A fault injection and impact domain analysis system based on aircraft

Abstract

The present invention discloses a fault injection and impact domain analysis system based on an aircraft; the impact domain analysis system is composed of a knowledge base management module, a system database management module, a first fault tree management module, a diagnostic tree management module, and a fault diagnosis module. The fault injection method includes establishing a knowledge base, a system database, and a first fault tree, constructing a diagnostic tree, and executing a fault injection process. A second fault tree is formed by analyzing abnormal nodes, and the fault injection location is determined. According to the impact domain range of the fault diagnosis, a threshold range or a specific value is set, and compared with the fault data in the system database to qualitatively identify the real fault source. The impact domain analysis system can improve the efficiency of troubleshooting and maintenance, accurately diagnose and locate aircraft faults, and ensure the safe operation of the aircraft.

Description

Technical Field

[0001] The present invention belongs to the technical field of aircraft fault analysis, and in particular relates to an aircraft-based fault injection and impact domain analysis system. Background Art

[0002] With the increasing complexity of engineering structures, complex systems with redundant centralized architectures often include components such as computers, sensors, and servo actuators. During actual flight, aircraft are susceptible to external environmental influences, such as strong wind disturbances, temperature fluctuations, and electromagnetic interference after entering a minefield. Various components of the aircraft may experience a variety of random failures, such as servo jamming, throttle motor failure, inertial navigation signal loss, and power supply voltage instability. This requires the aircraft's flight control system to respond promptly and perform a series of corresponding processing, such as fault detection, fault location, fault isolation, and system reconfiguration, to minimize losses.

[0003] System failures can be caused by one or more factors. Due to the vertical or horizontal spread of faults, traditional system fault lists often contain numerous false faults caused by the actual source. This makes system fault diagnosis and location extremely difficult, placing high demands on maintenance personnel's maintenance and troubleshooting capabilities, often relying on experienced professionals for analysis. Developing an aircraft-based fault diagnosis system that can eliminate false faults, locate the actual source, and assist maintenance personnel in locating and isolating system faults has become a pressing issue for those skilled in the art.

[0004] Patent publication number CN112766408A discloses a method for diagnosing minor aircraft faults based on principal component analysis and a multi-layered extreme learning machine. The technical solution adopted is "using principal component analysis for feature extraction and noise reduction; constructing a multi-layered extreme learning machine for online sequence learning training; extracting features and denoising the actual data of the system to be tested using principal component analysis, and then inputting the data into the multi-layered extreme learning machine network online. Based on the output weight matrix, the online network output value of the actual data is calculated, and the output value is compared with the label value of the fault sample to make a fault judgment." This fault analysis method makes it difficult to eliminate false fault sources, resulting in multiple fault sources being located, which need to be eliminated one by one. Summary of the Invention

[0005] The present invention aims to provide an aircraft-based fault injection and impact domain analysis system to solve the following technical problems raised in the background art:

[0006] The aircraft-based fault diagnosis system in the prior art is unable to eliminate false faults and locate the real fault source.

[0007] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0008] An aircraft-based fault injection and impact domain analysis system, the impact domain analysis system includes a knowledge base management module, a system database management module, a first fault tree management module, a diagnostic tree management module, and a fault diagnosis module;

[0009] Fault injection includes the following methods:

[0010] S1, establishing a system knowledge base of the flight control system and establishing the component information of the flight control system in the knowledge base management module;

[0011] S2, establish the system database of the flight control system and inject system variables and fault data into the system database management module;

[0012] S3: Take the direct fault as the top event, list the direct fault causes that lead to the top event to form intermediate events; continue to search for the next level of fault causes through the intermediate events until the basic fault cause that cannot be further divided is found; the basic fault cause is the bottom event, and the top event, intermediate events, and bottom events together constitute the first fault tree and the first fault tree library;

[0013] S4, determining the top event. In the first fault tree, using the fault variable influence range provided by the expert knowledge base as the top event diagnosis criterion, the direct fault cause leading to the top event is searched layer by layer. All fault causes and fault cause combinations leading to the top event are found and a diagnostic tree and a diagnostic tree library are formed.

