Digital thread based aviation equipment full life cycle management method and system

Abstract

The present application relates to the technical field of equipment life cycle management, and specifically includes an aviation equipment full life cycle management method and system based on digital threads, the method comprising: acquiring process parameters of each stage, determining first digital thread master data and a first layer of associated test sequences, second digital thread master data and a second layer of associated test sequences, mapping the deviation of the physical equipment state and the virtual simulation state, and identifying the performance difference point under the cross-stage data correlation link; setting a management strategy for collaborative optimization. The technical problem of being unable to identify the cross-stage performance difference point equipment performance fluctuation root cause positioning in real time, causing the reliability of aviation equipment to be limited is solved, the first and second digital thread master data and the associated test sequences are constructed, the deviation of the physical equipment and the virtual simulation state is accurately mapped, the management strategy is set in combination with the double thread data and the performance difference point, and full life cycle collaborative optimization is performed, and the technical effect of effectively improving the equipment reliability is achieved.

Description

Technical Field

[0001] This invention relates to the technical field of equipment lifecycle management, specifically to a method and system for full lifecycle management of aviation equipment based on digital threads. Background Technology

[0002] The entire life cycle of aviation equipment covers multiple stages, including design, manufacturing, service, and maintenance. The process parameters, performance status, and maintenance records of each stage are closely related and directly affect the reliability, safety, and service life of the equipment. Digital thread technology, by constructing a mapping link between the virtual and physical worlds, makes it possible to achieve unified management and in-depth mining of aviation equipment life cycle data, and is a key technical support for improving equipment management and ensuring flight safety.

[0003] Current lifecycle management of aviation equipment relies on empirical functions for status assessment and parameter optimization. It lacks a dynamic mapping mechanism between physical equipment and virtual simulation, making it impossible to identify performance differences across stages in real time. Furthermore, management strategies are mostly formulated for a single stage and lack the ability to coordinate and optimize the entire process. This results in limited reliability of aviation equipment and makes it difficult to meet the needs of modern aviation equipment for refined lifecycle management.

[0004] In summary, existing technologies suffer from the following problems: the full lifecycle management of aviation equipment relies on empirical functions for state assessment and parameter optimization, lacks the ability for collaborative optimization throughout the entire process, and thus cannot identify the root causes of performance fluctuations across different stages in real time, resulting in limited reliability of aviation equipment. Summary of the Invention

[0005] This application provides a digital thread-based method and system for the full lifecycle management of aviation equipment, aiming to solve the technical problem that the existing aviation equipment full lifecycle management relies on empirical functions for state assessment and parameter optimization, lacks full-process collaborative optimization capabilities, and thus cannot identify cross-stage performance differences in real time, resulting in unclear positioning of the root cause of equipment performance fluctuations and limited reliability of aviation equipment.

[0006] In view of the above problems, the technical solution to achieve the present application is as follows:

[0007] In a first aspect, this application provides a method for managing the entire lifecycle of aerospace equipment based on digital threads. The method includes: acquiring process parameters for each stage of the aerospace equipment's lifecycle, including machining accuracy data, assembly tolerances, and process execution records; determining first digital thread master data and a first-layer associated test sequence based on the process parameters for each stage, combined with material properties and performance parameters; determining second digital thread master data and a second-layer associated test sequence based on fault modes and maintenance parameters; mapping the deviation between the physical equipment state and the virtual simulation state using a digital thread analysis engine based on an equipment knowledge graph, and identifying performance difference points under cross-stage data association links; and setting management strategies based on the first digital thread master data and the first-layer associated test sequence, the second digital thread master data and the second-layer associated test sequence, and the performance difference points, to perform collaborative optimization throughout the entire lifecycle of the aerospace equipment.

[0008] Preferably, the first digital thread master data includes a 3D model identifier, material batch number, heat treatment parameters, and processing equipment ID; the test cases in the design stage perform strength simulation tests, and the test cases in the manufacturing stage perform dimensional tolerance tests, wherein the strength simulation test data and the dimensional tolerance test data are both associated with the first digital thread master data through timestamps.

[0009] Preferably, the second digital thread master data includes fault codes, spare parts replacement records, maintenance work order numbers, and testing equipment calibration certificate numbers; the service phase test cases perform vibration modal testing, and the maintenance phase test cases perform leak detection testing. The second-level associated test sequence uses a two-dimensional record of Pass / Fail + quantitative value, and non-conforming items automatically trigger the second digital thread master data marking.

[0010] Preferably, the digital thread analysis engine employs a graph neural network, inputting the physical equipment state vector and the virtual simulation vector to determine the cosine similarity; based on the cosine similarity, the source of deviation is located through the node attention mechanism of the graph neural network.

[0011] Preferably, the cosine similarity is used to perform preliminary screening in combination with a preset similarity threshold range. The attention weights of each node that exceed the similarity threshold are enhanced and abnormal nodes are marked. Starting from the abnormal nodes, the deviation between the physical equipment state and the virtual simulation state is evaluated on the downstream performance parameters through the edge weights of the equipment knowledge graph. When the influence exceeds the upper limit of the allowable influence, the deviation source deep tracing instruction is activated.

[0012] Preferably, according to the deviation source deep tracing instruction, a graph neural network is used to traverse the upstream process parameter chain of the abnormal node in reverse; based on the upstream process parameter chain of the abnormal node, the root link of the deviation is located, and a deep tracing report containing the cause of the deviation, the scope of influence, and correction suggestions is generated.

[0013] Preferably, based on the root cause of the deviation in the deep traceability report, the corresponding stage process parameters are extracted to construct a cross-stage data association link; the deviation transmission coefficient of multiple stage process parameters in the cross-stage data association link is evaluated through the digital thread analysis engine; and the performance difference association link is identified according to the deviation transmission coefficient of multiple stage process parameters in the cross-stage data association link.

[0014] Preferably, the cumulative impact index of deviations of process parameters at each stage is obtained; a deviation impact propagation matrix is ​​constructed, wherein the matrix elements of the deviation impact propagation matrix represent the weighting coefficients of the mutual influence between deviations of process parameters at different stages; and the deviation transmission coefficient is determined by matrix multiplication based on the cumulative impact index of deviations and the deviation impact propagation matrix.

[0015] Preferably, the performance difference points include potential difference points and existing difference points; the management strategy is configured by cross-iteration of the preventive cross-stage collaborative parameter combination corresponding to the potential difference points and the corrective cross-stage collaborative parameter combination corresponding to the existing difference points, and binding them with the first digital thread master data and the second digital thread master data.

