QEC detection method and system based on AR

Abstract

The invention relates to a QEC detection method and system based on AR. The method comprises the steps that a pre-constructed engine standard model is disassembled into part units, a three-dimensional model set in the installation process of all the part units is obtained, and the three-dimensional model obtained after installation of each part unit is finished is a part unit standard model; obtaining an engine real-time model based on AR equipment; when the difference degree between the engine real-time model and the standard model is higher than a first threshold value, a first part unit with the largest difference degree is obtained, whether the corresponding real-time model is matched with the three-dimensional model set or not is judged, if yes, the first part unit is maintained towards the corresponding standard model state, and if not, the first part unit is maintained; and when all the part units in contact with the first part unit are synchronously changed to the corresponding standard model states, the engine real-time model is obtained again to continue execution, and otherwise, a QEC request is sent out. Compared with the prior art, the method has the advantages that the safety and the efficiency of the whole QEC operation inspection and verification process are improved, and the like.

Description

Technical Field

[0001] This invention relates to the fields of infrastructure and IT support technology, and in particular to an AR-based QEC detection method and system. Background Technology

[0002] Aviation maintenance refers to the upkeep and repair of aircraft and their technical equipment to ensure aircraft safety. Aviation maintenance is a prerequisite and necessary condition for aircraft use and an important component of the aviation industry. The aircraft engine is the heart of the aircraft; its maintenance is both a part of overall aircraft maintenance and a crucial component of aviation mechanics. The most vulnerable areas for aircraft engine failure are the high-pressure compressor, combustion chamber, and high-pressure turbine. Currently, QEC (Quick Engine Change) is of paramount importance in aircraft engine maintenance. QEC is a frontline aviation operation performed on the core aircraft component, the engine, characterized by numerous component disassemblies and reassemblies, high frequency of disassembly and reassembly, and significant safety risks.

[0003] Currently, QEC (Quality Control and Inspection) work relies primarily on manual labor. Maintenance personnel still need to manually consult auxiliary equipment and fill out forms during the process. Furthermore, because QEC work is extremely complex, requiring a large number of personnel and numerous operational steps, the correlation between the actions of different personnel remains insufficient. The auxiliary equipment merely collects aircraft maintenance data for feedback; there is neither real-time guidance for QEC maintenance nor multi-step correlation analysis of the QEC process. It relies solely on the experience of maintenance personnel for real-time operation, making it difficult to guarantee the safety of the entire inspection and verification process. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology by providing an AR-based QEC inspection method and system. This method applies Augmented Reality (AR) technology to aircraft maintenance work, realizes a high degree of automation in QEC inspection, frees workers from tedious manual operations, and effectively improves the safety of the entire QEC inspection and verification process.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] According to a first aspect of the present invention, an AR-based QEC detection method is provided, comprising the following steps: S1, disassembling a pre-built standard engine model into component units, and obtaining a three-dimensional model group of all component units during the installation process, wherein the three-dimensional model at the end of installation of each component unit is the corresponding component unit standard model; S2, obtaining a real-time engine model based on a pre-acquired AR device; S3, determining whether the difference between the real-time engine model and the standard engine model is higher than a first preset threshold; if so, obtaining the first component unit with the largest difference and executing S4, otherwise not executing; S4, determining whether the real-time model corresponding to the first component unit matches the three-dimensional model group corresponding to the first component unit; if so, executing S5, otherwise issuing a QEC request; S5, repairing the first component unit to the corresponding component unit standard model state, and determining whether all component units in contact with the first component unit have synchronously changed to the corresponding component unit standard model state; if so, returning to S2 to continue execution, otherwise issuing a QEC request.

[0007] As a preferred technical solution, the specific construction process of the engine standard model includes: acquiring engine part information and constructing a separate model for each part; assembling a preliminary engine model based on the separate model of each part, pre-acquired part assembly information, and assembly joint model; acquiring engine flight data and using the flight data to correct the preliminary engine model to obtain a corrected engine model; acquiring historical engine disassembly and assembly information and calibrating it with the corrected engine model to obtain the engine standard model.

