Artificial intelligence based avionics fault diagnosis and prediction system

Abstract

The application discloses an artificial intelligence-based avionics equipment fault diagnosis and prediction system and relates to the technical field of avionics equipment fault diagnosis.The system comprises a receiving module, an acquisition module, an abnormal data generation module, a common cause probability generation module and a correlation analysis module, and realizes the following functions when the system is running: receiving sensing data, acquiring external environment data, generating abnormal data according to the sensing data, generating a common cause failure probability value according to the external environment data and a pre-constructed causal topology model, correlating the abnormal data and the common cause failure probability value, and judging whether the abnormal data is caused by a single-point hardware fault or environmental factor interference.The system has the beneficial effect of being able to distinguish between single-point hardware faults and environmental interference and improving the reliability of avionics equipment fault diagnosis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of avionics fault diagnosis, and in particular to an avionics fault diagnosis and prediction system based on artificial intelligence. BACKGROUND

[0002] As an important part of modern civil aircraft, the running state of avionics directly relates to the safety and economy of flight. With the increasing complexity of airborne systems, how to effectively diagnose and predict the faults of key avionics has become an important issue in aviation operation support.

[0003] In the prior art, the fault diagnosis of avionics mainly adopts a fault detection mechanism based on redundant sensor voting. The core principle is to configure multiple redundant sensors for key parameters, compare the readings of these sensors, and make decisions according to the "majority consensus" principle: when the readings of the majority of sensors are consistent, it is considered that the value is reliable, and the readings of the minority of sensors that differ from the value by more than a predetermined threshold are determined as faults and isolated. This method can effectively monitor the running state of the equipment in most flight stages and to some extent predict potential faults.

[0004] However, the effectiveness of this mechanism is based on the core assumption that "sensor faults occur independently in statistics". When the aircraft takes off and lands at an airport in a tropical or high-humidity area, the humidity of the runway environment may suddenly change in a very short time. Water or contaminants on the wet runway may be splashed by the wheels, or condense in a high-humidity environment, causing simultaneous contamination or physical blockage of multiple static pressure tubes, dynamic pressure tubes, temperature and humidity sensors, etc. This multi-sensor deviation caused by common environmental factors violates the basic assumption of "independent faults", making the readings between redundant sensors show a highly correlated consistent error, making it difficult to distinguish between single-point faults and environmental disturbances.

[0005] Therefore, an avionics fault diagnosis and prediction system based on artificial intelligence is proposed. SUMMARY

[0006] In view of the above prior art, the present application is proposed. The embodiments of the present application provide an avionics fault diagnosis and prediction system based on artificial intelligence, which can distinguish between single-point hardware faults and environmental disturbances, and improve the reliability of avionics fault diagnosis.

[0007] According to an aspect of the present application, an artificial intelligence-based avionics equipment fault diagnosis and prediction system is provided, comprising: a receiving module configured to receive sensor data from a plurality of onboard sensors; an obtaining module configured to obtain external environment data related to an external environment of an aircraft; an abnormal data generation module configured to generate abnormal data according to the sensor data and a preset physical constraint relationship, the abnormal data being used to indicate whether there is a reading abnormality of an onboard sensor and an attribute of the abnormality; a common cause probability generation module configured to generate a common cause failure probability value according to the external environment data and a pre-constructed causal topology model, the causal topology model defining a probability relationship in which an environmental disturbance affects a plurality of onboard sensors via a physical propagation path, and the common cause failure probability value representing a probability that a current environment causes a correlated common error of the plurality of onboard sensors; and a correlation analysis module configured to perform correlation analysis on the abnormal data and the common cause failure probability value, wherein if the abnormal data indicates that there is a reading abnormality of a single onboard sensor, and the common cause failure probability value is lower than a preset threshold, it is determined that the reading abnormality is caused by a single-point hardware fault; and if the abnormal data indicates that there are at least two of the onboard sensors having a coordinated reading abnormality matching a preset common cause failure feature, and the common cause failure probability value is higher than the preset threshold, it is determined that the reading abnormality is caused by the environment.

[0008] According to another aspect of the present application, an electronic device is provided, comprising a memory and a processor, the memory being configured to store computer executable instructions, and the processor being configured to execute the computer executable instructions, the computer executable instructions, when executed by the processor, realizing the functions of the system as described above.