[0016] In a second aspect, this application provides a digital thread-based full lifecycle management system for aviation equipment. The system includes: a process parameter acquisition module, which acquires process parameters for each stage of the aviation equipment's lifecycle, including machining accuracy data, assembly tolerances, and process execution records; a test sequence determination module, which determines first digital thread master data and a first-layer associated test sequence based on the process parameters for each stage, combined with material properties and performance parameters, and determines second digital thread master data and a second-layer associated test sequence based on fault modes and maintenance parameters; a performance difference point identification module, which maps the deviation between the physical equipment state and the virtual simulation state using a digital thread analysis engine based on the equipment knowledge graph, and identifies performance difference points under cross-stage data association links; and a collaborative optimization module, which sets management strategies based on the first digital thread master data and the first-layer associated test sequence, the second digital thread master data and the second-layer associated test sequence, and the performance difference points to perform collaborative optimization throughout the entire lifecycle of the aviation equipment.

[0017] In summary, one or more technical solutions provided in this application achieve the following: acquiring process parameters for each stage of the entire lifecycle of aviation equipment; constructing master data for the first and second digital threads and associated test sequences; structurally linking and connecting data from each stage; using a digital thread analysis engine based on equipment knowledge graphs to accurately map the deviation between physical equipment and virtual simulation states; identifying cross-stage performance differences in real time; and combining dual-thread data with performance differences to set management strategies and perform full lifecycle collaborative optimization, effectively improving the technical effect of equipment reliability. Attached Figure Description

[0018] Figure 1 This application provides a flowchart illustrating a digital thread-based approach to the full lifecycle management of aviation equipment.

[0019] Figure 2 This application provides a schematic diagram of the structure of a digital thread-based aviation equipment lifecycle management system.

[0020] Explanation of reference numerals in the attached diagram: Process parameter acquisition module M100, Test sequence determination module M200, Performance difference point identification module M300, Collaborative optimization module M400. Detailed Implementation

[0021] Example 1: The present application will be described in detail below with reference to the accompanying drawings, as follows... Figure 1 As shown, this application provides a method for full lifecycle management of aviation equipment based on digital threads, wherein the method includes:

[0022] S1: Obtain process parameters for each stage of the entire lifecycle of the aviation equipment, including machining accuracy data, assembly tolerances, and process execution records; S2: Based on the process parameters for each stage, determine the first digital thread master data and the first layer of associated test sequences by combining material properties and performance parameters, and determine the second digital thread master data and the second layer of associated test sequences by combining fault modes and maintenance parameters.

[0023] Specifically, process parameters cover key technical indicators of aerospace equipment at each stage of design, manufacturing, service, and maintenance, such as machining accuracy data (including dimensional tolerances, geometric tolerances, etc.), assembly tolerances (such as fit clearances, coaxiality, etc.), and process execution records (such as heat treatment temperature-time curves, welding parameter records, etc.). Among them, the first digital thread master data and the second digital thread master data correspond to the core datasets of the equipment under different management dimensions. The former focuses on material properties (such as material strength, coefficient of thermal expansion) and performance parameters (such as tensile strength, yield strength) in the design and manufacturing process, while the latter is associated with failure modes (such as fatigue crack propagation modes, corrosion types) and maintenance parameters (such as maintenance cycles, replacement part life) in the service and maintenance stages. The associated test sequence refers to a series of test procedures and specifications corresponding to the process parameters at each stage, used to verify the rationality of the process parameters and the conformity of the equipment performance.

[0024] Execution steps: Collect process parameters for each stage of the entire life cycle of aviation equipment. The amount of data is huge and the scope is wide. For example, in the manufacturing stage, the machining accuracy data of a single key component may include dimensional tolerances of dozens of inspection points (taking ±0.01mm accuracy control as an example). Assembly tolerances involve the fitting accuracy between multiple components. Based on these process parameters, the first digital thread master data is determined by combining material properties (such as the yield strength of titanium alloy materials can reach 900MPa) and performance parameters. Its role is to build the digital mapping foundation for equipment in the design and manufacturing stages. At the same time, the second digital thread master data is determined by combining failure modes (such as fatigue cracks in aero-engine blades starting from the blade root, with a propagation rate of about 0.05mm / flight hour) and maintenance parameters to serve intelligent management in the service and maintenance stages.

[0025] The construction of associated test sequences ensures the quality and validity of data at each stage. For example, strength simulation tests are conducted during the design stage (simulating the stress distribution of equipment under 1.5 times the maximum working load), and dimensional tolerance tests are conducted during the manufacturing stage. Preferably, laser scanning technology is used for high detection efficiency. These test data are precisely associated with the corresponding digital thread master data through timestamps, realizing data traceability and synchronous updates, laying a solid data foundation for subsequent full life cycle collaborative optimization.

[0026] S3: Based on the digital thread analysis engine of the equipment knowledge graph, map the deviation between the physical equipment state and the virtual simulation state, and identify the performance difference points under the cross-stage data association link; S4: Through the first digital thread master data and the first layer of association test sequence, the second digital thread master data and the second layer of association test sequence, and in combination with the performance difference points, set management strategies to carry out collaborative optimization throughout the entire life cycle of aviation equipment.

[0027] Specifically, the equipment knowledge graph refers to a semantic network model built based on the full lifecycle data of aviation equipment. It integrates information such as process parameters, performance indicators, and failure modes at each stage of the equipment, and represents entities and their relationships in the form of nodes and edges. For example, the processing accuracy data, material properties, and common failure modes of aero-engine blades are used as nodes, and the relationships between them (such as the influence of material properties on failure modes) are used as edges. The digital thread analysis engine is a computing module that runs on the equipment knowledge graph. It uses graph neural network and other technologies to quantitatively analyze the state mapping between physical equipment and virtual simulation. Here, deviation refers to the difference between the actual state vector of physical equipment (including real-time operating parameters such as temperature and pressure) and the state vector of virtual simulation (based on model-predicted parameters). It is quantified by indicators such as cosine similarity. For example, a cosine similarity of less than 0.95 indicates a significant deviation.

