CN118135143A - AR-based aeroengine maintenance modeling method - Google Patents
AR-based aeroengine maintenance modeling method Download PDFInfo
- Publication number
- CN118135143A CN118135143A CN202410555406.XA CN202410555406A CN118135143A CN 118135143 A CN118135143 A CN 118135143A CN 202410555406 A CN202410555406 A CN 202410555406A CN 118135143 A CN118135143 A CN 118135143A
- Authority
- CN
- China
- Prior art keywords
- aeroengine
- sample
- information
- maintained
- target sample
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012423 maintenance Methods 0.000 title claims abstract description 46
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 6
- 238000012545 processing Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 9
- 230000002159 abnormal effect Effects 0.000 description 4
- 230000003190 augmentative effect Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 239000000446 fuel Substances 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000002604 ultrasonography Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000000053 physical method Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Landscapes
- Testing Of Engines (AREA)
Abstract
The invention provides an AR-based aeroengine maintenance modeling method, which relates to the field of data processing and comprises the following steps: establishing an aero-engine model library for storing sample aero-engine three-dimensional models of various models; acquiring scanning information of an aeroengine to be maintained, and acquiring a target sample aeroengine three-dimensional model from an aeroengine model library; establishing a first knowledge graph and a second knowledge graph; acquiring historical operation parameter information of an aeroengine to be maintained, and determining candidate fault types by combining a first knowledge graph; acquiring running state information of the aeroengine to be maintained, and predicting fault information by combining the second knowledge graph and candidate fault types corresponding to the aeroengine to be maintained; based on the predicted fault information and the target sample aero-engine three-dimensional model, a waiting AR model is established, and the method has the advantages of assisting in aero-engine maintenance based on the AR technology, improving maintenance efficiency, reducing errors and optimizing maintenance flow.
Description
Technical Field
The invention relates to the field of data processing, in particular to an AR-based aeroengine maintenance modeling method.
Background
The aeroengine has complex structure and changeable state, the working environment of the air path component is very severe, and the aeroengine can bear higher centrifugal load, pneumatic load, high temperature and atmospheric temperature difference load, vibration alternating load and the like, and can inevitably generate faults. The engine state is evaluated by collecting parameters such as temperature, pressure, rotating speed, flow and the like of each section of the engine, and the faults of the engine are diagnosed and predicted, so that the efficiency of maintenance work of the aeroengine can be effectively improved.
The augmented reality (Augmented Reality, AR) technology is a technology for skillfully fusing virtual information with a real world, and widely uses various technical means such as multimedia, three-dimensional modeling, real-time tracking and registration, intelligent interaction, sensing and the like, and applies virtual information such as characters, images, three-dimensional models, music, videos and the like generated by a computer to the real world after simulation, wherein the two kinds of information are mutually complemented, so that the enhancement of the real world is realized. In the prior art, the maintenance of the aero-engine is mainly finished by means of manual experience, namely, after the fault data of the engine on the aircraft are collected and processed to a certain extent, all characteristic attributes of the fault data are observed and analyzed in a frequency doubling mode, and which fault type is manually judged. The maintenance of the aero-engine has low working efficiency, inaccurate diagnosis result and great consumption of manpower and material resources.
Therefore, it is desirable to provide an AR-based aero-engine repair modeling method that assists in aero-engine repair based on AR technology, improves repair efficiency, reduces errors, and optimizes repair procedures.
Disclosure of Invention
The invention provides an AR-based aeroengine maintenance modeling method, which comprises the following steps: establishing an aero-engine model library, wherein the aero-engine model library is used for storing sample aero-engine three-dimensional models of various models; acquiring scanning information of an aeroengine to be maintained; acquiring a target sample aeroengine three-dimensional model from the aeroengine model library based on the scanning information of the aeroengine to be maintained; establishing a first knowledge graph, wherein the first knowledge graph is used for recording the associated fault types of various types of sample aeroengines under different sample operation parameters; establishing a second knowledge graph, wherein the second knowledge graph is used for recording fault characteristics corresponding to each associated fault type; acquiring historical operation parameter information of the aeroengine to be maintained; determining a candidate fault type corresponding to an aeroengine to be maintained based on historical operation parameter information of the aeroengine to be maintained and a first knowledge graph; acquiring the running state information of the aeroengine to be maintained, and predicting the fault information of the aeroengine to be maintained based on the historical running state information, the second knowledge graph and the candidate fault type corresponding to the aeroengine to be maintained; and establishing an AR model corresponding to the aeroengine to be maintained based on the predicted fault information of the aeroengine to be maintained and the target sample three-dimensional model of the aeroengine.