[0008] As a preferred technical solution, the process of acquiring a group of three-dimensional models of a single part unit specifically includes: monitoring the installation process using pre-acquired AR equipment; acquiring all installation nodes of the current part unit from the start of installation to the end of installation; and acquiring all three-dimensional models of the position of the current part unit relative to the entire engine based on all installation nodes, thereby obtaining the corresponding group of three-dimensional models.

[0009] As a preferred technical solution, S2 specifically includes: acquiring an AR device and determining a target object; based on the AR device and the ranging principle, acquiring the distance between the AR device and the target object at a predetermined angle, and constructing three-dimensional coordinates; using at least three sets of three-dimensional coordinates at predetermined angles to obtain line, surface, and volume distribution data, and establishing a real-time engine model.

[0010] As a preferred technical solution, determining whether the difference between the real-time engine model and the standard engine model exceeds a first preset threshold involves the following steps: First, the structural surfaces in the real-time engine model and the standard engine model are set to obtain the data of the smallest unit. Second, based on the data of the smallest unit, the real-time engine model and the standard engine model are overlapped to obtain and compare the positions of all part units in the overlapped 3D model. Third, when the second part unit and the third part unit are located at the same position, the second part unit and the third part unit are compared to obtain the difference value. Fourth, after masking the second part unit and the third part unit, the real-time engine model and the standard engine model are compared again, until all part units are traversed. Fifth, all obtained difference values ​​are integrated, and the final difference value is calculated based on a preset weighting coefficient.

[0011] As a preferred technical solution, when the structural surface is curved, the data of each point in the structural surface is compared sequentially in the X, Y, and Z directions, and the data of the smallest unit is obtained by point set.

[0012] As a preferred technical solution, S3 further includes: when the difference between the real-time engine model and the standard engine model is higher than a first preset threshold, locating the difference region between the real-time engine model and the standard engine model; and displaying the exploded view of the standard engine model and the exploded view of the real-time engine model corresponding to the difference region.

[0013] As a preferred technical solution, determining whether the real-time model corresponding to the first part unit matches the three-dimensional model group corresponding to the first part unit includes: obtaining the real-time model corresponding to the first part unit; comparing the real-time model with each three-dimensional model in the three-dimensional model group corresponding to the first part unit; and determining whether the matching degree between the real-time model and any three-dimensional model in the three-dimensional model group is higher than a second preset threshold.

[0014] As a preferred technical solution, the execution process of the QEC request specifically includes: obtaining the correspondence between the part numbers of each component according to the engine standard model; obtaining the location of the deviated part unit and generating a maintenance task; generating the corresponding exploded view, flowchart and guidance diagram according to the maintenance task, and obtaining multiple branch steps; using the multiple branch steps as prerequisite steps according to the correspondence between the part numbers of each component, establishing corresponding node steps; and establishing a maintenance task tree according to the node steps and the corresponding branch steps.

[0015] According to a second aspect of the present invention, an AR-based QEC (Quality, Emergency, and Computation) inspection system is provided. The system is used to implement the method described above. The system includes a standard establishment module, a real-time acquisition module, a first judgment module, a second judgment module, an inspection module, and a QEC request module. The standard establishment module is used to disassemble a pre-built engine standard model into component units and acquire a group of three-dimensional models of all component units during the assembly process, wherein the three-dimensional model at the end of the assembly of each component unit is the corresponding component unit standard model. The real-time acquisition module is used to acquire a real-time engine model based on a pre-acquired AR device. The first judgment module is used to determine the degree of difference between the real-time engine model and the engine standard model. If the difference exceeds a first preset threshold, the first part unit with the greatest difference is obtained, and the second judgment module is started; otherwise, no action is taken. The second judgment module is used to determine whether the real-time model corresponding to the first part unit matches the three-dimensional model group corresponding to the first part unit. If so, the verification module is started; otherwise, the QEC request module is started to issue a QEC request. The verification module is used to repair the first part unit to the corresponding part unit standard model state and determine whether all part units in contact with the first part unit have synchronously changed to the corresponding part unit standard model state. If so, the real-time acquisition module is returned to continue execution; otherwise, the QEC request module is started to issue a QEC request.

