Avionic electromechanical equipment fault diagnosis and ar maintenance guidance method and system

Abstract

The application relates to an aviation electromechanical equipment fault diagnosis and AR maintenance guidance method and system, and belongs to the technical field of aviation maintenance engineering, and aims to solve the problems of no dynamic updating mechanism of maintenance knowledge, insufficient adaptability of fault coupling relationship, lack of airworthiness fusion and prior simulation verification of maintenance scheme, and multi-mode adaptation capability of field guidance in the prior art. The method comprises the following steps: constructing and dynamically updating a structured knowledge network and a self-adaptive maintenance rule base, outputting a diagnosis report with compliance identification through a double-layer diagnosis engine, obtaining an optimized maintenance scheme through simulation evaluation of a digital twin model, generating an augmented reality work card of multi-mode guidance, and optimizing resource scheduling and feeding back the rule base in combination with a dynamic coordination mechanism. The application realizes real-time dynamic updating of maintenance knowledge, accurate and efficient diagnosis of coupled faults, prior simulation verification of maintenance decision, and improvement of safety and efficiency of field operation through the above method.

Description

Technical Field

[0001] This invention belongs to the field of aviation maintenance engineering technology, specifically relating to a method and system for fault diagnosis and AR maintenance guidance of aviation electromechanical equipment. Background Technology

[0002] Currently, aviation electromechanical equipment is developing towards higher integration and complexity, resulting in increasingly diverse fault types and frequent coupled faults. This places higher demands on the accuracy of fault diagnosis and the efficiency of maintenance guidance. Traditional technologies often employ a model combining fixed technical manuals, decentralized information systems, and human experience. Diagnostic work is carried out through static knowledge bases or single data models, relying on manual comparison with airworthiness standards, development of maintenance plans, and on-site guidance to meet basic maintenance needs. However, this traditional model is difficult to adapt to the actual needs of complex scenarios: static knowledge bases cannot absorb new maintenance case experience in a timely manner, resulting in lagging knowledge updates and low accuracy in depicting dynamically changing fault coupling relationships, which can easily lead to diagnostic biases; the diagnostic process lacks deep integration with the minimum equipment list database for airworthiness requirements, leading to low efficiency in compliance judgment; maintenance plans lack pre-simulation verification, making it difficult to predict execution risks and effects; and on-site operation guidance is limited to text or diagrams, which is not adaptable to complex component disassembly and assembly and multi-step operations, affecting operational efficiency.

[0003] Existing technologies disclose a distributed expert collaborative troubleshooting system that aggregates information from multiple sources, including MCC (Maintenance Control Center), maintenance experts, and safety inspection departments, and utilizes AR smart glasses technology and software algorithms to generate maintenance work guidance. However, this system still suffers from the following shortcomings: It lacks a dynamic update mechanism for maintenance knowledge, resulting in insufficient adaptability to fault coupling relationships; maintenance plans lack deep integration with airworthiness standards, leading to low integration of compliance assessments; it lacks a pre-implementation simulation verification stage for maintenance plans, making it difficult to predict execution effects; and its multi-mode adaptability for on-site guidance is not refined, resulting in weak support for complex operational scenarios.

[0004] In view of this, it is very necessary for the present invention to provide a method and system for fault diagnosis and AR maintenance guidance of aircraft electromechanical equipment to solve the above-mentioned defects in the prior art. Summary of the Invention

[0005] To address the shortcomings of existing technologies, such as the lack of a dynamic update mechanism for maintenance knowledge, insufficient adaptability to fault coupling relationships, lack of deep integration of maintenance plans with airworthiness standards, low integration of compliance judgment, lack of pre-simulation verification of maintenance plans making it difficult to predict execution results, and insufficient refinement of multi-mode adaptability for on-site guidance, resulting in weak support for complex operational scenarios, this invention provides a method and system for fault diagnosis and AR maintenance guidance of aviation electromechanical equipment to solve the aforementioned technical problems.

[0006] In a first aspect, the present invention provides a method for fault diagnosis and AR maintenance guidance of aircraft electromechanical equipment, including: Step S1: Obtain multi-source information and fault information of aircraft electromechanical equipment, construct a structured knowledge network and an adaptive maintenance rule base based on the multi-source information, and dynamically update the adaptive maintenance rule base through feedback from historical maintenance cases. The dynamic update includes using an incremental learning algorithm to adjust the rule confidence in the structured knowledge network and quantitatively updating the weights of coupling relationship edges in the structured knowledge network based on multi-source information. The multi-source information includes historical maintenance case data, system schematic data, system safety assessment report data, and airworthiness directive information. The quantitative update of the coupling relationship edge weights is based on the co-occurrence probability of historical cases, the physical connection tightness of the system schematic, and the common-mode fault relationship in the system safety assessment report.