Further, the scanning information of the aeroengine to be maintained at least comprises laser radar scanning information and ultrasonic scanning information of the aeroengine to be maintained; based on the scanning information of the aero-engine to be maintained, acquiring a target sample aero-engine three-dimensional model from the aero-engine model library, wherein the method comprises the following steps: determining size information of the aeroengine to be maintained based on laser radar scanning information of the aeroengine to be maintained; and acquiring a target sample aeroengine three-dimensional model from the aeroengine model library based on the size information of the aeroengine to be maintained and the ultrasonic scanning information of the aeroengine to be maintained.
Further, the method further comprises: further comprises: acquiring laser radar scanning information of sample aeroengines of various models; for each model of sample aeroengine, determining size information of the sample aeroengine based on laser radar scanning information of the sample aeroengine; based on the size information of the sample aeroengines of each model, clustering the sample aeroengines of various models, and determining a plurality of sample aeroengine clustering clusters; and for each sample aeroengine cluster, determining ultrasonic scanning information of each model of sample aeroengine included in the sample aeroengine cluster at a plurality of candidate positions, and determining at least one target position corresponding to the sample aeroengine cluster from the plurality of candidate positions.
Further, the acquiring the scan information of the aeroengine to be repaired includes: determining a target sample aeroengine cluster from the plurality of sample aeroengine clusters based on the size information of the aeroengine to be repaired; and acquiring ultrasonic scanning information of the aeroengine to be maintained based on at least one target position corresponding to the target sample aeroengine cluster.
Further, based on the size information of the aeroengine to be repaired and the ultrasonic scanning information of the aeroengine to be repaired, obtaining a target sample aeroengine three-dimensional model from the aeroengine model library comprises the following steps: determining a target sample aeroengine from the target sample aeroengine cluster based on ultrasonic scanning information of sample aeroengines of each model included in the target sample aeroengine cluster at least at one target position and ultrasonic scanning information of the aeroengine to be maintained; and acquiring a sample aeroengine three-dimensional model corresponding to the target sample aeroengine from the aeroengine model library.
Further, determining a candidate fault type corresponding to the aeroengine to be maintained based on historical operating parameter information of the aeroengine to be maintained and a first knowledge graph comprises: for each sample operation parameter corresponding to the target sample aero-engine, calculating the similarity of the historical operation parameter information of the aero-engine to be maintained and the operation parameters of the sample operation parameters; determining a target sample operation parameter from a plurality of sample operation parameters corresponding to the target sample aeroengine based on the similarity of the historical operation parameter information of the aeroengine to be maintained and the operation parameter of each sample operation parameter corresponding to the target sample aeroengine; and determining the associated fault type of the target sample aeroengine under the target sample operation parameters from the first knowledge graph based on the target sample operation parameters and the target sample aeroengine, and taking the associated fault type as the candidate fault type corresponding to the aeroengine to be maintained.
Further, acquiring the running state information of the aero-engine to be maintained, including: and setting an operation state acquisition device at least one monitoring position of the aeroengine to be maintained, wherein the operation state acquisition device is used for acquiring operation state information of the aeroengine to be maintained, and the operation state information of the aeroengine to be maintained at least comprises temperature information, vibration information and sound information.
Further, predicting the fault information of the aeroengine to be maintained based on the historical operating state information, the second knowledge graph and the candidate fault type corresponding to the aeroengine to be maintained, including: for each candidate fault type, determining a state characteristic of the aircraft engine to be maintained corresponding to the candidate fault type based on historical operation parameter information of the aircraft engine to be maintained, determining a fault characteristic of the candidate fault type based on the second knowledge graph, and predicting fault information of the aircraft engine to be maintained based on the state characteristic of the aircraft engine to be maintained corresponding to the candidate fault type and the fault characteristic of the candidate fault type.
