable by the processor to perform operations including:

[0187] obtaining sensor data comprising voltage measurements over time of at least one electrical bus on the aircraft;

[0188] obtaining fault code data, the fault code data comprising fault codes signaled by the plurality of aircraft subsystems over time, wherein each of the fault codes is associated with a corresponding aircraft subsystem of the plurality of aircraft subsystems, causing the corresponding aircraft subsystem to output the fault code upon detection of an error;

[0189] calculating a voltage threshold, and indicating an abnormal voltage measurement in the sensor data when the voltage measurement exceeds the voltage threshold;

[0190] generating a set of events from the sensor data and the fault code data, wherein each event in the set of events occurs within a time interval over which (i) the sensor data indicates the abnormal voltage measurement or (ii) a fault code associated with a particular aircraft subsystem of the plurality of aircraft subsystems is signaled;

[0191] calculating a value of the statistical correlation between the set of events, wherein the value of the statistical correlation between the set of events indicates a likelihood that the set of events are causally related events occurring in time sequence;

[0192] constructing a temporal precursor network representing the set of events as a sequence of related events and further representing the temporal order of occurrence of the sequence of related events based on the value of the statistical correlation between the set of events exceeding a predetermined threshold;

[0193] indexing said values ​​of statistical correlation between said sequence of correlated events and said set of events in a summary table stored in said memory and separate from said sensor data and said fault code data; and

[0194] A display device is controlled to display the summary table and a visual representation of the temporal precursor network on a graphical user interface.

[0195] Clause 18. The system of clause 17, wherein the sequence of related events includes a first event and a second event, and wherein the operations further comprise:

[0196] calculating an average of the time period between the first event and the second event for past occurrences of the sequence of related events; and

[0197] calculating a standard deviation of the time period between the first event and the second event of the past occurrences of the sequence of related events,

[0198] Wherein controlling the display device to display the summary table and the visual representation of the temporal precursor network on the graphical user interface includes: controlling the display device to display on the graphical user interface, in the summary table (i) an identifier of the first event, (ii) an identifier of the second event, (iii) a value of the statistical correlation between the first event and the second event, (iv) the average value of the time period between the first event and the second event when the sequence of related events occurred in the past, and (v) the standard deviation of the time period between the first event and the second event when the sequence of related events occurred in the past.

[0199] Clause 19. The system of clause 17, wherein the operations further comprise:

[0200] referencing the reference data stored in the memory to determine the total number of flights of the aircraft during the period in which the sequence of related events occurred and the total number of flights of other aircraft and the total number of aircraft in which the sequence of related events occurred; and

[0201] The display device is controlled to display the total number of flights of the aircraft and the total number of flights of the other aircraft during the period in which the sequence of related events occurs, and the total number of aircraft in which the sequence of related events occurs.

[0202] Clause 20. The system of clause 17, wherein the operations further comprise:

[0203] referencing reference data stored in the memory to determine a total number of flights of the aircraft and a total number of flights of other aircraft during which the sequence of related events occurred when the environmental condition existed and a total number of aircraft during which the sequence of related events occurred when the environmental condition existed; and

[0204] The display device is controlled to display the total number of flights of the aircraft and the total number of flights of the other aircraft during which the sequence of related events occurs when the environmental condition exists, and the total number of aircraft in which the sequence of related events occurs when the environmental condition exists.

[0205] The description of different advantageous arrangements has been given for the purpose of illustration and description, and is not intended to be exhaustive or limited to the examples of the disclosed form. Many modifications and variations will be apparent to those of ordinary skill in the art. In addition, different advantageous examples may describe different advantages compared to other advantageous examples. The selected examples or examples are selected and described in order to best explain the principles of the examples, practical applications, and to enable those of ordinary skill in the art to understand the various examples of the present disclosure with various modifications suitable for the intended specific use.