[0016] Compared with the prior art, the present invention has the following beneficial effects:

[0017] 1. The method provided by this invention pre-establishes a standard engine model during QEC testing, and uses AR equipment to intelligently generate a real-time engine model, which is then compared with the standard engine model to guide the staff in maintenance operations in real time. After the operation steps are completed, matching and comprehensive testing are performed. This method can assist maintenance personnel in inspecting or verifying the actions of each step of QEC, achieving a high degree of automation in QEC testing and freeing staff from tedious manual operations. Furthermore, after the standard model state corresponding to the first part unit is repaired, a QEC request is issued when all part units in contact with it have not changed synchronously, which can ensure the correlation between maintenance steps and effectively improve the safety and efficiency of the entire QEC operation inspection and verification process.

[0018] 2. This invention establishes a maintenance task tree when executing QEC requests. Specifically, it automatically generates corresponding exploded diagrams, flowcharts, and guidance diagrams based on the confirmed maintenance tasks, refining each step of the operation. It establishes multi-level branch steps by combining the correspondence between the part numbers of each component in the pre-built engine standard model. It also uses AR equipment to acquire real-time information and performs comprehensive testing based on the constructed maintenance task tree. This is not just about obtaining aircraft maintenance data for feedback, but about integrating multiple information such as aircraft maintenance data, status data, and flight parameter data, and displaying them in multiple dimensions such as curves, reports, two-dimensional attitude, and three-dimensional playback. It integrates multi-dimensional maintenance support information, which can provide data support for aircraft maintenance support work and further ensure the safety of the entire operation inspection and verification process.

[0019] 3. When comparing the differences between the real-time engine model and the standard engine model, this invention compares the part unit models at the same location, finds the difference value, and after traversing the part units, obtains the degree of difference of the entire engine model and compares it with the first preset threshold. This fully considers the integrity of engine maintenance and the correlation between various part units, which can further improve the accuracy of QEC work and the safety of the entire operation inspection and verification process.

[0020] 4. By determining whether the matching degree between the real-time model corresponding to the first part unit and any three-dimensional model in the corresponding three-dimensional model group is higher than the second preset threshold, this invention can effectively determine whether the part unit with different properties is in a normal installation state, that is, the structure itself is not damaged and is in the installation process. This can avoid false detections that require QEC maintenance and improve work efficiency. Attached Figure Description

[0021] Figure 1 A flowchart illustrating the method of this invention;

[0022] Figure 2 A schematic diagram of the system structure is provided for this invention;

[0023] Among them: 100, Standard Establishment Module; 200, Real-time Acquisition Module; 300, First Judgment Module; 400, Second Judgment Module; 500, Verification Module; 600, QEC Request Module. Detailed Implementation

[0024] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0025] Example

[0026] like Figure 1As shown, this embodiment provides an AR-based QEC detection method, which specifically includes the following steps:

[0027] Step S1: Disassemble the pre-built standard engine model into component units and obtain the three-dimensional model group of all component units during the installation process. The three-dimensional model of each component unit at the end of installation is the corresponding component unit standard model.

[0028] The construction of a standard engine model requires acquiring relevant engine data. This data includes information on engine components and flight data. Component information includes the dimensions, shape, and materials of the air intake, fan, low-pressure compressor, high-pressure compressor, combustion chamber, high-pressure turbine, low-pressure turbine, and nozzle. Flight data includes the number of turbines, turbine arrangement, maximum output power, and maximum thrust. This data describes the basic structural parameters of the aircraft engine, thus enabling the creation of a more comprehensive engine model.

[0029] Based on relevant engine data, a standard engine model is constructed. The specific process is as follows:

[0030] 1) Establish structured engine basic information management: that is, obtain the part information such as the size, shape and material of the engine's air intake, fan, low-pressure compressor, high-pressure compressor, combustion chamber, high-pressure turbine, low-pressure turbine and nozzle, and use NATRAN software to build a separate model of each part;

[0031] 2) Connect to the management system of an airline to obtain engine-related data: that is, obtain the assembly information of each part and the model of each part assembly joint from the management system of an airline. Assemble each part and joint model according to the assembly rules to form a preliminary engine model.