[0009] According to another aspect of the present application, a computer storage medium is provided, having stored thereon computer executable instructions, the computer executable instructions, when executed by a processor, realizing the functions of the system as described above.

[0010] Compared with the prior art, by using the artificial intelligence-based avionics equipment fault diagnosis and prediction system according to the embodiments of the present application, the sensor data is obtained by the receiving module, the abnormal data and the common cause failure probability value are generated in combination with the external environment data, and the fault type is comprehensively judged by using the correlation analysis module, so that the coordinated error caused by the single-point hardware fault and the environmental interference can be distinguished, and the reliability of the avionics equipment fault diagnosis is improved. BRIEF DESCRIPTION OF DRAWINGS

[0011] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings. The drawings provided in the specification and the contents of the specification are to provide further understanding of the embodiments of the present application, and constitute a part of the specification, which are used to explain the present application together with the embodiments of the present application, and do not constitute a limitation to the present application. In the drawings, the same reference numerals generally designate the same components or steps.

[0012] Figure 1 A block diagram of an artificial intelligence-based avionics fault diagnosis and prediction system.

[0013] Figure 2 A data flow diagram of an artificial intelligence-based avionics fault diagnosis and prediction system.

[0014] Figure 3 A block diagram of an electronic device. DETAILED DESCRIPTION

[0015] Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part but not all of the embodiments of the present application, and the present application can be implemented in many different forms. Therefore, the attached drawings should not be used to limit and define the present application, and the present application should cover all changes falling within the scope of the appended claims and their equivalents.

[0016] Example system

[0017] Figures 1-2 An artificial intelligence-based avionics fault diagnosis and prediction system according to an embodiment of the present application is illustrated, including a receiving module, an obtaining module, an abnormal data generation module, a common cause probability generation module, and a correlation analysis module.

[0018] In the system, the receiving module is configured to receive sensing data from a plurality of onboard sensors.

[0019] The onboard sensor refers to a detection device installed on an aircraft for collecting flight parameters or device status, and can be implemented by using a static pressure pipe, a dynamic pressure pipe, a temperature and humidity sensor, an acceleration sensor, a wheel speed sensor, etc. The function of the onboard sensor is to provide real-time data input for fault diagnosis.

[0020] In the system, the obtaining module is configured to obtain external environment data related to the external environment of the aircraft.

[0021] The external environment data refers to physical parameters related to the external environment in which the aircraft is located, and can be implemented by using runway humidity, temperature, pollutant concentration, and atmospheric pressure data. The function of the external environment data is to provide a basis for judging the influence of environmental factors on the sensor.

[0022] In the system, the abnormal data generation module is configured to generate abnormal data according to the sensing data and a preset physical constraint relationship. The abnormal data is used to indicate whether there is an onboard sensor with a reading abnormality and the attribute of the abnormality.

[0023] Specifically, the generation of abnormal data includes the following steps:

[0024] First, a physical constraint residual of the sensing data is calculated according to a physical constraint relationship, the physical constraint relationship including at least one of the following:

[0025] An aerodynamic equation constraint for verifying consistency among aircraft airspeed, static pressure and total pressure data, the aerodynamic equation constraint specifically establishing a mathematical relationship among airspeed, static pressure and total pressure through Bernoulli equation, the constraint being able to detect inconsistency between airspeed and static pressure data when static pressure measurement is abnormal due to static pressure pipe blockage;

[0026] A kinematic relationship constraint for verifying consistency between airspeed data and wheel speed data when the aircraft touches the ground, the kinematic relationship constraint being activated in the landing gear touching ground stage, and identifying abnormal airspeed sensor due to runway water accumulation by comparing data difference between airspeed sensor and wheel speed sensor;

[0027] A geometric relationship constraint for verifying reading relationship of same type sensors in different installation positions under a specific flight attitude, the geometric relationship constraint being based on relationship between sensor installation position and aircraft aerodynamic shape, and establishing a theoretical reading deviation range of different position sensors under a specific pitch angle or roll angle.

[0028] Then, the physical constraint residual is standardized to generate a standardized residual.

[0029] Finally, it is judged whether the standardized residual exceeds a preset residual threshold, if yes, it is determined that the corresponding airborne sensor has reading abnormality, and abnormal data is generated, wherein the abnormal data includes an airborne sensor identifier of the reading abnormality, an abnormal severity and abnormal time information.