Further, based on the predicted fault information of the aeroengine to be maintained and the target sample three-dimensional model of the aeroengine, establishing an AR model corresponding to the aeroengine to be maintained, including: generating fault prompt information of different components of the three-dimensional model of the aircraft engine corresponding to the target sample based on the predicted fault information of the aircraft engine to be maintained; and fusing fault prompt information of different components corresponding to the three-dimensional model of the target sample aeroengine with the three-dimensional model of the target sample aeroengine, and establishing an AR model corresponding to the aeroengine to be maintained.
Further, based on the predicted fault information of the aeroengine to be maintained and the target sample three-dimensional model of the aeroengine, establishing an AR model corresponding to the aeroengine to be maintained, including: generating a suggested maintenance scheme corresponding to different components of the target sample aeroengine three-dimensional model based on the predicted fault information of the aeroengine to be maintained; generating a maintenance demonstration model corresponding to different components of the target sample aircraft engine three-dimensional model based on suggested maintenance schemes corresponding to different components of the target sample aircraft engine three-dimensional model; and fusing a suggested maintenance scheme corresponding to different components of the three-dimensional model of the target sample aeroengine, a maintenance demonstration model corresponding to different components of the three-dimensional model of the target sample aeroengine and the three-dimensional model of the target sample aeroengine, and establishing an AR model corresponding to the aeroengine to be maintained.
Compared with the prior art, the AR-based aeroengine maintenance modeling method provided by the invention has at least the following beneficial effects:
1. The method comprises the steps of pre-establishing an aeroengine model library, providing model support for subsequent AR model establishment, establishing a first knowledge graph and a second knowledge graph through analysis and processing of historical data of existing sample aeroengines of various models, providing a large amount of real data support for fault judgment of the aeroengine to be maintained, combing a process of predicting fault information of the aeroengine to be maintained by using the first knowledge graph and the second knowledge graph on the basis, improving accuracy and efficiency of predicting the fault information of the aeroengine to be maintained, further, establishing an AR model corresponding to the aeroengine to be maintained based on the predicted fault information of the aeroengine to be maintained and a target sample aeroengine three-dimensional model, providing more dynamic data support for maintenance personnel to check and maintain the aeroengine, realizing maintenance of the aeroengine based on the assistance of an AR technology, improving maintenance efficiency, reducing errors and optimizing the maintenance process.
2. The sample aeroengines in the same sample aeroengine cluster are determined by calculating the ultrasonic scanning similarity fluctuation parameters, and the target positions with larger structure differences exist, so that the aeroengines to be maintained are subjected to targeted scanning, the workload of ultrasonic scanning on the aeroengines to be maintained is reduced, meanwhile, the data volume of a three-dimensional model of the target sample aeroengines is reduced, and the modeling efficiency is improved.
3. Based on the predicted fault information of the aeroengine to be maintained, generating fault prompt information of different components of the three-dimensional model of the aeroengine corresponding to the target sample, carrying out fault auxiliary judgment data support on maintenance personnel in an AR mode, and simultaneously, generating maintenance demonstration models of different components of the three-dimensional model of the aeroengine corresponding to the target sample, and carrying out fault maintenance auxiliary data support on the maintenance personnel in an AR mode.
Drawings
The present specification will be further elucidated by way of example embodiments, which will be described in detail by means of the accompanying drawings. The embodiments are not limiting, in which like numerals represent like structures, wherein:
FIG. 1 is a flow diagram of an AR-based aeroengine repair modeling method according to some embodiments of the present description;
FIG. 2 is a flow diagram of acquiring scan information of an aircraft engine to be serviced according to some embodiments of the present description;
FIG. 3 is a schematic diagram of a first knowledge-graph, shown in accordance with some embodiments of the present description;
FIG. 4 is a schematic diagram of a second knowledge-graph, according to some embodiments of the present description;
FIG. 5 is a schematic illustration of establishing a corresponding AR model of an aircraft engine to be serviced, according to some embodiments of the present description;
FIG. 6 is a flow chart of clustering multiple model sample aircraft engines, according to some embodiments of the present description.