[0032] 3) Connect to the operating system and obtain engine flight data: that is, obtain flight parameters such as the number of turbines, turbine arrangement, maximum output power, and maximum thrust, and improve the initial engine model to obtain the corrected engine model;

[0033] 4) Engine historical disassembly and assembly records can be queried: that is, the engine disassembly and assembly data of each time can be obtained and finally checked against the engine correction model to obtain the engine standard model.

[0034] Only after establishing a standard engine model can workers quickly obtain specific data about the engine used in QEC (Quality, Engineering, and Manufacturing) and quickly locate and display the assembly process through exploded diagrams.

[0035] Engine standard model: This involves creating a 3D engine model, models of some sub-components, exploded views, and disassembly / assembly animations, and establishing a correspondence between these and the corresponding part numbers. Specifically:

[0036] 1) Establish a 3D model of the engine;

[0037] 2) Create models, exploded views, and assembly / disassembly animations for some sub-components;

[0038] 3) Establish the correspondence between the model, sub-component models, exploded views, disassembly and assembly animations, and the part numbers of each component.

[0039] It should be noted that the standard engine model formed in step S1 is a three-dimensional model under the theoretical condition that the engine is not currently experiencing a fault.

[0040] In step S1, based on engine-related data, the standard engine model is disassembled into component units, each of which is a detachable and installable component unit. The process of acquiring the 3D model set for a single component unit specifically includes: during the installation of each component unit, monitoring the installation process using pre-acquired AR equipment, acquiring all installation nodes for that component unit from the start to the end of installation, and acquiring the 3D model of the component unit's position relative to the entire engine at each installation node. All 3D models of each component unit throughout the entire installation process constitute its corresponding 3D model set. By traversing all component units, the 3D model set for all component unit installation processes can be obtained. The more installation nodes are segmented, the higher the reliability of the 3D model set.

[0041] When performing step S1, the part unit has multiple states throughout the entire installation process. The three-dimensional model of each state is recorded. The three-dimensional model at the end of the installation is the part unit model corresponding to the engine standard model, which is the standard model (standard position) of the part unit.

[0042] Step S2: Obtain a real-time engine model based on the pre-acquired AR device.

[0043] A real-time engine model refers to the engine model after maintenance by operators, i.e., the actual model. The specific process for obtaining a real-time engine model is as follows:

[0044] First, acquire the AR device and determine the target object; second, based on the AR device and the ranging principle, obtain the distance between the AR device and the target object at a predetermined angle, and construct the three-dimensional coordinates of a large number of dense points; finally, using at least three sets of three-dimensional coordinates at predetermined angles, obtain the line, surface and volume distribution data, and establish a real-time engine model, that is, a real-time three-dimensional model of the engine structure.

[0045] Step S3: Determine whether the difference between the real-time engine model and the standard engine model is higher than the first preset threshold. If so, obtain the first part unit with the largest difference and execute step S4; otherwise, do not execute.

[0046] The process of determining whether the difference between the real-time engine model and the standard engine model exceeds a first preset threshold includes:

[0047] Step S31: Set the structural surfaces in the two 3D models, the real-time engine model and the standard engine model, to obtain the data of the smallest unit;

[0048] Step S32: Based on the data of the smallest unit, the real-time engine model is overlaid with the standard engine model to obtain and compare the positions of all part units in the overlaid 3D model.

[0049] Step S33: When the second part unit and the third part unit are two different part units located at the same position, compare the two part units and obtain the difference value;

[0050] Specifically, the process involves: setting the structural surfaces of two part units as point sets, obtaining the three-dimensional coordinates of each point, and obtaining the ratio of the number of points that differ between the two part units to the total number of points, which is the difference value.

[0051] For example: On the structural surface of a part unit (second part unit) in a 3D model, a point set is formed by marking a point every 1 mm. There are 100 points in the point set. The 3D coordinates of these 100 points are obtained. Another part unit (third part unit) in the same position in another 3D model also has 100 points and 3D coordinates. By comparing the two sets of 3D coordinates, if the 3D coordinates of 90 points are the same and the 3D coordinates of 10 points are different, then the difference between the two part units is 10%, or 0.1.

[0052] Step S34: After masking the second and third part units, compare the two 3D models again until all part units have been found.

[0053] Step S35: Integrate all the obtained difference values ​​and calculate the final difference degree value according to the preset weight coefficient.