[0028] Cross-stage data correlation links refer to the logical correlation paths between process parameters at each stage from design to maintenance. For example, material strength parameters in the design stage affect fatigue life in the service stage through heat treatment processes in the manufacturing stage. Performance difference points are the specific locations in these correlation links where equipment performance fluctuates due to process deviations or parameter drifts. For example, a connector is designed to withstand a load of 1000N, but its load-bearing capacity is actually detected to have decreased to 800N in the service stage. This difference point is thus identified.

[0029] Execution steps: Relying on the digital thread analysis engine of the equipment knowledge graph, the real-time state data of the physical equipment (such as vibration frequency and pressure distribution during flight) is mapped and compared with the predicted state of the virtual simulation model. Taking the engine blade as an example, vibration frequency data (sampling frequency 10kHz) is collected in real time by 20 sensors arranged on the blade and compared with the vibration frequency predicted by the virtual simulation model based on the blade material properties (such as elastic modulus 210GPa) and geometry. The calculated cosine similarity is 0.89, which is significantly lower than the normal threshold of 0.95, thus identifying the deviation in the blade vibration performance.

[0030] Further edge weight analysis of the knowledge graph revealed a correlation between this deviation and the blade machining accuracy during the manufacturing stage (dimensional tolerance ±0.02mm exceeding ±0.05mm) and the thermal shock parameters during the service stage (temperature change rate 10℃ / s exceeding the design value 8℃ / s). A cross-stage data association link was constructed, and performance difference points were accurately located. Combining the master data of the first and second digital threads and the corresponding test sequences, management strategies were set for these performance difference points. For example, for blade vibration performance differences caused by machining deviations, machining parameters during the manufacturing stage were adjusted (such as optimizing the tool speed to 2000rpm and adjusting the feed rate to 0.1mm / r), and the frequency of blade vibration detection was increased in the subsequent maintenance stage. At the same time, the strength simulation test in the first-layer association test sequence was used to verify the impact of the modified machining parameters on the blade strength, and the sealing leak detection test in the second-layer association test sequence was used to ensure the effectiveness of maintenance measures. This achieved collaborative optimization of the entire life cycle of aerospace equipment and improved the reliability of engine blades.

[0031] Furthermore, by combining material properties and performance parameters to determine the first digital thread master data and the first layer of associated test sequences, the method of this application includes:

[0032] The first digital thread master data includes a 3D model identifier, material batch number, heat treatment parameters, and processing equipment ID; the test cases in the design stage perform strength simulation tests, and the test cases in the manufacturing stage perform dimensional tolerance tests. The strength simulation test data and the dimensional tolerance test data are both associated with the first digital thread master data through timestamps.

[0033] Specifically, the first digital thread master data refers to a set of core datasets constructed during the design and manufacturing phases of aerospace equipment. This includes 3D model identifiers (used to uniquely identify the equipment's geometry and structure), material batch numbers (for tracing material origin and quality), heat treatment parameters (such as temperature-time curves, affecting material properties), and processing equipment IDs (used to track the status and parameters of processing equipment). Design-phase test cases refer to a series of strength simulation test schemes developed for the equipment design model, such as simulating the stress distribution of the equipment under extreme loads. Manufacturing-phase test cases are dimensional tolerance test schemes developed for the actual manufacturing process, such as checking whether the dimensions of parts are within tolerance ranges. Timestamp association refers to establishing a precise correspondence between the data generated during the testing process (such as strength simulation results and dimensional inspection results) and the first digital thread master data according to time sequence, ensuring data traceability and synchronization.

[0034] Execution steps: Determine the specific content of the first digital thread master data. For example, in the design phase of aerospace structural components, the 3D model is identified as ABC123, the material batch number is MAT-xxxx year xx month xx day, the heat treatment parameters are 550℃ for 2 hours, and the processing equipment ID is MACH-001. During the design phase, execute strength simulation test cases to simulate the stress distribution of the structural components under 1.5 times the maximum working load and generate strength simulation test data. These data are associated with the first digital thread master data through timestamps (accurate to the second) to ensure that the test results can accurately reflect the performance of a specific batch of materials and a specific heat treatment process.

[0035] During the manufacturing phase, dimensional tolerance test cases are executed, and laser scanning technology is used to inspect the dimensions of the processed structural parts. The dimensional tolerance test data is also linked to the master data of the first digital thread through timestamps to achieve traceability of the processing process. This ensures the integrity and consistency of data in the design and manufacturing phases, providing accurate data support for subsequent performance evaluation and optimization. For example, by analyzing the linked data, it was found that the strength simulation results of a certain batch of materials (MAT-xxxx year xx month xx day) under specific heat treatment parameters (550℃ for 2 hours) deviated from the actual test results. The heat treatment process parameters can be adjusted in time (such as increasing the holding time to 2.5 hours) to optimize the manufacturing process, thereby improving the reliability and quality stability of aerospace equipment.

[0036] Furthermore, by combining fault modes and maintenance parameters to determine the second digital thread master data and the second-level associated test sequence, the method of this application includes:

[0037] The second digital thread master data includes fault codes, spare parts replacement records, maintenance work order numbers, and testing equipment calibration certificate numbers. During the service phase, test cases perform vibration modal testing, and during the maintenance phase, test cases perform leak detection testing. The second-level associated test sequence uses a two-dimensional record of Pass / Fail + quantitative value. Non-conforming items automatically trigger the second digital thread master data to mark them.

[0038] Specifically, the second digital thread master data refers to the key data set during the service and maintenance phases of aviation equipment. This includes fault codes (used to identify specific fault types, such as ENG-OIL-P-01 corresponding to low engine oil pressure), spare parts replacement records (recording the part numbers, replacement times, etc.), maintenance work order numbers (used to track the execution of each maintenance task), and calibration certificate numbers for testing equipment (ensuring the accuracy of testing equipment meets standards, such as CAL-VIB-xxxx year xx month xx day for a vibration detector). Service phase test cases refer to tests conducted on the equipment in actual operating environments. The system can detect various testing methods, such as vibration modal testing, which is used to evaluate the vibration characteristics of equipment under different working conditions; maintenance phase test cases are testing methods developed for key performance indicators such as sealing during equipment maintenance; Pass / Fail + quantitative value dual-dimensional recording means that when recording test results, not only is the test passed (Pass / Fail) determined, but also the specific test values ​​(such as the specific value of the leakage rate in the sealing leak detection test) are recorded to more comprehensively reflect the equipment performance; when the test result is unqualified, the system automatically triggers the second digital thread master data marking, which means that the system automatically marks the corresponding second digital thread master data to quickly locate the problem.