[0007] The incremental learning algorithm tunes the rule confidence in structured knowledge networks as follows: When validating rules for new cases, press C. new = C old + α · (1-C old )calculate; When a new case negates the rule, press C. new = C old • β calculation; Among them, C old C represents the confidence level before the rule update. new α represents the confidence level after the rule update, β represents the reward learning rate, and β represents the penalty decay coefficient. α and β are dynamically fine-tuned based on the historical trigger frequency of the rule or the recentity of the case.

[0008] By adopting the above scheme, the knowledge network construction and updating are made more targeted by clarifying the specific types of multi-source information and the basis for the quantitative update of the edge weights of coupling relationships, and the adaptability of fault coupling relationships is further improved. By dynamically adjusting the rule confidence through quantitative formulas and combining the dynamically fine-tuned reward and punishment parameters, the rule confidence update is made more flexible, and the iteration efficiency and reliability of the adaptive maintenance rule base are improved.

[0009] Step S2: Input the fault information into the dual-layer diagnostic engine. The upper-layer fast matching engine filters the rule subgraph, and the lower-layer graph calculation and inference engine performs circular confidence propagation calculation based on the rule subgraph. Simultaneously, it compares with the minimum equipment list database to complete the compliance analysis and outputs a diagnostic report with compliance identification. The circular confidence propagation calculation of the lower-level graph computation inference engine includes evidence fusion, which combines the results of the minimum equipment list database verification to rank the causes of failure; Evidence fusion includes the process of propagating fault information to intermediate nodes, the intermediate nodes propagating the fused evidence to fault nodes, and the fault nodes receiving multi-source evidence and updating their confidence. The ranking dimensions include confidence level and compliance status. The compliance status includes prohibition of release and release under restricted conditions.

[0010] By adopting the above technical solution, and by refining the dual ranking dimensions of evidence fusion propagation path and fault cause, the calculation of circular confidence propagation becomes more operable, thereby improving the accuracy of fault diagnosis.

[0011] Step S3: Input the diagnostic report into the digital twin model calibrated with aircraft condition monitoring data, perform multi-dimensional dynamic weighted simulation evaluation of candidate maintenance solutions, and output the optimized maintenance solution; The dimensions of the multi-dimensional dynamic weighted simulation assessment include estimated working hours, resource costs, technical compliance risks, and secondary damage risks; Among them, dynamic weighting adjusts the weight ratio of each dimension through a preset weight strategy template corresponding to the fault urgency level; When the emergency level of the fault is aircraft grounding failure, increase the weight of estimated working hours and on-site availability of key aviation materials in the simulation evaluation, and decrease the weight of resource cost in the simulation evaluation.

[0012] By adopting the above technical solution, we can clarify the multi-dimensional evaluation indicators and the dynamic weighting implementation method, making the maintenance plan evaluation more comprehensive and in line with the actual scenario. By adjusting the weights that are tied to the urgency level of the fault, we can ensure that the evaluation results are adapted to the needs of different fault scenarios. For aircraft parking faults, we can optimize the evaluation weight ratio, prioritize maintenance efficiency, and quickly select maintenance plans that are suitable for emergency faults.

[0013] Step S4: Based on the optimized maintenance plan, real-time resource status, and personnel qualifications, generate an augmented reality work card with component verification and multi-mode guidance capabilities; The component verification of augmented reality (AR) work cards uses image recognition and optical character recognition technologies. When the number of consecutive failures of image recognition exceeds a preset threshold, or the confidence level of the results obtained by optical character recognition and image recognition is lower than a preset threshold, the system switches to an augmented positioning guidance mode based on ultra-wideband or Bluetooth beacons to provide directional guidance in the augmented reality field of view.

[0014] By adopting the above technical solution, which integrates dual recognition technology and multi-mode guidance switching mechanism, it can adapt to the component verification needs in complex on-site environments, avoid work stoppages caused by the failure of a single recognition mode, improve the efficiency of component search and operation through precise positioning guidance, reduce the risk of human error, and ensure the smoothness of on-site operations.

[0015] Step S5: Through a dynamic coordination mechanism, optimize the scheduling system resources and collaboration of each link according to the urgency of the fault. After the fault repair is completed, feed the case data back to the adaptive maintenance rule base to complete the update. The dynamic coordination mechanism includes hierarchical management of background learning tasks; When the emergency level of the fault is aircraft grounding fault, a rapid response strategy of prioritizing diagnosis and simplifying solutions is activated, and model retraining and large-scale knowledge mining tasks that are not related to the current fault diagnosis are suspended.