[0054] It should be noted that when the structural surface is planar, the data of each point on the plane can usually be used directly, and will not be elaborated further; when the structural surface is curved, the data of each point on the structural surface is compared in the X, Y, and Z directions in turn, and the data of the smallest unit is obtained by point set conversion.

[0055] The process involves creating a group for the first part unit that has a difference, finding all adjacent part units in the difference model, performing a difference search, and adding the difference-causing part units to the current group and deleting them from the difference model to avoid repeated searches, until the last part unit is found. The difference model is a real-time model used for comparison with the standard model.

[0056] All parts units that have differences are normalized, the point set of the parts units that have differences is obtained, a difference degree value is generated according to the weight coefficient, and the difference degree value is compared with the first preset threshold.

[0057] In addition, the first part unit with the greatest degree of difference is obtained as follows: when the difference between the real-time engine model and the standard engine model is greater than a first preset threshold, all part units in the difference model are obtained, and the real-time model of all part units is compared with the standard model of the part unit in turn to obtain the difference value of the part unit, and the part unit with the greatest degree of difference is obtained.

[0058] When performing step S3, if there is a significant difference between the real-time engine model and the standard engine model, it means that some component units have deviated from their standard positions. The component unit with the largest deviation is the most likely cause of failure and needs to be inspected.

[0059] Step S3 further includes: when the difference between the real-time engine model and the standard engine model exceeds a first preset threshold, locating the differentiated region between the real-time engine model and the standard engine model, and simulating and displaying the exploded view of the standard engine model and the exploded view of the real-time engine model corresponding to the differentiated region. Display methods include: display via AR glasses and upload / display via a mobile network device.

[0060] Step S4: Determine whether the real-time model corresponding to the first part unit with the greatest difference matches the 3D model group corresponding to the first part unit. If yes, proceed to step S5; otherwise, issue a QEC request.

[0061] The specific judgment process for this step is as follows:

[0062] Step S41: Obtain the real-time model of the part unit corresponding to the first part unit;

[0063] Step S42: Compare the real-time model of the part unit with each 3D model in the 3D model group corresponding to the part unit;

[0064] Step S43: Determine whether the matching degree between the real-time model of the part unit and any three-dimensional model in the three-dimensional model group is higher than the preset second preset threshold. If so, proceed to step S5.

[0065] Specifically, if the difference between the real-time model of the part unit and any 3D model is less than a preset value, that is, the matching degree between the real-time model and this 3D model is higher than a second preset threshold, for example, the difference is less than 0.5%, that is, the matching degree is higher than 99.5%.

[0066] When executing step S4, it can be effectively determined whether the component unit with the difference is in a normal installation state (at this time, the matching degree between the real-time model of the component unit and one of the three-dimensional models in the three-dimensional model group is higher than the second preset threshold). That is, the structure itself is not damaged and is in the process of installation. This means that the component is only loose (accidental loosening or loosening caused by human maintenance forgetting to restore) and there is no stress structure damage. At this time, further inspection is required for this state. Conversely, if the component is not in a normal installation state, it can be directly identified that the component is abnormally deformed, detached, or in an abnormal position. It can be identified that the component is damaged and QEC maintenance is required. At this time, a QEC request is directly issued.

[0067] Step S5: Repair the first part unit to the corresponding part unit standard model state. Determine whether all part units in contact with the first part unit have changed synchronously to the corresponding part unit standard model state. If yes, return to step S2 to continue execution; otherwise, issue a QEC request.

[0068] The prerequisite for executing step S5 is that the component unit with the difference in nature is in the normal installation state. At this time, the maintenance personnel are notified to restore the component to the standard position according to the installation state steps. At the same time, the AR device (such as AR glasses) is used to obtain whether all other components in contact with the component are restored to the standard position synchronously during the restoration process. If so, it is determined that the component unit was simply forgotten to be restored, and the component can still perform normal functions. It has also successfully constrained all components in contact with it to perform normal functions. This situation indicates that the abnormal risk of the component unit has been initially eliminated. Conversely, if the component unit cannot be restored to the standard position, or cannot constrain all components in contact with it to return to the standard position, it means that although the component unit structure is not damaged, its function is damaged, and QEC maintenance is still required. At this time, a QEC request is issued directly.