[0030] In the system, the common cause probability generation module is configured to generate a common cause failure probability value according to external environment data and a pre-constructed causal topology model, the causal topology model defining a probability relationship that an environmental disturbance influences multiple airborne sensors via a physical propagation path, and the common cause failure probability value representing a probability that the current environment causes correlated common errors of the multiple airborne sensors.

[0031] The construction of the causal topology model includes the following steps:

[0032] First, historical flight data is obtained and sensor calibration data .

[0033] Then, an environmental variable in the external environment data and a measurement of the airborne sensor are respectively associated as an environmental node and a sensor node;

[0034] Specifically, a set of external environment variables and a set of airborne sensor measurement quantities are extracted, Establishing as an environmental node, each sensor measurement quantity Establishing as a sensor node, obtaining a node set .

[0035] Next, according to the physical propagation mechanism, the association relationship from the environmental node to the sensor node affected by it is established, and the initial topology structure is formed, wherein the physical propagation mechanism is represented by a corresponding physical model, at least including:

[0036] An aerodynamic model: using fluid mechanics equations or linearized models to describe the mapping between air pressure, velocity field disturbance and sensor reading error;

[0037] A structural dynamics model: using structural vibration dynamics equations to establish the coupling relationship between external load, structural response and sensor signal;

[0038] A heat conduction and radiation model: using heat conduction equations to establish the relationship between external temperature disturbance and sensor temperature drift;

[0039] An electromagnetic interference model: using Maxwell's equations to establish the coupling path of lightning or electromagnetic pulse to the sensor circuit;

[0040] When the physical model determines that there is an effective propagation path from to , a directed edge is established in the topology to form the initial topology structure.

[0041] Then, according to the association relationship in the initial topology structure, the propagation parameter representing the influence degree of environmental disturbance on the sensor reading is generated, that is, on each edge of the initial topology structure, the propagation parameter is defined to describe the conditional probability influence of environmental disturbance on sensor reading: , wherein is given by a physical model or a statistical regression model;

[0042] Finally, according to the historical flight data and sensor calibration data, the propagation parameter is estimated by maximum likelihood estimation or Bayesian inference:

[0043]

[0044] , wherein and are the sensor reading and environmental observation at the t-th moment, respectively, if the Bayesian method is used, the prior distribution can be introduced into the above likelihood function to calculate the posterior distribution ;

[0045] A causal topology model for calculating the common cause failure probability value wherein, is a set of causal edges constrained by physical propagation mechanisms, is a set of propagation parameters, the model can calculate the common cause failure probability of multiple sensors appearing correlated errors by Bayesian inference or belief propagation algorithm given the current environment observation:

[0046]

[0047] wherein, is a set of airborne sensors that are currently detected to have abnormal readings.

[0048] In the system, the correlation analysis module is configured to correlate the abnormal data with the common cause failure probability value:

[0049] If the abnormal data indicates that a single airborne sensor has a reading abnormality, and the common cause failure probability value is lower than a preset threshold, it is determined that the reading abnormality is caused by a single-point hardware failure.

[0050] Specifically, if only one airborne sensor in the abnormal data has a reading abnormality, the abnormal data indicates that a single airborne sensor has a reading abnormality.

[0051] If the abnormal data indicates that at least two airborne sensors have a coordinated reading abnormality matching a preset common cause failure feature, and the common cause failure probability value is higher than a preset threshold, it is determined that the reading abnormality is caused by the environment.

[0052] Specifically, the preset common cause failure feature can include a time feature, an abnormality degree feature, and an abnormality position feature. Determining whether at least two airborne sensors have a coordinated reading abnormality matching the preset common cause failure feature includes the following steps:

[0053] First, according to the abnormal time information contained in the abnormal data, it is determined whether the reading abnormalities of the at least two airborne sensors occur synchronously within a preset time window. If so, a first determination evidence indicating that there is a coordinated reading abnormality is generated.

[0054] Then, according to the abnormal severity contained in the abnormal data, it is determined whether the reading abnormalities of the at least two airborne sensors satisfy a preset severity correlation condition. If so, a second determination evidence indicating that there is a coordinated reading abnormality is generated.