[0014] S5, executing the aircraft fault injection and impact domain analysis process; loading system variables, fault diagnosis information, the first fault tree, and the diagnostic tree into the fault diagnosis module; analyzing the aircraft operating data and extracting data record markers related to the diagnostic tree; forming a second fault tree for the system fault by analyzing the marked abnormal nodes, and determining the location of the fault injection; determining the impact domain range of one or more fault diagnoses;

[0015] Based on the acquired fault diagnosis's impact domain, a corresponding threshold range or specific value is set. The fault phenomenon is compared with the fault data in the system database to form corresponding fault diagnosis criteria to narrow the fault scope. Based on the characteristic parameters of the fault phenomenon, a mathematical model is used to calculate the fault value. Each fault value is then judged individually, for example, using circuit equations and parameters such as voltage and current. When the fault value is greater than or less than the set value, the node is identified as the true fault source. False faults are eliminated and one or more true fault sources are located.

[0016] Furthermore, in S1, the system knowledge base management module is provided with a data editing window.

[0017] Furthermore, in S1, the component information of the flight control system includes product hierarchy, system alarm information, FMECA table data, maintenance technical information, maintenance methods, and maintenance history information.

[0018] Furthermore, in step S2, the specific implementation method of establishing the composition information of the flight control system is to manage the FC, 1394 bus technology, and PCIE bus data recorded during the flight recording process, and perform data parsing on the recorded data by loading the corresponding ICD data file; wherein the bus data parsing model library performs data processing logic modeling, constructs a multi-type bus data parsing model library such as FC, 1394, PCIE, etc., parses the bus recording data file into a visual data file and injects it into the system database management module, wherein the parsed data file format includes subject, date, time, data packet timestamp, packet length, and signal value.

[0019] Furthermore, in step S3, the direct fault is taken as the top event, and the direct fault causes that lead to the top event are listed and connected with corresponding symbols to form intermediate events;

[0020] Furthermore, the system knowledge base stores an expert knowledge base; in step S5, the impact domain threshold of the fault diagnosis is input based on the expert system data.

[0021] Furthermore, the diagnostic information is collected by a diagnostic information collection system; and data from the diagnostic information collection system and the fault database are used as expert knowledge base data.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] The present invention utilizes a two-step fault diagnosis process. Based on the component information, system variables, and fault data of the aircraft system stored in the system knowledge base management module and the system database management module, the fault injection location is determined and a fault tree is formed. The first fault tree diagnosis is performed and the impact domain of the fault diagnosis is determined to perform preliminary fault elimination. The second fault diagnosis uses the impact domain information after the first fault diagnosis as a priori constraint and the first fault tree as a model. From top to bottom, each fault test point is subjected to a causal logic analysis strictly according to the fault hierarchy. After the fault list is imported, the system will traverse and compare the fault phenomena in the fault list with the fault data of the expert knowledge base stored in the system knowledge base based on the data provided in the database, form corresponding fault diagnosis criteria, and perform more accurate positioning of the aircraft system fault, thereby achieving the purpose of positioning the real fault source and eliminating other false faults.

[0024] The aircraft-based fault injection and impact domain analysis system of the present invention narrows the scope of fault troubleshooting in the entire tree through the first fault tree diagnosis, and then performs a second, more detailed fault diagnosis. This not only reduces the workload of fault troubleshooting, but also guides maintenance personnel to accurately locate, troubleshoot and isolate faults, solving the problem of low efficiency and low accuracy in fault diagnosis and location when maintenance personnel use traditional system fault lists. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 Schematic diagram of the influence domain analysis system of the present invention;

[0026] Figure 2 This is a flow chart of the fault diagnosis, impact and analysis of the present invention;

[0027] Figure 3 This is the first fault tree logic causal relationship of the present invention. DETAILED DESCRIPTION

[0028] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0029] Example

[0030] A fault injection and impact domain analysis system based on aircraft, such as Figure 1 As shown; the impact domain analysis system includes a knowledge base management module, a system database management module, a first fault tree management module, a diagnostic tree management module and a fault diagnosis module;

[0031] like Figure 2 As shown, fault injection includes the following methods:

[0032] S1, such as Figure 2 As shown, a system knowledge base of the flight control system is established, and the composition information of the flight control system is established in the knowledge base management module; it is mainly used to manage the product hierarchy of the flight control system, system alarm information, FMECA table data, maintenance technical information, maintenance methods / plans, maintenance history information, etc., and provides a system knowledge base data editing window.