[0039] Execution steps: Clarify the specific composition of the second digital thread master data. For example, in the maintenance record of an aircraft engine, the fault code ENG-OIL-P-01 indicates that the lubricating oil pressure is below the normal range; the spare parts replacement record shows that the lubricating oil pump (spare part number SP-001) was replaced; the maintenance work order number is MW--xxxx year xx month xx day-001, recording the maintenance log, time, and other information of this maintenance task; the calibration certificate number of the testing equipment is CAL-VIB--xxxx year xx month xx day, indicating that the vibration testing equipment has been calibrated and is within the validity period; during the service phase, execute vibration modal test cases by installing 15 vibration sensors on the engine to collect vibration frequency and amplitude data of the engine at different speeds (from 50% thrust to 100% thrust) and evaluate whether its vibration modes meet the design requirements.

[0040] During the maintenance phase, leak detection test cases are performed, using a helium mass spectrometer to test the engine's fuel system, achieving a detection accuracy of 1×10⁻⁶. -6 Pa·m 3 / s; Test results are recorded using a dual-dimensional approach of Pass / Fail + quantitative values. For example, the vibration modal test result is Pass, with a vibration frequency of 125.3Hz ± 0.5Hz; the leak detection test result is Fail, with a leakage rate of 3.5 × 10⁻⁶. -5 Pa·m 3 / s. When a non-conformity occurs (such as a leak detection test failure), it is automatically marked in the corresponding position in the second digital thread master data to quickly locate the problem. In this way, key performance information of equipment during service and maintenance is recorded and marked in real time and accurately, providing strong support for subsequent performance evaluation, fault diagnosis and maintenance strategy optimization, and effectively improving the reliability and maintenance efficiency of the equipment.

[0041] Furthermore, based on the digital thread analysis engine of the equipment knowledge graph, the method of this application includes:

[0042] The digital thread analysis engine employs a graph neural network, inputting the physical equipment state vector and the virtual simulation vector to determine the cosine similarity; based on the cosine similarity, it locates the source of deviation through the node attention mechanism of the graph neural network.

[0043] Specifically, the digital thread analysis engine employs graph neural network technology, a deep learning model specifically designed for processing graph-structured data, which can effectively capture the complex relationships between nodes; the physical equipment state vector refers to quantifying the real-time operating parameters of physical equipment (such as temperature, pressure, vibration frequency, etc.) into a multi-dimensional vector; the virtual simulation vector refers to the multi-dimensional vector of equipment state predicted by the virtual simulation model based on design parameters and physical laws; cosine similarity is an index that measures the similarity between two vectors, with a value range of [-1, 1]. The closer the value is to 1, the higher the similarity. It is commonly used in text similarity, image similarity, and, here, state vector similarity analysis; the node attention mechanism is an attention mechanism in graph neural networks that highlights nodes that are more important for specific tasks (such as deviation source localization) by assigning different weights to different nodes.

[0044] Execution steps: Using the graph neural network in the digital thread analysis engine, input the physical equipment state vector and the virtual simulation vector. For example, in the real-time monitoring of an aero-engine, the physical equipment state vector contains parameters such as the engine's real-time temperature (850℃), pressure (12MPa), and vibration frequency (125Hz), while the virtual simulation vector is based on the engine design model and predicts parameters such as temperature (840℃), pressure (11.8MPa), and vibration frequency (123Hz) under the same operating conditions. By calculating the cosine similarity between the two vectors, the cosine similarity is found to be 0.85, which is lower than the normal threshold of 0.9, indicating a significant deviation.

[0045] The node attention mechanism of a graph neural network was used to locate the source of deviation. In the node attention calculation, the temperature node was found to have the highest attention weight, reaching 0.65 (while other nodes such as pressure and vibration frequency had weights of 0.2 and 0.15, respectively), indicating that temperature difference is the main source of deviation. Further analysis revealed that the high temperature of the physical equipment was due to the actual cooling airflow (45 kg / s) being lower than the flow rate assumed in the virtual simulation model (50 kg / s), resulting in insufficient cooling efficiency. This method allows for precise identification of the root cause of equipment performance deviations, providing a clear direction for subsequent optimization and adjustments, such as adjusting cooling system parameters or optimizing the cooling airflow assumptions in the virtual simulation model, thereby improving the equipment's operational performance and reliability.

[0046] Furthermore, this application's method utilizes the node attention mechanism of a graph neural network to locate the source of deviation, and includes:

[0047] Using the cosine similarity and a preset similarity threshold range for initial screening, attention weights of nodes exceeding the similarity threshold are enhanced and abnormal nodes are marked. Starting from the abnormal nodes, the influence of the deviation between the physical equipment state and the virtual simulation state on downstream performance parameters is evaluated through the edge weights of the equipment knowledge graph. When the influence exceeds the upper limit of the allowable influence, the deviation source deep tracing instruction is activated.

[0048] Specifically, cosine similarity is used to quantify the similarity between the physical equipment state vector and the virtual simulation vector. It is calculated by dividing the dot product of the two vectors by the product of their magnitudes, and its value ranges from -1 to 1. The closer the value is to 1, the higher the similarity. The preset similarity threshold range is a cosine similarity range pre-set based on the normal operating state of the equipment, usually between 0.9 and 1. Exceeding this range indicates an abnormal deviation. The similarity threshold value is the lower limit of the threshold range. When the cosine similarity is lower than this value, further analysis of the corresponding node is triggered. Node attention weight enhancement analysis refers to the process of adding attention weights to abnormal nodes in the graph neural network. The intention weight is amplified to more accurately locate the source of deviation; the edge weight of the equipment knowledge graph reflects the tightness of the relationship between nodes in the knowledge graph, and the larger the weight, the stronger the relationship; downstream performance parameters refer to the subsequent key performance indicators affected by the deviation, such as engine thrust and fuel efficiency; the upper limit of allowable influence is the maximum allowable value of the deviation's impact on downstream performance parameters, exceeding this value indicates that the deviation's impact on equipment performance is too serious and further measures are required; the deviation source deep tracing command is a command used to deeply analyze the root cause of the deviation, and its activation will start a deeper level of data tracing and analysis process.