Claims

1. A method (200) for identifying a correlation event in an aircraft (100), the method comprising: (202) obtaining sensor data, the sensor data comprising voltage measurements over time of at least one electrical bus (112) on the aircraft (100); (204) obtaining fault code data, the fault code data comprising fault codes signaled over time by a plurality of aircraft subsystems (108) on the aircraft (100), wherein each of the fault codes is associated with a corresponding aircraft subsystem of the plurality of aircraft subsystems (108), causing the corresponding aircraft subsystem to output the fault code upon detection of an error; (206) calculating a voltage threshold, and indicating an abnormal voltage measurement in the sensor data when the voltage measurement exceeds the voltage threshold; (208) generating a set of events from the sensor data and the fault code data, wherein each event in the set of events occurs within a time interval over which (i) the sensor data indicates the abnormal voltage measurement or (ii) a fault code associated with a particular aircraft subsystem of the plurality of aircraft subsystems (108) is signaled; (210) calculating a value of the statistical correlation between the set of events, wherein the value of the statistical correlation between the set of events indicates a likelihood that the set of events are causally related events that occur in a temporal sequence; (212) constructing a temporal precursor network (128) based on the value of the statistical correlation between the set of events exceeding a predetermined threshold, which represents the set of events as a sequence of correlated events and further represents the timing of occurrence of the sequence of correlated events, wherein the sequence of correlated events includes a first event in which the sensor data indicates the abnormal voltage measurement and a second event in which the fault code associated with the particular aircraft subsystem of the plurality of aircraft subsystems is signaled; (214) indexing, in a summary table (126) stored in the memory (120) and separate from the sensor data and the fault code data, the values ​​of the statistical correlation between the sequence of related events and the set of events; as well as (216) controlling a display device (104) to display the summary table (126) and a visual representation of the time precursor network (128) on a graphical user interface, wherein controlling the display device to display the summary table on the graphical user interface comprises: controlling the display device to display on the graphical user interface, in the summary table: (i) for the first event, a first identifier of an electrical bus of the at least one electrical bus on which the abnormal voltage measurement exists; (ii) for the second event, a second identifier of the fault code and a third identifier of the specific aircraft subsystem; and (iii) displaying a value of a statistical correlation between the first event and the second event.

2. The method of claim 1 , wherein calculating the value of the statistical correlation between the set of events comprises: (218) Calculating the value of the statistical correlation between the set of events based on (i) the total number of events in the set of events, (ii) the total number of past occurrences of the sequence of related events in which the first event preceded the second event by more than a predetermined threshold time period, (iii) the total number of times the first event has occurred, and (iv) the total number of times the second event has occurred.

3. The method according to claim 1, further comprising: Prior to generating the set of events from the sensor data and the fault code data: (220) removing from the sensor data voltage measurements obtained during a time period when the rotational speed of the rotor of the aircraft (100) exceeds a threshold rpm; and (222) removing from the fault code data fault codes that were signaled during a time period when the rotational speed of the rotor of the aircraft (100) exceeded the threshold rpm.

4. The method of claim 1 , wherein the values ​​indexing the sequence of related events and the statistical correlation between the set of events in the summary table (126) stored in the memory and separate from the sensor data and the fault code data include: (224) updating the summary table (126) to indicate the total number of flights of the aircraft (100) and the total number of flights of other aircraft during the period in which the sequence of related events occurred and the total number of aircraft in the sequence in which the sequence of related events occurred.

5. The method of claim 1 , wherein controlling the display device ( 104 ) to display the summary table ( 126 ) and the visual representation of the temporal precursor network ( 128 ) on the graphical user interface comprises: (226) Controlling the display device (104) to depict in the visual representation (i) each event of the sequence of related events as a node and (ii) depicting the sequence of related events as two or more nodes (130, 132) connected by a link (134), wherein the link (134) indicates the timing of occurrence of the sequence of related events.