[0016] By adopting the above technical solution, and through background task hierarchical management and emergency fault rapid response strategies, the system resource scheduling priority is optimized to ensure efficient handling of emergency faults.

[0017] Secondly, the technical solution of the present invention also provides an aviation electromechanical equipment fault diagnosis and AR guidance system, including a rule base construction module, a fault diagnosis module, a solution evaluation module, a work card generation module, and a coordination feedback module; The rule base construction module is used to acquire multi-source information and fault information of aviation electromechanical equipment, construct a structured knowledge network and an adaptive maintenance rule base based on the multi-source information, and dynamically update the adaptive maintenance rule base through feedback from historical maintenance cases. The dynamic update includes using incremental learning algorithms to adjust the rule confidence in the structured knowledge network and quantitatively updating the weights of coupling relationships in the structured knowledge network based on multi-source information. The fault diagnosis module is used to receive fault information and input it into the two-layer diagnosis engine. The upper-layer fast matching engine filters the rule subgraph, and the lower-layer graph calculation and inference engine performs circular confidence propagation calculation based on the rule subgraph. Simultaneously, it compares with the minimum equipment list database to complete compliance analysis and outputs a diagnosis report with compliance identification. When the lower-level graph computation inference engine performs the circular confidence propagation calculation, it includes evidence fusion and combines compliance identifiers formed by comparing with the minimum equipment list database to rank the causes of failures. Evidence fusion includes the process of propagating fault information to intermediate nodes, the intermediate nodes propagating the fused evidence to fault nodes, and the fault nodes receiving multi-source evidence and updating their confidence. The ranking dimensions include confidence level and compliance status. The compliance status includes prohibition of release and release under restricted conditions.

[0018] The solution evaluation module is used to receive diagnostic reports and input digital twin models calibrated with aircraft condition monitoring data, perform multi-dimensional dynamic weighted simulation evaluation of candidate maintenance solutions, and output optimized maintenance solutions. The work card generation module is used to generate augmented reality work cards with component verification and multi-mode guidance capabilities based on optimized maintenance plans, real-time resource status, and personnel qualifications. The coordination and feedback module is used to optimize the scheduling of system resources and the collaboration of various units based on the urgency of the fault through a dynamic coordination mechanism. After the fault repair is completed, the case data is fed back to the adaptive maintenance rule base for updating.

[0019] The beneficial effects of this invention are as follows: The method and system for fault diagnosis and AR maintenance guidance of aviation electromechanical equipment provided by this invention, by constructing a structured knowledge network and an adaptive maintenance rule base, incrementally learns and adjusts the confidence of rules, updates the edge weights of coupling relationships with multi-source information, and combines historical case feedback to update the rule base, thereby achieving dynamic iteration of maintenance knowledge and accurate adaptation of fault coupling relationships. This solves the problems of lack of dynamic update mechanism for maintenance knowledge and insufficient adaptability of fault coupling relationships. Furthermore, by using a lower-level graph calculation inference engine to simultaneously compare with the minimum equipment list database during the circular confidence propagation calculation, a diagnostic report with compliance indicators is output, achieving fault diagnosis and airworthiness standards. The deep integration of these technologies addresses the issues of insufficient airworthiness integration and low integration of compliance assessment in maintenance solutions. By inputting diagnostic reports into a digital twin model calibrated with aircraft condition monitoring data, multi-dimensional dynamic weighted simulation evaluations are performed on candidate maintenance solutions, enabling pre-verification and effect prediction of maintenance solutions. This solves the problems of lack of pre-simulation verification and difficulty in predicting the execution effect of maintenance solutions. Furthermore, by generating augmented reality work cards with component verification and multi-mode guidance capabilities, combined with a dynamic coordination mechanism to adapt to on-site resources and personnel, multi-mode precise guidance in complex operational scenarios is achieved, solving the problems of insufficient refinement of on-site guidance multi-mode adaptation capabilities and weak support for complex scenarios.

[0020] Furthermore, the design principle of this invention is reliable, the structure is simple, and it has a very wide range of application prospects. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of the aviation electromechanical equipment fault diagnosis and AR maintenance guidance method provided by the present invention.

[0023] Figure 2 This is a schematic diagram of the fault diagnosis and AR maintenance guidance system for aviation electromechanical equipment provided by the present invention. Detailed Implementation

[0024] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0026] Example 1: like Figure 1 As shown, this embodiment of the invention provides a method for fault diagnosis and AR maintenance guidance of aircraft electromechanical equipment, including the following steps: Step S1: Obtain multi-source information and fault information of aircraft electromechanical equipment, construct a structured knowledge network and an adaptive maintenance rule base based on the multi-source information, and dynamically update the adaptive maintenance rule base through feedback from historical maintenance cases. The dynamic update includes using an incremental learning algorithm to adjust the rule confidence in the structured knowledge network and quantitatively updating the weights of coupling relationship edges in the structured knowledge network based on multi-source information. The multi-source information includes historical maintenance case data, system schematic data, system safety assessment report data, and airworthiness directive information. The quantitative update of the coupling relationship edge weights is based on the co-occurrence probability of historical cases, the physical connection tightness of the system schematic, and the common-mode fault relationship in the system safety assessment report.