[0049] Execution steps: The cosine similarity is used to quantitatively evaluate the matching degree between the physical equipment state and the virtual simulation state. For example, in the real-time monitoring of an aero-engine, the obtained cosine similarity is 0.85, which is lower than the preset similarity threshold of 0.9, indicating a significant deviation. Enhancement analysis is performed on each node that exceeds the similarity threshold (set to 0.8). It is found that the attention weight of the temperature node increases from the initial 0.6 to 0.8, and the attention weight of the pressure node increases from 0.5 to 0.7, indicating that these two nodes may be the main sources of deviation. The nodes are identified and marked as anomalous nodes. Starting from these anomalous nodes, the edge weights of the equipment knowledge graph are used for analysis. The larger the edge weight, the stronger the association. If the edge weight between the temperature node and the downstream performance parameter engine thrust is 0.9, and the edge weight between the pressure node and fuel efficiency is 0.85, the influence of temperature deviation on engine thrust is found to be 0.76 (0.8 × 0.95, where 0.95 is the temperature deviation coefficient), and the influence of pressure deviation on fuel efficiency is found to be 0.68 (0.7 × 0.97, where 0.97 is the pressure deviation coefficient).

[0050] When the impact exceeds the upper limit of the allowable impact, the deviation source deep tracing command is automatically activated, and the deep tracing process for the root cause of the deviation is initiated to further analyze the specific cause of the deviation. For example, the temperature deviation may be caused by a cooling system failure, and the pressure deviation may be caused by a decline in the performance of the fuel pump. Through this deep tracing, the root cause of the problem can be accurately located, providing a clear direction for subsequent optimization and adjustment, thereby improving the operating performance and reliability of the equipment.

[0051] Furthermore, the method for activating the deviation source depth tracing command in this application includes:

[0052] According to the deviation source deep tracing instruction, a graph neural network is used to traverse the upstream process parameter chain of the abnormal node in reverse; based on the upstream process parameter chain of the abnormal node, the root link of the deviation is located, and a deep tracing report containing the cause of the deviation, the scope of impact, and correction suggestions is generated.

[0053] Specifically, the deviation source deep tracing command is a system command that, once activated, triggers a deep analysis process of the deviation root cause. Graph neural network reverse traversal refers to starting from a marked abnormal node and tracing back along the edges of the knowledge graph to the upstream process parameter chain to locate the initial link causing the deviation. For example, if the abnormal node is an engine overheating during service, the reverse traversal will trace back to the cooling system processing parameters during the manufacturing stage, the cooling channel geometry model during the design stage, etc. The upstream process parameter chain refers to a link composed of a series of interrelated process parameters from design to manufacturing and then to service. These parameters collectively affect the final performance of the equipment. The deviation root link refers to the specific process link where the root cause of the deviation lies, such as a deviation in heat treatment parameters during the processing of a certain component. The deep tracing report includes the specific cause of the deviation (such as dimensional deviation caused by processing equipment failure), the scope of the deviation's impact (which systems or performance indicators are affected), and corrective suggestions for the deviation (such as adjusting processing parameters or replacing faulty equipment).

[0054] Execution steps: Based on the deviation source depth tracing instruction, a graph neural network is used to traverse the upstream process parameter chain backward from the abnormal node. For example, in the monitoring of aero-engines, the abnormal node is found to be the engine vibration frequency exceeding the standard. The backward traversal reveals that its upstream process parameter chain includes the blade machining accuracy and heat treatment parameters in the manufacturing stage, as well as the blade 3D model parameters in the design stage. Through analysis, it is found that the tool wear of the processing equipment (such as a five-axis CNC machining center) in the manufacturing stage of the blade causes the actual blade size to deviate from the design size (the blade exit angle deviation reaches 0.5°, exceeding the allowable tolerance of ±0.2°), which is defined as the root of the deviation.

[0055] A deep traceability report is generated based on the upstream process parameter chain of the abnormal node. The report indicates that the deviation was caused by the failure to replace the worn cutting tools of the machining equipment in a timely manner. The impact includes a decrease in engine vibration performance (vibration amplitude increased by 30%), which may lead to premature fatigue cracks in the blades (expected lifespan shortened by 20%). The corrective suggestions include replacing the cutting tools of the machining equipment, optimizing the tool wear monitoring and early warning mechanism, and adjusting the blade machining program to compensate for dimensional deviations. This deep traceability and report generation mechanism can provide strong support for the continuous improvement of equipment and effectively improve the reliability and performance stability of the equipment.

[0056] Furthermore, to identify performance differences across data association links, the method in this application includes:

[0057] Based on the root cause of the deviation in the deep traceability report, the corresponding stage process parameters are extracted to construct a cross-stage data association link; the deviation transmission coefficient of multiple stage process parameters in the cross-stage data association link is evaluated through the digital thread analysis engine; and the performance difference association link is identified based on the deviation transmission coefficient of multiple stage process parameters in the cross-stage data association link.

[0058] Specifically, cross-stage data correlation links refer to connecting interrelated process parameters in different lifecycle stages of equipment (such as design, manufacturing, and service) through data links to form a complete data path. This is used to analyze the transmission relationship and impact of process parameters between stages. For example, the 3D model parameters in the design stage will affect the machining accuracy in the manufacturing stage, and the assembly tolerances in the manufacturing stage will affect the equipment performance in the service stage. Deviation transmission coefficients are used to quantify the degree of impact of deviations in upstream process parameters on downstream process parameters and the final equipment performance during cross-stage transmission. For example, parameter deviations in the design stage are amplified or reduced through the manufacturing process, thereby affecting the equipment performance indicators in the service stage. Performance difference correlation links refer to the correlation path between equipment performance differences and cross-stage process parameter deviations. By analyzing this link, the specific causes and scope of impact of performance differences can be clarified.

[0059] Execution steps: Based on the root cause of deviation in the deep traceability report, extract the corresponding process parameters and construct a cross-stage data association link. For example, in the case of aero-engines, the root cause of deviation is the blade exit angle deviation caused by tool wear in the blade machining equipment during the manufacturing stage. Extract the blade 3D model angle parameters (design angle is 30°) from the design stage, the actual machining angle parameters (actual angle is 30.5°, deviation is 0.5°) from the manufacturing stage, and the blade vibration frequency parameters from the service stage. Construct a cross-stage data association link, and use the deviation transmission coefficient evaluation model in the digital thread analysis engine to determine that the deviation transmission coefficient from the design to the manufacturing stage is 0.8 (meaning that 80% of the design angle deviation will be transmitted to the manufacturing stage). The deviation transmission coefficient to the service stage is 1.2 (meaning that the angular deviation in the manufacturing stage will amplify the vibration frequency deviation in the service stage by a factor of 1.2). Based on these deviation transmission coefficients, the correlation link of performance differences is identified as follows: blade exit angle deviation → blade vibration frequency deviation → decrease in overall engine vibration performance. Further analysis shows that a blade exit angle deviation of 0.5° leads to a vibration frequency deviation of 5Hz in the service stage, which exceeds the allowable range (±3Hz) and affects the reliability of the engine. Through this cross-stage analysis, the root causes and transmission paths of performance differences can be accurately identified, providing a basis for formulating targeted optimization measures, such as adjusting the processing parameters in the manufacturing stage or optimizing the model in the design stage, thereby improving the overall performance and reliability of the equipment.