6. The method of claim 1, wherein controlling the display device (104) to display the summary table (126) and the visual representation of the temporal precursor network (128) on the graphical user interface comprises: (230) controlling the display device (104) to display, on the graphical user interface, in the summary table (126), a fourth identifier of an aircraft subsystem of the plurality of aircraft subsystems (108) associated with the abnormal voltage measurement.

7. The method of claim 1, wherein the sequence of related events comprises at least three events, and wherein controlling the display device (104) to display the summary table (126) and the visual representation of the temporal precursor network (128) on the graphical user interface comprises: (232) Controlling the display device (104) to display on the graphical user interface, in the summary table (126), (i) a corresponding identifier for each of the at least three events and (ii) the value of the statistical correlation between the at least three events.

8. The method according to claim 1, wherein the method further comprises: (234) calculating an average of the time period between the first event and the second event for past occurrences of the sequence of related events; and (236) calculating a standard deviation of the time period between the first event and the second event in the past occurrences of the sequence of related events, Wherein controlling the display device (104) to display the summary table (126) and the visual representation of the temporal precursor network (128) on the graphical user interface includes: (238) controlling the display device (104) to display on the graphical user interface, in the summary table (126) (i) an identifier of the first event, (ii) an identifier of the second event, (iii) a value of the statistical correlation between the first event and the second event, (iv) the average value of the time period between the first event and the second event when the sequence of related events occurred in the past, and (v) the standard deviation of the time period between the first event and the second event when the sequence of related events occurred in the past.

9. The method according to claim 1, further comprising: (240) referring to the reference data stored in the memory (120) to determine the total number of flights of the aircraft (100) during the period when the sequence of related events occurred and the total number of flights of other aircraft and the total number of aircraft in which the sequence of related events occurred; and (242) controlling the display device (104) to display the total number of flights of the aircraft (100) during the period in which the sequence of related events occurred and the total number of flights of the other aircraft as well as the total number of aircraft in which the sequence of related events occurred.

10. The method according to claim 1, further comprising: (244) referencing reference data stored in the memory (120) to determine a total number of flights of the aircraft (100) during which the sequence of related events occurred when the environmental condition existed and a total number of flights of other aircraft and a total number of aircraft during which the sequence of related events occurred when the environmental condition existed; and (246) controlling the display device (104) to display the total number of flights of the aircraft (100) and the total number of flights of the other aircraft during the period when the sequence of related events occurred when the environmental condition existed, and the total number of aircraft during which the sequence of related events occurred when the environmental condition existed.

11. A non-transitory computer-readable medium (120) having instructions (124) stored thereon, the instructions (124) when executed by one or more processors (114) of a computing device (106) causing the computing device to perform operations comprising: obtaining sensor data, the sensor data comprising voltage measurements over time of at least one electrical bus (112) on the aircraft (100); obtaining fault code data, the fault code data comprising fault codes signaled over time by a plurality of aircraft subsystems (108) on the aircraft (100), wherein each of the fault codes is associated with a corresponding aircraft subsystem of the plurality of aircraft subsystems (108), causing the corresponding aircraft subsystem to output the fault code upon detection of an error; calculating a voltage threshold, and indicating an abnormal voltage measurement in the sensor data when the voltage measurement exceeds the voltage threshold; generating a set of events from the sensor data and the fault code data, wherein each event in the set of events occurs within a time interval over which (i) the sensor data indicates the abnormal voltage measurement or (ii) a fault code associated with a particular aircraft subsystem of the plurality of aircraft subsystems (108) is signaled; calculating a value of the statistical correlation between the set of events, wherein the value of the statistical correlation between the set of events indicates a likelihood that the set of events are causally related events occurring in time sequence; constructing a temporal precursor network (128) representing the set of events as a sequence of correlated events and further representing the timing of occurrence of the sequence of correlated events based on the value of the statistical correlation between the set of events exceeding a predetermined threshold, wherein the sequence of correlated events includes a first event in which the sensor data indicates the abnormal voltage measurement and a second event in which the fault code associated with the particular aircraft subsystem of the plurality of aircraft subsystems is signaled; indexing said sequence of related events and said values ​​of statistical correlation between said set of events in a summary table (126) stored in memory and separate from said sensor data and said fault code data; as well as Controlling a display device (104) to display the summary table (126) and a visual representation of the time precursor network (128) on a graphical user interface, wherein controlling the display device to display the summary table on the graphical user interface includes: controlling the display device to display on the graphical user interface, in the summary table: (i) for the first event, a first identifier of an electrical bus of the at least one electrical bus on which the abnormal voltage measurement exists; (ii) for the second event, a second identifier of the fault code and a third identifier of the specific aircraft subsystem; and (iii) displaying a value of a statistical correlation between the first event and the second event.