[0027] In this embodiment, historical maintenance case data is extracted from complete cases spanning the past 5-10 years from the airline's maintenance management system, covering fields such as fault phenomena, diagnostic processes, maintenance measures, and effect verification; system schematic diagram data requires obtaining detailed electromechanical system drawings corresponding to the aircraft model to clarify the physical connection relationships between components; system safety assessment report data uses FMEA (Failure Mode and Effects Analysis) reports to extract common mode fault-related information; and airworthiness directive information is obtained through the Civil Aviation Administration's airworthiness information platform to ensure that the rule base complies with the latest airworthiness requirements.

[0028] The structured knowledge network uses fault type, component name, fault cause, and maintenance measures as nodes.

[0029] The incremental learning algorithm tunes the rule confidence in structured knowledge networks as follows: When validating rules for new cases, press C. new = C old+ α · (1-C old )calculate; When a new case negates the rule, press C. new = C old • β calculation; Among them, C old C represents the confidence level before the rule update. new α represents the confidence level after the rule update, β represents the reward learning rate, and β represents the penalty decay coefficient. α and β are dynamically fine-tuned based on the historical trigger frequency of the rule or the recentity of the case.

[0030] For example, when the historical trigger frequency of the rule is higher than 50 times / year, α=0.3 and β=0.7; when the trigger frequency is lower than 10 times / year, α=0.15 and β=0.85. When the case is a recent case within the last 6 months, both α and β increase by 20%, while cases older than 3 years decrease by 30%.

[0031] In this embodiment, step S1 continuously extracts knowledge from multi-source engineering data. Multi-source engineering data includes, but is not limited to: aircraft maintenance manuals, fault isolation manuals, component maintenance manuals, historical maintenance case libraries, airworthiness directives, service bulletins, system schematics, and system safety assessment reports. Step S1 parses the structured manuals to generate initial diagnostic rules expressed in IF-THEN format, forming the initial framework of the knowledge network. Furthermore, step S1 automatically parses the unstructured regulatory text of newly received airworthiness directives; for example, when receiving an AD directive regarding a potential internal leak in a certain type of actuator, step S1 uses natural language processing technology to extract the faulty component, pattern, effective date, and urgency level, and transforms it into a high-priority diagnostic rule with a mandatory effective date, injecting it into the structured knowledge network in real time to ensure the knowledge base is synchronized with the latest regulations.

[0032] Calculating the coupling effect factor is crucial for accurately characterizing fault coupling relationships and achieving precise diagnosis. The calculation of the coupling effect factor is based on quantification of multi-source information. In this embodiment, taking the hydraulic pump (denoted as node C1) and hydraulic filter (denoted as node C2) as examples, the calculation specifically includes: Statistical analysis of historical co-occurrence frequency: Based on the historical maintenance case library, the number and frequency of co-occurrence cases where C1 failure leads to C2 blockage or vice versa are statistically analyzed to obtain the basic weights of the historical statistical dimensions.

[0033] Extracting the physical connection tightness: Confirm the physical connection relationship between C1 and C2 from the system schematic diagram. If they are directly connected by series pipes, assign a higher first-level weight; if they are indirectly connected through other components, reduce the weight according to the path length.

[0034] Analyze common-mode failure relationships: Query the system security assessment report to confirm whether C1 and C2 share the same critical resource or belong to the same redundant channel. If a common-mode failure relationship exists, assign an additional second-level weight to characterize the implicit functional association; otherwise, do not add an additional weight.

[0035] Introducing a time decay factor: A time decay factor is introduced to reflect the recency of knowledge. The weight of recent case data is set higher than that of earlier cases, so that recent cases contribute more to the update.

[0036] Taking all the above factors into account, the coupling influence factor between C1 and C2 is dynamically updated using the following formula:

[0037] in, and These represent the edge weights before and after the dynamic update, respectively. This represents the historical statistical weight based on historical co-occurrence frequency. Represents the physical connection weights based on the density of physical connections. This represents the security association weight based on common-mode fault relationships. Indicates the time decay factor. The value decreases as time goes on after the case occurred. , , These are configurable normalization coefficients. This is used to balance the contributions of different information sources. The learning rate controls the adjustment range of this update relative to the historical weights.