[0060] Furthermore, to evaluate the deviation propagation coefficient of process parameters across multiple stages in the cross-stage data association link, the method of this application further includes:

[0061] Obtain the cumulative impact index of the deviation of the process parameters at each stage; construct the deviation impact propagation matrix, wherein the matrix elements of the deviation impact propagation matrix represent the weight coefficients of the mutual influence of deviations between process parameters at different stages; and determine the deviation transmission coefficient by matrix multiplication based on the cumulative impact index and the deviation impact propagation matrix.

[0062] Specifically, the cumulative impact index of deviations is an indicator used to quantify the cumulative impact of deviations in process parameters at each stage on the final performance of equipment. This index comprehensively considers the magnitude of the deviation of a single process parameter and its transmission effect at different stages. For example, if the machining accuracy of a certain component deviates during the manufacturing stage, its cumulative impact index is 0.7, which means that this deviation may have a 70% impact on the equipment performance in subsequent stages. The matrix elements in the deviation impact propagation matrix represent the weighting coefficients of the mutual influence between deviations of process parameters at different stages, used to describe the transmission law of deviations between different stages. For example, between the design and manufacturing stages, the weighting coefficient of a certain process parameter is 0.6, indicating that the deviation in the design stage has a 60% probability of affecting the process parameters in the manufacturing stage. Matrix multiplication is a basic mathematical operation used to multiply the cumulative impact index of deviations by the deviation impact propagation matrix to determine the deviation transmission coefficient, thereby quantifying the degree of transmission of deviations in the cross-stage data association link.

[0063] Execution steps: Obtain the cumulative impact index of deviations in process parameters at each stage. For example, in the design stage, the cumulative impact index of deviations caused by errors in 3D model parameters is 0.5; in the manufacturing stage, the cumulative impact index of deviations caused by decreased precision of processing equipment is 0.7; and in the service stage, the cumulative impact index of maintenance parameter deviations is 0.6. Then, construct the deviation impact propagation matrix. If the deviation impact propagation matrix is ​​a 3×3 matrix, where the matrix elements represent the mutual influence weight coefficients of deviations between different stages, for example, the influence weight of the design stage on the manufacturing stage is 0.6, the influence weight of the manufacturing stage on the service stage is 0.8, and the direct influence weight of the design stage on the service stage is 0.3.

[0064] By multiplying the cumulative impact index vector of deviations with the deviation impact propagation matrix using matrix multiplication, a deviation transmission coefficient vector is obtained. For example, the deviation transmission coefficient from the design stage to the manufacturing stage is 0.5 × 0.6 = 0.3, the deviation transmission coefficient from the manufacturing stage to the service stage is 0.7 × 0.8 = 0.56, and the deviation transmission coefficient from the design stage to the service stage is 0.5 × 0.3 = 0.15. Through these deviation transmission coefficients, key influence paths in the performance difference correlation chain can be identified. Furthermore, the vibration frequency deviation of aero-engine blades mainly originates from the processing deviation in the manufacturing stage, which in turn mainly originates from the model parameter deviation in the design stage. It is necessary to focus on the deviation transmission path from design to manufacturing and then to the service stage. By optimizing design parameters and manufacturing processes to reduce performance differences in the service stage, more targeted measures can be taken to improve the reliability and performance of equipment.

[0065] Furthermore, by combining the aforementioned performance difference points with management strategies, and conducting collaborative optimization throughout the entire lifecycle of aviation equipment, the method of this application includes:

[0066] The performance difference points include potential difference points and existing difference points; cross-iteration is performed using the preventive cross-stage collaborative parameter combination corresponding to the potential difference points and the corrective cross-stage collaborative parameter combination corresponding to the existing difference points, and the management strategy is configured by binding it with the first digital thread master data and the second digital thread master data.

[0067] Specifically, potential differences and existing differences are two manifestations of equipment performance differences. Potential differences refer to performance differences that are not yet apparent but have the potential to develop into significant problems. They are usually caused by minor deviations in early process parameters or potential process risks, such as potential mismatches in material selection during the design phase or slight deviations in processing parameters during the manufacturing phase. Existing differences, on the other hand, refer to equipment performance differences that can already be clearly observed. These differences have usually already had a significant impact on the operational performance of the equipment, such as excessive wear of equipment components during the service phase or performance degradation detected during the maintenance phase.

[0068] Preventative cross-stage collaborative parameter combinations are a series of parameter adjustment schemes developed in advance for potential discrepancies, involving process parameters at multiple stages. They aim to prevent potential discrepancies from developing into serious problems through early optimization. Corrective cross-stage collaborative parameter combinations, on the other hand, are corrective measures developed for existing discrepancies. They correct existing performance differences by adjusting process parameters at relevant stages. These parameter combinations need to be linked to the first digital thread master data (focusing on core data in the design and manufacturing stages) and the second digital thread master data (focusing on core data in the service and maintenance stages) to ensure that the management strategy can comprehensively cover all key aspects of the equipment's entire life cycle.

[0069] Execution steps: Classify and identify performance differences, distinguishing between potential and existing differences. For example, in monitoring aircraft engine blades, analysis of the first and second digital thread master data and associated test sequences reveals slight deviations in blade dimensional tolerances during manufacturing (potential differences), while vibration modal testing during service reveals blade vibration frequencies exceeding the normal range (existing differences). For potential differences, develop preventative cross-stage collaborative parameter combinations, such as adjusting blade 3D model parameters during design and optimizing machining equipment precision control parameters during manufacturing. Bind relevant parameters to the first digital thread master data to ensure these optimization measures are implemented throughout the subsequent manufacturing and service processes.