12. The non-transitory computer readable medium (120) of claim 11, wherein the operations further comprise: Prior to generating the set of events from the sensor data and the fault code data: removing from the sensor data voltage measurements obtained during a time period when the rotational speed of a rotor of the aircraft (100) exceeds a threshold rpm; and Fault codes signaled during a time period when the rotational speed of the rotor of the aircraft (100) exceeds the threshold rpm are removed from the fault code data.

13. The non-transitory computer-readable medium (120) of claim 11, wherein controlling the display device (104) to display the summary table (126) and the visual representation of the temporal precursor network (128) on the graphical user interface comprises: The display device (104) is controlled to depict in the visual representation (i) each event of the sequence of related events as a node and (ii) the sequence of related events as two or more nodes connected by a link, the link indicating the timing of occurrence of the sequence of related events.

14. A system (102), comprising: a plurality of aircraft subsystems (108) on the aircraft (100); and A fault diagnosis computing device (106) having a processor (114) and a memory (120), the memory (120) storing instructions (124) executable by the processor to perform operations including: obtaining sensor data, the sensor data comprising voltage measurements over time of at least one electrical bus (112) on the aircraft (100); obtaining fault code data, the fault code data comprising fault codes signaled by the plurality of aircraft subsystems (108) over time, wherein each of the fault codes is associated with a respective aircraft subsystem of the plurality of aircraft subsystems (108), causing the respective aircraft subsystem to output the fault code upon detection of an error; calculating a voltage threshold, and indicating an abnormal voltage measurement in the sensor data when the voltage measurement exceeds the voltage threshold; generating a set of events from the sensor data and the fault code data, wherein each event in the set of events occurs within a time interval over which (i) the sensor data indicates the abnormal voltage measurement or (ii) a fault code associated with a particular aircraft subsystem of the plurality of aircraft subsystems (108) is signaled; calculating a value of the statistical correlation between the set of events, wherein the value of the statistical correlation between the set of events indicates a likelihood that the set of events are causally related events occurring in time sequence; constructing a temporal precursor network (128) representing the set of events as a sequence of correlated events and further representing the timing of occurrence of the sequence of correlated events based on the value of the statistical correlation between the set of events exceeding a predetermined threshold, wherein the sequence of correlated events includes a first event in which the sensor data indicates the abnormal voltage measurement and a second event in which the fault code associated with the particular aircraft subsystem of the plurality of aircraft subsystems is signaled; indexing, in a summary table (126) stored in the memory and separate from the sensor data and the fault code data, the values ​​of the statistical correlation between the sequence of related events and the set of events; as well as Controlling a display device (104) to display the summary table (126) and a visual representation of the time precursor network (128) on a graphical user interface, wherein controlling the display device to display the summary table on the graphical user interface includes: controlling the display device to display on the graphical user interface, in the summary table: (i) for the first event, a first identifier of an electrical bus of the at least one electrical bus on which the abnormal voltage measurement exists; (ii) for the second event, a second identifier of the fault code and a third identifier of the specific aircraft subsystem; and (iii) displaying a value of a statistical correlation between the first event and the second event.

Images