[0038] Step S2: Input the fault information into the dual-layer diagnostic engine. The upper-layer fast matching engine filters the rule subgraph, and the lower-layer graph calculation and inference engine performs circular confidence propagation calculation based on the rule subgraph. Simultaneously, it compares with the Minimum Equipment List (MEL) database to complete the compliance analysis and outputs a diagnostic report with compliance identification. The circular confidence propagation calculation of the lower-level graph computation inference engine includes evidence fusion, which combines the results of the minimum equipment list database verification to rank the causes of failure; Evidence fusion includes the process of propagating fault information to intermediate nodes, the intermediate nodes propagating the fused evidence to fault nodes, and the fault nodes receiving multi-source evidence and updating their confidence. The ranking dimensions include confidence level and compliance status. The compliance status includes prohibition of release and release under restricted conditions.

[0039] In this embodiment, the belief propagation algorithm is used to calculate the circular belief propagation, and the mathematical expression is:

[0040] in, For factor functions, Let i be a factor node, i be a variable node, and t be the iteration round. This refers to the message passed from factor node c to variable node i during the t-th iteration. The message represents the confidence level for the "value state of variable i" (e.g., "fault occurred / not occurred"), while \ indicates that it does not include the value. For a set, The product; The factor nodes correspond to rule knowledge, and the i variable node contains fault information.

[0041] In this embodiment, the upper-layer fast matching engine filters the structured knowledge network based on the source of fault information (e.g., engine system, power system) and initially associates it with possible causes such as "engine rotor imbalance", "generator bearing wear" and "engine-generator connection shaft failure", forming a rule subgraph.

[0042] The lower-level graph computational inference engine performs circumferential confidence propagation calculations on the rule subgraph. It is assumed that "connecting shaft failure" receives the highest confidence because it can explain two fault phenomena simultaneously. After the circumferential confidence propagation calculation is completed, the lower-level graph computational inference engine immediately accesses the MEL database and compares it with the MEL database. It finds that "engine-generator connecting shaft failure" corresponds to the "No Release" item in the MEL, outputs a diagnostic report with a compliance indicator, lists the confidence rankings, and adds a prominent "MEL: NO-GO" label next to the fault cause in the diagnostic report, linking to the specific clause, prompting maintenance personnel to prioritize troubleshooting this fault before releasing the vehicle.

[0043] Step S3: Input the diagnostic report into the digital twin model calibrated with aircraft condition monitoring data, perform multi-dimensional dynamic weighted simulation evaluation of candidate maintenance solutions, and output the optimized maintenance solution; The dimensions of the multi-dimensional dynamic weighted simulation assessment include estimated working hours, resource costs, technical compliance risks, and secondary damage risks; Among them, dynamic weighting adjusts the weight ratio of each dimension through a preset weight strategy template corresponding to the fault urgency level; The fault level is determined by a dynamic coordination mechanism based on the fault's impact on flight safety, operational efficiency, and real-time scheduling. Pre-defined fault urgency levels include, but are not limited to, the following three levels: Aircraft Grounding Fault (AOG): The fault prevents the aircraft from performing its next scheduled flight mission, requiring immediate grounding for maintenance. Important Level: The fault affects aircraft performance or poses a potential safety risk, but can be addressed with limited delays or after the execution of a subsequent limited flight segment. Routine Level: The fault does not affect the safety of this release and is included in subsequent scheduled maintenance plans.

[0044] Specifically, when the emergency level of the fault is aircraft grounding failure, the weight of estimated working hours and on-site availability of key aviation materials in the simulation evaluation is increased, while the weight of resource cost in the simulation evaluation is decreased.

[0045] In this embodiment, when the fault urgency level is Aircraft Grounding Fault (AOG), the corresponding AOG weight template is adopted. The AOG weight template sets a scaling factor greater than 1 for the weights of the estimated completion time and the on-site availability of key aviation materials, and sets a scaling factor equal to 1 for other dimensions. The scaled weights are calculated based on the AOG weight template and normalized to generate a weight set for the current assessment, thereby automatically improving the assessment weights of the above two dimensions.

[0046] In this embodiment, the entire process data of a real "slow leakage of hydraulic system" case recorded in the aircraft electromechanical equipment is used. The entire process data includes the pressure decay curve and the change of oil volume in the tank. The pipeline resistance parameters and pump efficiency curve in the digital twin model are back-simulated and fitted by gradient descent method or genetic algorithm so that the pressure decay characteristics simulated by the digital twin model have an error of less than 5% with the real data.