[0070] To address existing discrepancies, corrective cross-stage collaborative parameter combinations are developed. For example, during the maintenance phase, the frequency of blade inspections and repair process parameters are increased; during the service phase, engine operating parameters are adjusted to reduce blade vibration stress. These corrective measures are then linked to the second digital thread master data for real-time tracking and evaluation of their effectiveness. Through this cross-iterative approach, combined with comprehensive information from the first and second digital thread master data, management strategies are dynamically adjusted and optimized. For instance, it was found during iteration that improving machining accuracy by 0.01 mm (achieved by adjusting machining equipment parameters) and optimizing material properties during the design phase (increasing material strength by 5%) effectively reduces blade vibration frequency deviation. Simultaneously, adopting new repair processes during the maintenance phase extends blade fatigue life. Continuous optimization in this manner effectively improves the reliability and safety of aerospace equipment, ensuring stable performance throughout its entire lifecycle.

[0071] In summary, the beneficial effects of the embodiments of this application are:

[0072] This application provides a digital thread-based method and system for managing the entire lifecycle of aerospace equipment. It acquires process parameters for each stage of the equipment's lifecycle, including machining accuracy data, assembly tolerances, and process execution records. Based on these parameters, and combining material properties and performance parameters, it determines the first digital thread master data and the first-layer associated test sequence. It also determines the second digital thread master data and the second-layer associated test sequence based on fault modes and maintenance parameters. Using a digital thread analysis engine based on the equipment knowledge graph, it maps the deviation between the physical equipment state and the virtual simulation state, identifying performance differences across data links. By using the first and second digital thread master data and their associated test sequences, and combining these performance differences with management strategies, it achieves collaborative optimization throughout the entire lifecycle of the aerospace equipment. It has achieved the acquisition of process parameters at each stage of the entire life cycle of aviation equipment, constructed master data of the first and second digital threads and related test sequences, structured association and connection of data at each stage, and accurately mapped the deviation between physical equipment and virtual simulation state using the digital thread analysis engine of equipment knowledge graph. It has also identified cross-stage performance difference points in real time, combined with dual-thread data and performance difference points to set management strategies and carry out full life cycle collaborative optimization, effectively improving the technical effect of equipment reliability.

[0073] Example 2 is based on the same inventive concept as the digital thread-based aviation equipment lifecycle management method in the previous examples, such as... Figure 2 As shown in the embodiment of this application, a digital thread-based full lifecycle management system for aviation equipment is provided, wherein the system includes:

[0074] M100 Process Parameter Acquisition Module: Acquires process parameters for each stage of the entire lifecycle of aerospace equipment, including machining accuracy data, assembly tolerances, and process execution records.

[0075] Test sequence determination module M200: Based on the process parameters of each stage, combined with material properties and performance parameters, determines the first digital thread master data and the first layer of associated test sequence, and combined with fault modes and maintenance parameters, determines the second digital thread master data and the second layer of associated test sequence.

[0076] M300 Performance Difference Identification Module: Based on the digital thread analysis engine of the equipment knowledge graph, it maps the deviation between the physical equipment state and the virtual simulation state, and identifies performance difference points under the cross-stage data association link.

[0077] Collaborative optimization module M400: Through the first digital thread master data and the first layer of associated test sequences, the second digital thread master data and the second layer of associated test sequences, and in combination with the performance difference point setting management strategy, collaborative optimization is carried out throughout the entire life cycle of aviation equipment.

[0078] Furthermore, the test sequence determination module M200 is used to perform the following method:

[0079] The first digital thread master data includes a 3D model identifier, material batch number, heat treatment parameters, and processing equipment ID; the test cases in the design stage perform strength simulation tests, and the test cases in the manufacturing stage perform dimensional tolerance tests. The strength simulation test data and the dimensional tolerance test data are both associated with the first digital thread master data through timestamps.

[0080] Furthermore, the test sequence determination module M200 is also used to perform the following method:

[0081] The second digital thread master data includes fault codes, spare parts replacement records, maintenance work order numbers, and testing equipment calibration certificate numbers. During the service phase, test cases perform vibration modal testing, and during the maintenance phase, test cases perform leak detection testing. The second-level associated test sequence uses a two-dimensional record of Pass / Fail + quantitative value. Non-conforming items automatically trigger the second digital thread master data to mark them.

[0082] Furthermore, the performance difference point identification module M300 is used to perform the following method:

[0083] The digital thread analysis engine employs a graph neural network, inputting the physical equipment state vector and the virtual simulation vector to determine the cosine similarity; based on the cosine similarity, it locates the source of deviation through the node attention mechanism of the graph neural network.

[0084] Furthermore, the performance difference point identification module M300 is also used to perform the following method:

[0085] Using the cosine similarity and a preset similarity threshold range for initial screening, attention weights of nodes exceeding the similarity threshold are enhanced and abnormal nodes are marked. Starting from the abnormal nodes, the influence of the deviation between the physical equipment state and the virtual simulation state on downstream performance parameters is evaluated through the edge weights of the equipment knowledge graph. When the influence exceeds the upper limit of the allowable influence, the deviation source deep tracing instruction is activated.

[0086] Furthermore, the performance difference point identification module M300 is used to perform the following method:

[0087] According to the deviation source deep tracing instruction, a graph neural network is used to traverse the upstream process parameter chain of the abnormal node in reverse; based on the upstream process parameter chain of the abnormal node, the root link of the deviation is located, and a deep tracing report containing the cause of the deviation, the scope of impact, and correction suggestions is generated.

[0088] Furthermore, the performance difference point identification module M300 is also used to perform the following method:

[0089] Based on the root cause of the deviation in the deep traceability report, the corresponding stage process parameters are extracted to construct a cross-stage data association link; the deviation transmission coefficient of multiple stage process parameters in the cross-stage data association link is evaluated through the digital thread analysis engine; and the performance difference association link is identified based on the deviation transmission coefficient of multiple stage process parameters in the cross-stage data association link.

[0090] Furthermore, the performance difference point identification module M300 is also used to perform the following method:

[0091] Obtain the cumulative impact index of the deviation of the process parameters at each stage; construct the deviation impact propagation matrix, wherein the matrix elements of the deviation impact propagation matrix represent the weight coefficients of the mutual influence of deviations between process parameters at different stages; and determine the deviation transmission coefficient by matrix multiplication based on the cumulative impact index and the deviation impact propagation matrix.