[0033] S2, such as Figure 2 As shown, a system database of the flight control system is established, and system variables and fault data are injected into the system database management module; the system database management module is mainly used to manage the system variables and fault data of the flight control system, and provides a system variable and fault data editing window.

[0034] S3: Take the direct fault as the top event, list the direct fault causes that lead to the top event to form intermediate events; continue to search for the next level of fault causes through the intermediate events until the basic fault cause that cannot be further divided is found; the basic fault cause is the bottom event, and the top event, intermediate events, and bottom events together constitute the first fault tree and the first fault tree library;

[0035] The first fault tree management module is used to manage the first fault tree. The first fault tree is a special inverted tree-like logic causal relationship diagram, which uses event symbols, logic gate symbols and transfer symbols to describe the causal relationship between various events in the system.

[0036] By analyzing various factors that may cause aircraft system failure, including hardware, software, environmental changes, and human factors, etc. Take aircraft engine failure as an example, take it as the top event, and list the direct failure causes that lead to the top event, such as hardware, software or human factors, and connect them with corresponding symbols to form intermediate events. Continue to trace the next level of failure causes in the same way until all the causes of system failure are found, and find the most basic cause that cannot be divided further, that is, the bottom event, to form the first fault tree library. For example, Figure 3 As shown in the figure, a fault tree logic causal relationship is specifically described. The top event is used to find the intermediate event 1 or intermediate event 2, and the search continues through the intermediate event 1 or intermediate event 2. There are two situations when continuing to search: one is to get the bottom event directly, and the other is to get the intermediate event. After getting the intermediate event, it is necessary to continue searching until only the bottom event can be found. For example, in Figure 3 In the example, we search through intermediate event 1 to find bottom event 1 or intermediate event 3. After finding intermediate event 3, we continue searching and find bottom event 4 and bottom event 5. We search through intermediate event 2 to find bottom event 2 or bottom event 3. In other words, the search ends only when the final result is the bottom event.

[0037] S4, determine the top event. In the first fault tree, use the fault variable influence range provided by the expert knowledge base as the top event diagnosis criterion, and find the direct fault cause leading to the top event layer by layer; find all the fault causes and fault cause combinations leading to the top event and form a diagnostic tree and a diagnostic tree library; Figure 2 As shown in the figure, the expert knowledge base is a database containing expert experience and knowledge. This expert knowledge base collects the knowledge and experience accumulated by experts in a specific field, including failure modes, causes, and repair methods. It can be a structured database or a knowledge base containing documents, articles, cases, and rules.

[0038] The specific process is to first determine a fault test point, find out all its variable information, and then form a fault diagnosis criterion based on the influence range of the fault variable provided by the expert knowledge base. After the criterion is passed, continue to find the cause of the fault in the next layer, and then find out the direct cause of the fault event layer by layer, draw a logical relationship diagram (tree model), and finally find out all the causes and cause combinations that lead to the top event (that is, identify all fault modes that lead to the top event) to form a diagnostic tree. The purpose of the diagnostic tree is to find out the probability of the top event of the first fault tree, the combination of fault causes, discover some reliability and safety weaknesses, and perform more detailed fault diagnosis on the first fault tree. Through the analysis of the above two processes, the various combinations of causes of system failures and the probability of occurrence are determined to form a diagnostic tree library;

[0039] S5, such as Figure 2 As shown, the aircraft fault injection and impact domain analysis process is executed. System variables, fault diagnosis information, the first fault tree, and the diagnostic tree are loaded into the fault diagnosis module. Aircraft operating data is analyzed and data record tags related to the diagnostic tree are extracted. By analyzing the marked abnormal nodes, a second fault tree for the system fault is formed and the location of the fault injection is determined. The impact domain of one or more fault diagnoses is determined. The fault phenomenon is traversed and compared with the fault data in the system database to form corresponding fault diagnosis criteria to narrow the fault scope. Based on the characteristic parameters of the fault phenomenon, a mathematical model is used to calculate the fault value. Each fault value is judged separately. For example, the fault value is calculated using circuit equations and parameters such as voltage and current. For example, for a simple capacitor, the relationship between current (I) and capacitor voltage (Vc) can be expressed by the following equation: I = C * dVc / dt. Where C is the capacitor's capacitance and dVc / dt is the rate of change of the capacitor voltage over time. Assuming that we have measured the change in the capacitor voltage over time and know that the capacitor's capacitance is 10μF, we can use this data to calculate the capacitor's fault value. For example, assume that at time t = 0, the capacitor voltage is 0V, and at time t = 1s, the capacitor voltage is 5V. We can substitute this data into the above equation and solve it to calculate the capacitor's fault value. First, we can calculate the rate of change of the capacitor voltage over time: dVc / dt = (5V - 0V) / (1s - 0s) = 5V / s. Then, we can substitute this rate of change into the equation and solve for the current: I = 10μF * 5V / s = 50μA; therefore, we calculate the capacitor's fault value to be 50μA. When the fault value is greater than or less than the set value, the node is identified as a true fault source. Eliminate false faults and locate one or more true fault sources.