[0069] To further ensure that all engine risks are eliminated, after the abnormal risks of this part have been initially eliminated, step S2 should be executed again. If the real-time model of the engine corresponds to the standard model at this time, the inspection is complete; otherwise, the inspection of other parts and components should be carried out again.

[0070] During the execution of steps S1 to S5, the model is acquired through AR devices (such as AR glasses) worn by maintenance personnel.

[0071] In steps S4 and S5, when the QEC request is executed, a maintenance task is generated and a maintenance task tree is established by obtaining the fault location, that is, the location of the deviated part unit.

[0072] Workers can effectively detect engine fault locations by checking the engine's real-time status and comparing it with a standard engine model; or they can detect engine fault locations by scanning real-time images of the engine with video and comparing them with a standard engine model.

[0073] After obtaining the engine fault location, this location is uploaded. Based on the fault source, a corresponding repair process, or repair task, is intelligently generated. Only then is the repair task for QEC (Quality, Engine, and Control) truly determined. Simultaneously, QEC package management needs to be established, creating detailed information on QEC package components for each engine model. This serves as the basis for enabling rapid disassembly and assembly of QEC components and managing missing parts records.

[0074] The aforementioned process of establishing a maintenance task tree includes:

[0075] 1) Once the maintenance task is confirmed, the system automatically generates an exploded view, flowchart, and instruction diagram for the maintenance task; each step is broken down into a branch step.

[0076] 2) Based on the "correspondence between part numbers of each component" obtained in step S1, the subsequent step that can only be executed after multiple preceding steps are completed is the node step, and the multiple preceding steps are the branch steps of this node step.

[0077] 3) By analogy, establish first-level node steps, multiple branch steps of first-level node steps (second-level node steps), multiple branch steps of second-level node steps (third-level node steps), and so on, and you will get the maintenance task tree.

[0078] The specific application process of the method to AR devices provided in this embodiment is as follows:

[0079] When performing step S3, task permissions are granted to the AR device.

[0080] AR devices include multi-functional glasses, specifically including a camera, display screen, processor, and human-computer interaction terminal; AR device terminals can see their own maintenance tasks and mark them on a standard engine model; AR device terminals can understand the assembly process of the operation through exploded views.

[0081] The task standard execution guidance materials are distributed to AR devices and displayed on the screen. The task standard execution guidance materials include the QEC work target location, tool usage, and detailed explanation of QEC operation steps. The operation materials include text, images, and videos. The application of AR devices can fully free up workers' hands and improve work efficiency.

[0082] During QEC operations: AR devices acquire real-time operational footage and generate a real-time engine model;

[0083] At this point, the worker performs real-time QEC (Quality, Engineering, and Control) work according to the exploded view of the engine and detailed maintenance procedures displayed on the AR device. The AR device captures the worker's real-time work footage and synchronously guides the worker through each subsequent step of the standard procedure; of course, the AR device also generates a real-time engine model based on the worker's real-time work footage captured by the AR device.

[0084] It should be noted that since each AR device can only capture a small portion of the operation screen, the real-time engine model generated by the AR device should be a partial sub-component model; however, it is still displayed on the complete model.

[0085] When the QEC operation reaches a node step in the maintenance task tree, it acquires the operation videos of all branch steps under that node step, performs comprehensive inspection, and stores them. In practice, the inspection in the QEC operation can still be performed using steps S2 to S5, specifically divided into two steps: inspection and storage.

[0086] According to the maintenance task tree, a node step can only be executed after multiple branch steps are completed. This means that a node step can only be executed normally if multiple branch steps are completed according to the procedure. Therefore, when a node step needs to be executed, the AR device executing the node step automatically obtains the operation status of the branch step, that is, it obtains the real-time engine operation model of other AR devices executing the branch step obtained in step S4 and the operation video recorded by these AR devices. The specific steps are as follows:

[0087] First, integrate the real-time engine operation models from multiple AR devices and determine whether the integrated model is safe and compliant. If it is, proceed directly to the node step; otherwise, proceed to the next step.