[0055] Next, according to the identification of the airborne sensors contained in the abnormal data, the physical installation positions of the at least two airborne sensors are extracted from a preset identification-installation position mapping table. It is determined whether the physical installation positions of the at least two airborne sensors are adjacent. If so, a third determination evidence indicating that there is a coordinated reading abnormality is generated.

[0056] Finally, if there is at least one of the first determination evidence, the second determination evidence and the third determination evidence, it is determined that at least two airborne sensors exist a collaborative reading anomaly matching the preset common cause failure feature.

[0057] In summary, the core innovation of the present application is to combine the physical constraint relationship with the causal topology model, and effectively distinguish single-point hardware failure and sensor collaborative failure caused by environmental factors through the correlation analysis of abnormal data and common cause failure probability value. The traditional method relies on the sensor redundant voting mechanism and cannot identify related errors caused by common environmental factors. The present application establishes the causal influence probability model of environmental variables to the sensor group, combines the time, space and physical quantity characteristics matching of abnormal data, realizes the accurate identification of common cause failure, and solves the misjudgment problem caused by the synchronous failure of multiple sensors in the harsh environment such as high humidity.

[0058] Exemplary electronic device

[0059] Figure 3 A block diagram of an electronic device according to embodiments of the present application is illustrated.

[0060] As Figure 3 shown, the electronic device includes one or more processors and a memory.

[0061] The processor can be a central processing unit (CPU) or other form of processing unit having data processing and / or instruction execution capabilities, and can control other components in the electronic device to perform desired functions.

[0062] The memory can include one or more computer program products that can include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory, for example, can include random access memory (RAM), cache, and / or the like. Non-volatile memory, for example, can include read-only memory (ROM), hard disk, flash memory, and / or the like.

[0063] In one example, the electronic device can further include input and output devices, which are interconnected through a bus system and / or other forms of connection mechanism (not shown).

[0064] Of course, in order to simplify, Figure 3 only some of the components in the electronic device related to the present application are shown, and components such as buses, input / output interfaces, and the like are omitted. In addition, the electronic device can also include any other appropriate components according to specific application circumstances.

[0065] Exemplary computer-readable medium

[0066] Embodiments of the present application can also be computer readable storage media storing computer program instructions thereon which, when executed by a processor, cause the processor to perform the steps described in the above "Exemplary Methods" section of this specification according to various embodiments of the present application.

[0067] The computer readable storage medium can be any combination of one or more computer readable media. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium include an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0068] The above describes the basic principles of the present application in conjunction with specific embodiments. However, it should be noted that the advantages, benefits, effects and the like mentioned in the present application are only examples and are not limiting, and these advantages, benefits, effects and the like are not necessarily required for each embodiment of the present application. In addition, the above specific details are only for the purpose of example and understanding, and are not limiting, and the above details do not necessarily require the specific details to be implemented.

[0069] The block diagrams of the devices, apparatuses, equipment, systems involved in the present application are only illustrative examples and are not intended to require or imply the connection, arrangement, configuration as shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, equipment, systems can be connected, arranged, configured in any manner. Words such as "include", "contain", "have" and the like are open-ended words, meaning "including but not limited to", and can be used interchangeably. The words "or" and "and" used herein mean the word "and / or", and can be used interchangeably unless the context clearly indicates otherwise. The word "such as" used herein means the phrase "such as but not limited to", and can be used interchangeably.

[0070] It should also be noted that in the devices, apparatuses and methods of the present application, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions of the present application.