[0040] The fault diagnosis module uses the first fault tree as a model and conducts a causal logic analysis on each fault test point from top to bottom in strict accordance with the fault hierarchy. Figure 3 shown.

[0041] Specifically, the fault diagnosis module is used to detect and locate equipment failures. It locates the cause of equipment failures by analyzing aircraft operating data, log records, and reported error information. The fault diagnosis module process imports aircraft operating data, log records, and reported error information, pre-processes the data, extracts data associated with system variables, and saves them to the system variable memory. After the data processing is completed, the diagnosis model is configured according to the fault diagnosis module. In the execution of the diagnosis model, each test point is executed in sequence. The value range of the variable configured for each test point is used as the basis for judgment. When executing the test point, the data in the system variable memory is compared with the value range of the test point variable. If it is within the range, it indicates that there is no fault. Otherwise, it indicates a fault.

[0042] The specific process of S5 is as follows: load the system variables, fault diagnosis information, the first fault tree and the diagnostic tree into the fault diagnosis module, execute the first fault diagnosis process, and mark all abnormal nodes that appear this time by extracting data records related to the diagnostic tree model from the aircraft operation data (according to the implementation requirements, it may be the fault code, faulty equipment, abnormal variables, etc. reported by the self-test equipment) to form a second fault tree for the system fault and determine the location of the fault injection, and finally determine the impact domain range of one or more fault diagnoses to achieve the purpose of preliminary troubleshooting. Then execute the secondary fault diagnosis process within the impact domain range. According to the impact domain range of the fault diagnosis obtained in the first diagnosis, set the corresponding threshold range or specific value, where the impact domain threshold of the fault diagnosis is based on the data of the intelligent diagnostic information acquisition system and the fault library as the expert system data input. Using the experience accumulated by experts in the long-term work process, the system can save the expert knowledge base in the system knowledge base to form a system knowledge base that combines fault phenomena, fault causes and solutions. After the fault list is imported, the system will traverse and compare the fault phenomena in the fault list with the fault data in the database based on the data provided in the system knowledge base to form the corresponding fault diagnosis criteria. When the fault value is greater than or less than the set value, the node is identified as a real fault source, and then false faults are eliminated and one or more real fault sources are located. Fault information, troubleshooting strategies and fault isolation methods are pushed to users by matching the knowledge base.

[0043] In a preferred embodiment, in S1, the system knowledge base management module is provided with a data editing window.

[0044] In a preferred embodiment, in S1, the component information of the flight control system includes product hierarchy, system alarm information, FMECA table data, maintenance technical information, maintenance methods, and maintenance history information.

[0045] In a preferred embodiment, in step S2, the specific implementation method of establishing the composition information of the flight control system is to manage the FC, 1394 bus technology, and PCIE bus data recorded during the flight recording process, and perform data parsing on the recorded data by loading the corresponding ICD (electrical interface file) data file; wherein the bus data parsing model library performs data processing logic modeling, constructs a multi-type bus data parsing model library such as FC, 1394, PCIE, etc., parses the bus recording data file into a visual data file and injects it into the system database management module, wherein the parsed data file format includes subject, date, time, data packet timestamp, packet length, and signal value.

[0046] FC is a high-speed serial transmission bus proposed by the X3T11 group of the American Institute for Standardization (ANSI) in 1988. 1394 bus technology is one of the more advanced avionics bus technologies. It is based on the IEEE-1394b protocol and has been modified and constrained to propose the AS5643 protocol. PCIE is a high-speed serial computer expansion bus standard, mainly used to expand computer system bus data throughput and improve device communication speed.