[0047] For the diagnosed "hydraulic actuator internal leakage" problem, the digital twin model evaluated "replacing the actuator only" (Option A) and "replacing the actuator and cleaning the system" (Option B). In the standard mode, each dimension (work time, cost, risk) is calculated with standard weights. If this fault is identified as an AOG (Area of ​​Goods) by the dynamic coordination mechanism, the digital twin model automatically switches to the AOG evaluation strategy template. The AOG evaluation strategy template specifies that the scaling factor for estimated completion time is set to 2.5, the scaling factor for critical aircraft parts availability on-site is set to 2.0, and the scaling factor for other dimensions is 1.0. The digital twin model calculates the scaled weights based on the standard base weights and then normalizes them. After this dynamic adjustment, the influence of estimated completion time and critical aircraft parts availability on-site in the overall score is increased, raising the priority score of repair solutions that can be executed more quickly and for which the required aircraft parts are immediately available.

[0048] Step S4: Based on the optimized maintenance plan, real-time resource status, and personnel qualifications, generate an augmented reality work card with component verification and multi-mode guidance capabilities; The component verification of the augmented reality work card adopts image recognition and optical character recognition technology. When the number of consecutive failures of image recognition exceeds the preset threshold, or the confidence level of the results obtained by optical character recognition and image recognition is lower than the preset threshold, it switches to the augmented positioning guidance mode based on ultra-wideband or Bluetooth beacons to provide directional guidance in the augmented reality field of view.

[0049] In this embodiment, the operator (possessing a Class II fuselage signature but no engine-specific authorization) wears an AR device to perform the task of replacing the lubricating oil pump.

[0050] When an augmented reality work card is generated, the operator's qualifications are verified from the personnel qualification database. When the augmented reality work card reaches the step of "removing and installing the engine oil low-pressure pipeline," it is recognized that removing and installing the engine oil low-pressure pipeline requires engine-specific authorization, which the operator does not possess. In the AR interface, the removal and installation of the engine oil low-pressure pipeline is automatically locked and displayed in gray, and a prompt appears: "This operation requires engine-authorized personnel. Please contact authorized personnel."

[0051] When disassembling and assembling the lubricating oil pump, the AR device first attempted to identify the pump body part number using a camera. Due to oil contamination on the pump body surface, the AR device failed to identify the part number. The system then switched to enhanced positioning guidance mode: utilizing the ultra-wideband (UWB) positioning network deployed in aviation electromechanical equipment, a 3D arrow was continuously displayed in the AR interface to guide the operator to the spare parts rack. A 3D outline of the target lubricating oil pump was also displayed floating above the rack, assisting in quickly and accurately selecting it from multiple similar parts.

[0052] When a torque wrench is required for tightening, the AR device verifies compliance by identifying the wrench model. Throughout the entire operation, all step confirmations, part number scanning (success or failure), tool usage, operator information, and timestamps are automatically recorded in an unalterable maintenance log, meeting airworthiness traceability requirements.

[0053] Step S5: Through a dynamic coordination mechanism, optimize the scheduling system resources and collaboration of each link according to the urgency of the fault. After the fault repair is completed, feed the case data back to the adaptive maintenance rule base to complete the update. The dynamic coordination mechanism includes hierarchical management of background learning tasks; When the emergency level of the fault is aircraft grounding fault, a rapid response strategy of prioritizing diagnosis and simplifying solutions is activated, and model retraining and large-scale knowledge mining tasks that are not related to the current fault diagnosis are suspended.

[0054] Example 2: like Figure 2As shown, this embodiment also provides an aircraft electromechanical equipment fault diagnosis and AR maintenance guidance system, including a rule base construction module 1, a fault diagnosis module 2, a solution evaluation module 3, a work card generation module 4, and a coordination feedback module 5; Rule base construction module 1 is used to acquire multi-source information and fault information of aircraft electromechanical equipment, construct a structured knowledge network and an adaptive maintenance rule base based on multi-source information, and dynamically update the adaptive maintenance rule base through feedback from historical maintenance cases. The dynamic update includes using an incremental learning algorithm to adjust the rule confidence in the structured knowledge network and quantitatively updating the weights of coupling relationships in the structured knowledge network based on multi-source information. Fault diagnosis module 2 is used to receive fault information and input it into the dual-layer diagnosis engine. The upper-layer fast matching engine filters the rule subgraph, and the lower-layer graph calculation and inference engine performs circular confidence propagation calculation based on the rule subgraph. Simultaneously, it compares with the minimum equipment list database to complete compliance analysis and outputs a diagnosis report with compliance identification. When the lower-level graph computation inference engine performs the circular confidence propagation calculation, it includes evidence fusion and combines compliance identifiers formed by comparing with the minimum equipment list database to rank the causes of failures. Evidence fusion includes the process of propagating fault information to intermediate nodes, the intermediate nodes propagating the fused evidence to fault nodes, and the fault nodes receiving multi-source evidence and updating their confidence. The ranking dimensions include confidence level and compliance status. The compliance status includes prohibition of release and release under restricted conditions.