[0071] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0072] The above description has been presented to enable any person skilled in the art to make or use the application. Numerous modifications to the aspects described herein will be readily apparent to those skilled in the art, and the principles defined herein can be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An artificial intelligence-based avionics equipment fault diagnosis and prediction system, characterized in that, The system includes: The receiving module is used to receive sensing data from multiple airborne sensors; The acquisition module is used to acquire external environment data related to the aircraft's external environment. An abnormal data generation module is used to generate abnormal data based on the sensing data and the preset physical constraint relationship. The abnormal data is used to indicate whether there are airborne sensors with abnormal readings and the attributes of the abnormality. The common cause probability generation module is used to generate a common cause failure probability value based on the external environment data and the pre-built causal topology model. The causal topology model defines the probabilistic relationship of environmental disturbances affecting multiple airborne sensors through physical propagation paths. The common cause failure probability value represents the probability that the current environment causes multiple airborne sensors to have correlated common errors. The correlation analysis module is used to perform correlation analysis between the abnormal data and the common cause failure probability value: if the abnormal data indicates that a single airborne sensor has an abnormal reading and the common cause failure probability value is lower than a preset threshold, then the abnormal reading is determined to be caused by a single point of hardware failure; if the abnormal data indicates that at least two airborne sensors have a coordinated abnormal reading that matches a preset common cause failure feature and the common cause failure probability value is higher than a preset threshold, then the abnormal reading is determined to be caused by the environment. The construction of the causal topology model includes: acquiring historical flight data and sensor calibration data; associating environmental variables in the external environment data and the measurements of the airborne sensors as environmental nodes and sensor nodes, respectively; establishing the correlation relationship from the environmental nodes to the sensor nodes affected by them according to the physical propagation mechanism to form an initial topology; generating propagation parameters representing the degree of influence of environmental disturbances on the airborne sensor readings based on the correlation relationship in the initial topology; and estimating the propagation parameters based on the historical flight data and sensor calibration data to obtain the causal topology model used to calculate the common cause failure probability value.

2. The avionics equipment fault diagnosis and prediction system based on artificial intelligence according to claim 1, characterized in that, The step of generating abnormal data based on the relationship between the sensing data and preset physical constraints includes: Calculate the physical constraint residuals of the sensing data based on the physical constraint relationships; The physical constraint residuals are standardized to generate standardized residuals; Determine whether the standardized residual exceeds a preset residual threshold. If so, determine that the corresponding airborne sensor has an abnormal reading and generate the abnormal data. The abnormal data includes the airborne sensor identifier of the abnormal reading, the severity of the abnormality, and the time of the abnormality.

3. The avionics equipment fault diagnosis and prediction system based on artificial intelligence according to claim 1 or 2, characterized in that, The physical constraint relationship includes at least one of the following: Aerodynamic equation constraints used to verify the consistency between aircraft airspeed, static pressure and total pressure data; Kinematic constraints used to verify the consistency between airspeed data and wheel speed data when the aircraft touches down; Geometric constraints used to verify the reading relationships of the same type of sensors installed at different locations under a specific flight attitude.

4. The avionics equipment fault diagnosis and prediction system based on artificial intelligence according to claim 3, characterized in that, The indication that at least two of the airborne sensors exhibit coordinated reading anomalies matching preset common-cause failure characteristics includes: Based on the abnormal time information contained in the abnormal data, it is determined whether the abnormal readings of the at least two airborne sensors occurred synchronously within a preset time window. If so, first evidence indicating the existence of the coordinated reading abnormality is generated.

5. The avionics equipment fault diagnosis and prediction system based on artificial intelligence according to claim 4, characterized in that, The indication that at least two of the airborne sensors exhibit a coordinated reading anomaly matching a preset common-cause failure characteristic also includes: Based on the severity of the anomalies contained in the abnormal data, it is determined whether the reading anomalies of the at least two airborne sensors meet the preset severity correlation conditions. If so, a second determination evidence indicating the existence of the cooperative reading anomaly is generated.

6. The avionics equipment fault diagnosis and prediction system based on artificial intelligence according to claim 5, characterized in that, Before determining that at least two airborne sensors exhibit a cooperative reading anomaly matching a preset common-cause failure characteristic, the following steps are also included: Based on the identifiers of the airborne sensors contained in the abnormal data, the physical installation locations of at least two airborne sensors are extracted from a preset identifier-installation location mapping table. Determine whether the physical installation locations of the at least two airborne sensors are adjacent; if so, generate third evidence indicating the presence of the abnormal synergy readings.

7. The avionics equipment fault diagnosis and prediction system based on artificial intelligence according to claim 6, characterized in that, If at least one of the first determination evidence, the second determination evidence, and the third determination evidence exists, then it is determined that the at least two airborne sensors have a cooperative reading anomaly that matches the preset common cause failure characteristics.

8. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the functions of the system as described in any one of claims 1 to 7.

9. A computer storage medium storing computer-executable instructions thereon, characterized in that: When the computer-executable instructions are executed by the processor, they implement the functions of the system as described in any one of claims 1 to 7.

Images