[0092] Furthermore, the collaborative optimization module M400 is used to perform the following method:

[0093] The performance difference points include potential difference points and existing difference points; cross-iteration is performed using the preventive cross-stage collaborative parameter combination corresponding to the potential difference points and the corrective cross-stage collaborative parameter combination corresponding to the existing difference points, and the management strategy is configured by binding it with the first digital thread master data and the second digital thread master data.

[0094] In summary, any step can be stored as a computer instruction or program in an unrestricted computer memory and can be called and recognized by an unrestricted computer processor; no further restrictions are imposed here.

[0095] Furthermore, the above technical solutions only embody the preferred technical solutions of the embodiments of this application. Any changes that those skilled in the art may make to certain parts of these solutions embody the novel principles of the embodiments of this application. Obviously, those skilled in the art can make various modifications and variations to this application without departing from the scope of this application.

Claims

1. A method for full lifecycle management of aviation equipment based on digital threads, characterized in that, The method includes: Obtain process parameters for each stage of the entire lifecycle of aerospace equipment, including machining accuracy data, assembly tolerances, and process execution records; Based on the process parameters of each stage, the first digital thread master data and the first layer of associated test sequence are determined in combination with material properties and performance parameters, and the second digital thread master data and the second layer of associated test sequence are determined in combination with fault modes and maintenance parameters. Based on the digital thread analysis engine of the equipment knowledge graph, the deviation between the physical equipment state and the virtual simulation state is mapped, and the performance difference points under the cross-stage data association link are identified. By using the first digital thread master data and the first layer of associated test sequences, the second digital thread master data and the second layer of associated test sequences, and combining the performance difference points to set management strategies, collaborative optimization is carried out throughout the entire life cycle of aviation equipment. Among them, the digital thread analysis engine based on the equipment knowledge graph includes: The digital thread analysis engine uses a graph neural network to determine the cosine similarity by inputting the physical equipment state vector and the virtual simulation vector. Based on the cosine similarity, the source of deviation is located through the node attention mechanism of the graph neural network; Among them, the node attention mechanism of graph neural networks is used to locate the source of deviation, including: Using the cosine similarity and a preset similarity threshold range, a preliminary screening is performed. The attention weights of nodes that exceed the similarity threshold are then enhanced and abnormal nodes are marked. Starting from the abnormal node, the impact of the deviation between the physical equipment state and the virtual simulation state on downstream performance parameters is evaluated through the edge weights of the equipment knowledge graph. When the impact exceeds the upper limit of the allowable impact, the deviation source depth tracing command is activated; The instructions for activating the depth tracing of deviation sources include: According to the deviation source depth tracing instruction, the upstream process parameter chain of the abnormal node is traversed in reverse using a graph neural network; Based on the upstream process parameter chain of the abnormal node, the root cause of the deviation is located, and a deep traceability report containing the cause of the deviation, the scope of impact, and correction suggestions is generated. Identifying performance differences across data association links in different stages includes: Based on the root cause of the deviation in the deep traceability report, the corresponding stage process parameters are extracted, and a cross-stage data association link is constructed. The deviation propagation coefficient of process parameters in multiple stages of the cross-stage data association link is evaluated using the digital thread analysis engine. Based on the deviation transmission coefficient of process parameters in multiple stages in the cross-stage data association link, identify the performance difference association link; The evaluation of the deviation propagation coefficient of process parameters in multiple stages of the cross-stage data association link further includes: Obtain the cumulative impact index of deviations in process parameters at each stage; Construct a deviation influence propagation matrix, wherein the matrix elements of the deviation influence propagation matrix represent the weighting coefficients of the mutual influence between deviations of process parameters at different stages; The deviation transmission coefficient is determined by matrix multiplication based on the deviation cumulative impact index and the deviation impact propagation matrix.

2. The method for full lifecycle management of aviation equipment based on digital threads as described in claim 1, characterized in that, The method for determining the first digital thread master data and the first layer of associated test sequences by combining material properties and performance parameters includes: The first digital thread master data includes a 3D model identifier, material batch number, heat treatment parameters, and processing equipment ID; During the design phase, test cases are used for strength simulation testing, and during the manufacturing phase, test cases are used for dimensional tolerance testing. The strength simulation test data and dimensional tolerance test data are both associated with the first digital thread master data through timestamps.

3. The method for full lifecycle management of aviation equipment based on digital threads as described in claim 2, characterized in that, The method for determining the second digital thread master data and the second-layer associated test sequence by combining fault modes and maintenance parameters includes: The second digital thread master data includes fault codes, spare parts replacement records, maintenance work order numbers, and testing equipment calibration certificate numbers; During the service phase, test cases are used for vibration modal testing, and during the maintenance phase, test cases are used for leak detection. The second-level associated test sequence uses a dual-dimensional record of Pass / Fail and quantitative values. Non-conforming items automatically trigger the second digital thread master data marker.

4. The method for full lifecycle management of aviation equipment based on digital threads as described in claim 1, characterized in that, In conjunction with the aforementioned performance differences, a management strategy is established to perform collaborative optimization throughout the entire lifecycle of aviation equipment. The method includes: The performance differences include potential differences and existing differences. The management strategy is configured by cross-iterping the preventive cross-stage collaborative parameter combination corresponding to the potential difference point and the corrective cross-stage collaborative parameter combination corresponding to the existing difference point, and binding them with the first digital thread master data and the second digital thread master data.

5. A digital thread-based full lifecycle management system for aviation equipment, characterized in that, The system is used to implement the digital thread-based full lifecycle management method for aviation equipment according to any one of claims 1-4, wherein the system comprises: Process parameter acquisition module: Acquires process parameters for each stage of the entire lifecycle of aerospace equipment, including machining accuracy data, assembly tolerances, and process execution records; Test sequence determination module: Based on the process parameters of each stage, combined with material properties and performance parameters, determine the first digital thread master data and the first layer of associated test sequence, and combined with fault modes and maintenance parameters, determine the second digital thread master data and the second layer of associated test sequence; Performance Difference Identification Module: Based on the digital thread analysis engine of the equipment knowledge graph, it maps the deviation between the physical equipment state and the virtual simulation state, and identifies performance difference points under the cross-stage data association link; Collaborative optimization module: By combining the first digital thread master data and the first layer of associated test sequences, the second digital thread master data and the second layer of associated test sequences, and the management strategy set for the performance difference points, collaborative optimization is carried out throughout the entire life cycle of aviation equipment.

Images