[0088] Based on the operation videos recorded by these AR devices, erroneous branch steps are identified, feedback is provided for improvement, and node steps can still be performed after correction.

[0089] After executing a node step, the data for that node step and its branch steps are stored.

[0090] When all QEC maintenance tasks are completed, all AR device forms are collected, and a task summary is performed. Therefore, this method eliminates the need for manual form filling and summarization during QEC testing; the AR devices directly assist in the process, making the summary more accurate and faster.

[0091] Furthermore, this embodiment also provides an AR-based QEC detection system, which is used to implement the various steps of the aforementioned method, including a standard establishment module 100, a real-time acquisition module 200, a first judgment module 300, a second judgment module 400, an inspection module 500, and a QEC request module 600. The standard establishment module 100 is used to disassemble the pre-built engine standard model into component units and obtain the 3D model group of all component units during the installation process. The 3D model of each component unit at the end of installation is the corresponding component unit standard model. The real-time acquisition module 200 is used to acquire the engine real-time model based on the pre-acquired AR device. The first judgment module 300 is used to judge whether the difference between the engine real-time model and the engine standard model is higher than a first preset threshold. If so, the first component unit with the largest difference is acquired and the second judgment module 400 is started. Otherwise, it is not executed. The second judgment module 400 is used to judge whether the real-time model corresponding to the first component unit matches the 3D model group corresponding to the first component unit. If so, the verification module 500 is started. Otherwise, the QEC request module 600 is started to issue a QEC request. The verification module 500 is used to repair the first component unit to the corresponding component unit standard model state and judge whether all component units in contact with the first component unit have synchronously changed to the corresponding component unit standard model state. If so, it returns to the real-time acquisition module 200 to continue execution. Otherwise, the QEC request module 600 is started to issue a QEC request. The specific execution steps for each module are the same as those for the aforementioned method, and will not be repeated here.

[0092] In summary, the AR-based QEC (Quality, Emergency, and Critical Engine) detection method and system provided by this invention establishes a standard engine model, uses AR devices to acquire real-time operational images, generates a real-time engine model, and matches the standard model with the real-time model. During the execution of each node step, it acquires models and videos of all its branch steps for comprehensive detection. Throughout the implementation process, tasks can be automatically assigned, and real-time guidance and recording can be provided on the AR devices, eliminating the need for manual document review and form filling. Furthermore, it intelligently generates engine models and maintenance task trees, providing real-time guidance to personnel during maintenance operations. After each operation step is completed, matching and comprehensive detection are performed. Utilizing AR technology assists maintenance personnel in inspecting or verifying QEC, effectively avoiding low-level errors caused by human factors and improving the accuracy of safety detection. In some other embodiments, this method can also enable communication between multiple AR devices, facilitating control and management of AR devices via a control center.

[0093] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. An AR-based QEC detection method, characterized in that, Includes the following steps: S1, disassemble the pre-built standard engine model into part units, and obtain the three-dimensional model group of all part units during the installation process. The three-dimensional model of each part unit at the end of installation is the corresponding part unit standard model. S2, based on pre-acquired AR devices, acquires a real-time model of the engine; S3, determine whether the difference between the real-time engine model and the standard engine model is higher than a first preset threshold. If so, obtain the first part unit with the largest difference and execute S4; otherwise, do not execute. S4. Determine whether the real-time model corresponding to the first part unit matches the three-dimensional model group corresponding to the first part unit. If yes, execute S5; otherwise, issue a QEC request. S5, repair the first part unit to the corresponding part unit standard model state, determine whether all part units in contact with the first part unit have changed synchronously to the corresponding part unit standard model state, if so, return to S2 to continue execution, otherwise issue a QEC request.

2. The AR-based QEC detection method according to claim 1, characterized in that, The specific construction process of the engine standard model includes: Obtain information about engine parts and build a separate model for each part; Based on the individual model of each part, the pre-acquired part assembly information, and the assembly joint model, a preliminary engine model is assembled. Acquire engine flight data and use the flight data to correct the initial engine model to obtain a corrected engine model; The engine's historical disassembly and assembly information is obtained and compared with the engine correction model to obtain the engine standard model.