[0055] Solution evaluation module 3 is used to receive diagnostic reports and input digital twin models calibrated with aircraft condition monitoring data, perform multi-dimensional dynamic weighted simulation evaluation of candidate maintenance solutions, and output optimized maintenance solutions; Work card generation module 4 is used to generate augmented reality work cards with component verification and multi-mode guidance capabilities based on optimized maintenance plans, real-time resource status and personnel qualifications. The coordination and feedback module 5 is used to optimize the scheduling of system resources and the collaboration of each unit according to the urgency of the fault through a dynamic coordination mechanism. After the fault repair is completed, the case data is fed back to the adaptive maintenance rule base to complete the update.

[0056] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. The methods disclosed in the embodiments are described simply because they correspond to the systems disclosed in the embodiments; relevant details can be found in the method section.

[0057] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0058] In the embodiments provided by this invention, it should be understood that the disclosed systems, methods, and approaches can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or units may be electrical, mechanical, or other forms.

[0059] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0060] In addition, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit.

[0061] Similarly, in the various embodiments of the present invention, each processing unit can be integrated into a functional module, or each processing unit can exist physically, or two or more processing units can be integrated into a functional module.

[0062] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0063] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0064] The above-disclosed embodiments are merely preferred embodiments of the present invention, but the present invention is not limited thereto. Any non-creative variations that can be conceived by those skilled in the art, as well as any improvements and modifications made without departing from the principles of the present invention, should fall within the protection scope of the present invention.

Claims

1. A method for fault diagnosis and AR maintenance guidance of aircraft electromechanical equipment, characterized in that, Includes the following steps: Step S1: Obtain multi-source information and fault information of aircraft electromechanical equipment, construct a structured knowledge network and an adaptive maintenance rule base based on the multi-source information, and dynamically update the adaptive maintenance rule base through feedback from historical maintenance cases. The dynamic update includes using an incremental learning algorithm to adjust the rule confidence in the structured knowledge network and quantitatively updating the weights of coupling relationship edges in the structured knowledge network based on multi-source information. Step S2: Input the fault information into the dual-layer diagnostic engine. The upper-layer fast matching engine filters the rule subgraph, and the lower-layer graph calculation and inference engine performs circular confidence propagation calculation based on the rule subgraph. Simultaneously, it compares with the minimum equipment list database to complete the compliance analysis and outputs a diagnostic report with compliance identification. The circular confidence propagation calculation of the lower-level graph computation inference engine includes evidence fusion, which combines the results of the minimum equipment list database verification to rank the causes of failure; Evidence fusion includes the process of propagating fault information to intermediate nodes, the intermediate nodes propagating the fused evidence to fault nodes, and the fault nodes receiving multi-source evidence and updating their confidence. The ranking dimensions include confidence level and compliance status, with compliance status including prohibition of passage and passage allowed under restricted conditions; The confidence propagation algorithm is used to calculate the circular confidence propagation. The mathematical expression is as follows: in, For factor functions, Let i be a factor node, i be a variable node, and t be the iteration round. This refers to the message passed from factor node a to variable node i in the t-th iteration. The confidence level of the message is indicated by the backslash "\", which indicates that the message is excluded. For a set, The product; The factor nodes correspond to rule knowledge, and the i-variable node contains fault information; Step S3: Input the diagnostic report into the digital twin model calibrated with aircraft condition monitoring data, perform multi-dimensional dynamic weighted simulation evaluation of candidate maintenance solutions, and output the optimized maintenance solution; Step S4: Based on the optimized maintenance plan, real-time resource status, and personnel qualifications, generate an augmented reality work card with component verification and multi-mode guidance capabilities; Step S5: Through a dynamic coordination mechanism, optimize the scheduling system resources and collaboration of each link according to the urgency of the fault. After the fault repair is completed, feed the case data back to the adaptive maintenance rule base to complete the update.

2. The method for fault diagnosis and AR maintenance guidance of aircraft electromechanical equipment according to claim 1, characterized in that, In step S1, the multi-source information includes historical maintenance case data, system schematic diagram data, system safety assessment report data, and airworthiness directive information. The quantitative update of the coupling relationship edge weights is based on the statistical co-occurrence probability of historical cases, the physical connection tightness of the system schematic diagram, and the common-mode fault relationship in the system safety assessment report.