[0047] In a preferred embodiment, in step S3, the direct fault is taken as the top event, and the direct fault causes leading to the top event are listed and connected with corresponding symbols to form intermediate events;

[0048] In a preferred embodiment, the system knowledge base stores an expert knowledge base; in step S5, the impact domain threshold of the fault diagnosis is input based on expert system data.

[0049] In a preferred embodiment, the diagnostic information is collected by a diagnostic information collection system; and the data from the diagnostic information collection system and the fault database are used as expert knowledge base data.

[0050] In the description of the present invention, it should be understood that the terms "coaxial", "bottom", "one end", "top", "middle", "the other end", "upper", "one side", "top", "inside", "front", "center", "both ends", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0051] In the present invention, unless otherwise clearly stipulated and limited, the terms "install", "set", "connect", "fix", "screw" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integrated connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. Unless otherwise clearly defined, ordinary technicians in this field can understand the specific meanings of the above terms in the present invention according to the specific circumstances.

[0052] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An aircraft-based fault injection and impact domain analysis system, characterized by: The impact domain analysis system includes a knowledge base management module, a system database management module, a first fault tree management module, a diagnosis tree management module and a fault diagnosis module; Fault injection includes the following methods: S1, establishing a system knowledge base of the flight control system and establishing the component information of the flight control system in the knowledge base management module; S2, establish the system database of the flight control system and inject system variables and fault data into the system database management module; S3: Take the direct fault as the top event and list the direct fault causes that led to the top event to form intermediate events. Continue to search for the next level of fault causes through the intermediate events until the basic fault cause that cannot be further divided is found. The basic fault cause is the bottom event, and the top event, the intermediate event and the bottom event together constitute the first fault tree and the first fault tree library; S4, determining the top event. In the first fault tree, using the fault variable influence range provided by the expert knowledge base as the top event diagnosis criterion, the direct fault cause leading to the top event is searched layer by layer. All fault causes and fault cause combinations leading to the top event are found and a diagnostic tree and a diagnostic tree library are formed. S5, execute the aircraft fault injection and impact domain analysis process; Loading system variables, fault diagnosis information, a first fault tree, and a diagnostic tree into a fault diagnosis module; Analyze aircraft operating data and extract data record markers related to the diagnostic tree; form a second fault tree for system faults by analyzing the marked abnormal nodes and determine the location of fault injection; determine the impact domain of one or more fault diagnoses; According to the obtained fault diagnosis influence domain range, the corresponding threshold range or specific value is set; the fault phenomenon is traversed and compared with the fault data in the system database to form the corresponding fault diagnosis judgment criteria; the fault value is calculated based on the characteristic parameters of the fault phenomenon, and each fault value is judged separately. When the fault value is greater than or less than the set value, the node is characterized as the real fault source.

2. The aircraft-based fault injection and impact domain analysis system according to claim 1, characterized in that: In S1, the system knowledge base management module is provided with a data editing window.

3. The aircraft-based fault injection and impact domain analysis system according to claim 1, characterized in that: In step S2, the specific implementation method of establishing the composition information of the flight control system is to manage the FC, 1394 bus technology, and PCIE bus data recorded during the flight recording process, and perform data parsing on the recorded data by loading the corresponding ICD data file; wherein the bus data parsing model library performs data processing logic modeling, constructs the FC, 1394, and PCIE multi-type bus data parsing model library, parses the bus recording data file into a visual data file and injects it into the system database management module, wherein the parsed data file format includes subject, date, time, data packet timestamp, packet length, and signal value.

4. The aircraft-based fault injection and impact domain analysis system according to claim 1, characterized in that: In step S3, the direct fault is taken as the top event, and the direct fault causes that lead to the top event are listed and connected with corresponding symbols to form intermediate events.

5. The aircraft-based fault injection and impact domain analysis system according to claim 1, characterized in that: The system knowledge base stores an expert knowledge base; in step S5, the impact domain threshold of the fault diagnosis is input based on the expert system data.

6. The aircraft-based fault injection and impact domain analysis system according to claim 5, characterized in that: The diagnostic information is collected by the diagnostic information collection system; the data of the diagnostic information collection system and the fault database are used as the expert knowledge base data.

Images