3. The AR-based QEC detection method according to claim 1, characterized in that, The process of obtaining a group of 3D models for a single part unit specifically includes: Monitor the installation process using pre-acquired AR devices; Retrieve all installation nodes of the current part unit from the start of installation to the end of installation; Based on all the installation nodes, obtain all the three-dimensional models of the current part unit relative to the entire engine, and obtain the corresponding three-dimensional model group.

4. The AR-based QEC detection method according to claim 1, characterized in that, S2 specifically includes: Acquire AR devices and identify target objects; Based on the AR device and the ranging principle, the distance between the AR device and the target object at a predetermined angle is obtained, and three-dimensional coordinates are constructed. By using at least three sets of three-dimensional coordinates at predetermined angles, line, surface, and volume distribution data are obtained, and a real-time engine model is established.

5. The AR-based QEC detection method according to claim 1, characterized in that, The process of determining whether the difference between the real-time engine model and the standard engine model exceeds a first preset threshold includes: The structural surfaces in the real-time engine model and the standard engine model are set together to obtain the data of the smallest unit. Based on the data of the smallest unit, the real-time engine model is overlaid with the standard engine model to obtain and compare the positions of all part units in the overlaid 3D model. When the second part unit and the third part unit are located in the same position, compare the second part unit and the third part unit and obtain the difference value; After disabling the second and third part units, compare the real-time engine model with the standard engine model again, until all part units are traversed; All the obtained difference values ​​are integrated, and the final difference value is calculated based on the preset weighting coefficients.

6. The AR-based QEC detection method according to claim 5, characterized in that, When the structural surface is curved, the data of each point in the structural surface is compared sequentially in the X, Y, and Z directions, and the data of the smallest unit is obtained by point set conversion.

7. The AR-based QEC detection method according to claim 1, characterized in that, S3 further includes: when the difference between the real-time engine model and the standard engine model exceeds a first preset threshold... Locate the region of difference between the real-time engine model and the standard engine model; This displays exploded views of the standard engine model and the real-time engine model corresponding to the differentiated regions.

8. The AR-based QEC detection method according to claim 1, characterized in that, The process of determining whether the real-time model corresponding to the first part unit matches the 3D model group corresponding to the first part unit includes: Obtain the real-time model corresponding to the first part unit; The real-time model is compared with each three-dimensional model in the three-dimensional model group corresponding to the first part unit; Determine whether the matching degree between the real-time model and any three-dimensional model in the three-dimensional model group is higher than a second preset threshold.

9. The AR-based QEC detection method according to claim 1, characterized in that, The execution process of the QEC request specifically includes: Based on the engine standard model, obtain the correspondence between the part numbers of each component; Obtain the location of the deviated part unit and generate a maintenance task; Based on the maintenance task, generate corresponding exploded views, flowcharts, and guidance diagrams, and obtain multiple branch steps; Based on the correspondence between the part numbers of each component, the multiple branch steps are used as prerequisite steps to establish corresponding node steps; Based on the node steps and corresponding branch steps, a maintenance task tree is established.

10. An AR-based QEC detection system, characterized in that, The system is used to implement the method as described in any one of claims 1-9, and the system includes a standard establishment module, a real-time acquisition module, a first judgment module, a second judgment module, a verification module, and a QEC request module. The standard establishment module is used to disassemble the pre-built engine standard model into part units and obtain a group of three-dimensional models during the installation process of all part units. The three-dimensional model of each part unit at the end of installation is the corresponding part unit standard model. The real-time acquisition module is used to acquire a real-time model of the engine based on a pre-acquired AR device. The first judgment module is used to determine whether the difference between the real-time engine model and the standard engine model is higher than a first preset threshold. If so, the first part unit with the largest difference is obtained and the second judgment module is started; otherwise, it is not executed. The second judgment module is used to determine whether the real-time model corresponding to the first part unit matches the three-dimensional model group corresponding to the first part unit. If yes, the verification module is started; otherwise, the QEC request module is started to issue a QEC request. The inspection module is used to repair the first part unit to the corresponding part unit standard model state, and to determine whether all part units in contact with the first part unit have synchronously changed to the corresponding part unit standard model state. If so, it returns to the real-time acquisition module to continue execution; otherwise, it starts the QEC request module to issue a QEC request.

Images