3. The method for fault diagnosis and AR maintenance guidance of aircraft electromechanical equipment according to claim 2, characterized in that, In step S1, the incremental learning algorithm tunes the rule confidence in the structured knowledge network as follows: When validating rules for new cases, press C. new = C old + α · (1-C old )calculate; When a new case negates the rule, press C. new = C old • β calculation; Among them, C old C represents the confidence level before the rule update. new α represents the confidence level after the rule update, β represents the reward learning rate, and β represents the penalty decay coefficient. α and β are dynamically fine-tuned based on the historical trigger frequency of the rule or the recentity of the case.

4. The method for fault diagnosis and AR maintenance guidance of aircraft electromechanical equipment according to claim 1, characterized in that, In step S3, the dimensions of the multi-dimensional dynamic weighted simulation evaluation include estimated working hours, resource costs, technical compliance risks, and secondary damage risks; Among them, dynamic weighting adjusts the weight ratio of each dimension through a preset weight strategy template corresponding to the fault urgency level.

5. The method for fault diagnosis and AR maintenance guidance of aircraft electromechanical equipment according to claim 4, characterized in that, The fault level is determined by the dynamic coordination mechanism based on the degree of impact of the fault on flight safety and operational efficiency, as well as the real-time scheduling situation. The pre-defined emergency levels for faults include at least the following three levels: Aircraft grounding fault: The fault prevents the aircraft from performing the next scheduled flight mission and requires immediate grounding for maintenance; Importance level: The malfunction affects aircraft performance or poses a potential safety risk, but can be addressed with limited delay or after the execution of a subsequent limited segment; Standard level: The fault does not affect the safety of this release and will be included in the subsequent regular maintenance plan.

6. The method for fault diagnosis and AR maintenance guidance of aircraft electromechanical equipment according to claim 1, characterized in that, In step S4, the component verification of the augmented reality work card adopts image recognition and optical character recognition technology. When the number of consecutive failures of image recognition exceeds a preset threshold, or the confidence level of the results obtained by optical character recognition and image recognition is lower than a preset threshold, the system switches to the augmented positioning guidance mode based on ultra-wideband or Bluetooth beacons.

7. The method for fault diagnosis and AR maintenance guidance of aircraft electromechanical equipment according to claim 1, characterized in that, In step S5, the dynamic coordination mechanism includes hierarchical management of background learning tasks; When the emergency level of the fault is aircraft grounding fault, a rapid response strategy of prioritizing diagnosis and simplifying solutions is activated, and model retraining and large-scale knowledge mining tasks that are not related to the current fault diagnosis are suspended.

8. An aircraft electromechanical equipment fault diagnosis and AR maintenance guidance system, characterized in that, The system is used to implement the aircraft electromechanical equipment fault diagnosis and AR maintenance guidance method as described in any one of claims 1 to 7; The system includes: a rule base construction module, a fault diagnosis module, a solution evaluation module, a work card generation module, and a coordination and feedback module; The rule base construction module is used to acquire multi-source information and fault information of aviation electromechanical equipment, construct a structured knowledge network and an adaptive maintenance rule base based on the multi-source information, and dynamically update the adaptive maintenance rule base through feedback from historical maintenance cases. The dynamic update includes using incremental learning algorithms to adjust the rule confidence in the structured knowledge network and quantitatively updating the weights of coupling relationships in the structured knowledge network based on multi-source information. The fault diagnosis module is used to receive fault information and input it into the two-layer diagnosis engine. The upper-layer fast matching engine filters the rule subgraph, and the lower-layer graph calculation and inference engine performs circular confidence propagation calculation based on the rule subgraph. Simultaneously, it compares with the minimum equipment list database to complete compliance analysis and outputs a diagnosis report with compliance identification. The solution evaluation module is used to receive diagnostic reports and input digital twin models calibrated with aircraft condition monitoring data, perform multi-dimensional dynamic weighted simulation evaluation of candidate maintenance solutions, and output optimized maintenance solutions. The work card generation module is used to generate augmented reality work cards with component verification and multi-mode guidance capabilities based on optimized maintenance plans, real-time resource status, and personnel qualifications. The coordination and feedback module is used to optimize the scheduling of system resources and the collaboration of various units based on the urgency of the fault through a dynamic coordination mechanism. After the fault repair is completed, the case data is fed back to the adaptive maintenance rule base for updating.

9. The aircraft electromechanical equipment fault diagnosis and AR maintenance guidance system according to claim 8, characterized in that, When the lower-level graph computation inference engine of the fault diagnosis module performs the circular confidence propagation calculation, it includes evidence fusion and sorts the causes of faults by combining the compliance identifiers formed by comparing with the minimum equipment list database. Evidence fusion includes the process of propagating fault information to intermediate nodes, the intermediate nodes propagating the fused evidence to fault nodes, and the fault nodes receiving multi-source evidence and updating their confidence. The ranking dimensions include confidence level and compliance status, with compliance status including prohibition of release and release under restricted conditions.

Images