Detailed Description
In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings that are required to be used in the description of the embodiments will be briefly described below. It is apparent that the drawings in the following description are only some examples or embodiments of the present specification, and it is possible for those of ordinary skill in the art to apply the present specification to other similar situations according to the drawings without inventive effort. Unless otherwise apparent from the context of the language or otherwise specified, like reference numerals in the figures refer to like structures or operations.
FIG. 1 is a flow diagram of an AR-based aeroengine repair modeling method, as shown in FIG. 1, according to some embodiments of the present description, which may include the following steps.
And step 110, establishing an aero-engine model library.
The model library of the aero-engine is used for storing sample three-dimensional models of the aero-engines with various models.
The method specifically comprises the following steps:
S11, data collection and pretreatment: and collecting relevant data of the sample aeroengine, including the structure, the size, the working principle and the like. Such data may come from technical documents, CAD drawings, physical measurements, etc.
S12, preprocessing the collected relevant data of the sample aero-engine, including cleaning, format conversion and arrangement, so as to facilitate subsequent modeling.
S13, three-dimensional modeling: three-dimensional modeling of sample aircraft engines was performed using specialized three-dimensional modeling software (e.g., solidWorks, 3ds Max, blender, etc.). Based on the collected data, a three-dimensional model of each component of the sample aeroengine is created and the accuracy, proportion and authenticity of the details are ensured.
S14, texture and material mapping: texture and materials are added to the three-dimensional model of the sample aeroengine, so that the model looks more realistic. Specialized texture creation software can be used to create and edit textures, which are then mapped to the surface of the model.
S15, developing an interaction function: according to the requirements of AR application, interactive functions such as rotation, scaling, perspective and the like are developed, so that a user can view the three-dimensional model of the sample aeroengine in all directions. These interactive functions are implemented using an AR development kit (Augmented Reality Software Development Kit, SDK) and programming languages (e.g., unity, unreal Engine, etc.).
S16, optimizing and testing: the three-dimensional model of the sample aeroengine is optimized to ensure that it can run smoothly in AR applications without jamming or delay. Testing is performed under different equipment and environments, and compatibility and stability of a three-dimensional model of the sample aeroengine are ensured.
Step 120, obtaining scanning information of the aeroengine to be maintained.
In some embodiments, the scan information of the aircraft engine to be serviced includes at least lidar scan information and ultrasound scan information of the aircraft engine to be serviced.
Specifically, obtaining laser radar scan information of an aircraft engine to be serviced may include the steps of:
s21, selecting and preparing equipment: a lidar device suitable for aircraft engine scanning is selected. The lidar device should have high accuracy, high resolution and good stability to ensure the accuracy of the scanned data. And (3) calibrating and testing the laser radar equipment to ensure that the performance of the laser radar equipment reaches the optimal state.
S22, preparing an aeroengine: the aero-engine is placed in a stable and easy-to-scan position, so that the engine is ensured not to move or vibrate during the scanning process. If the engine is in a complex environment, it may be necessary to clear surrounding obstructions so that the lidar device can scan the engine without obstruction.
S23, scanning: and setting laser radar equipment, and adjusting scanning parameters such as scanning speed, resolution and scanning angle to adapt to the characteristics of the aeroengine. And starting the laser radar equipment, and carrying out omnibearing scanning on the aeroengine. During the scanning process, equipment stability is ensured, and a scanning path is continuous, so that the surface data of the complete aeroengine are acquired.
The ultrasonic scan information of the aircraft engine to be serviced may include ultrasonic scan information at a plurality of locations of the aircraft engine to be serviced.
FIG. 2 is a flow diagram of acquiring scan information of an aircraft engine to be serviced, as shown in FIG. 2, according to some embodiments of the present disclosure, including:
Acquiring laser radar scanning information of sample aeroengines of various models;
for each model of sample aero-engine, determining size information of the sample aero-engine based on laser radar scanning information of the sample aero-engine;
Based on the size information of the sample aeroengines of each model, clustering the sample aeroengines of various models, and determining a plurality of sample aeroengine clustering clusters;
For each sample aeroengine cluster, determining ultrasonic scanning information of each model of sample aeroengine included in the sample aeroengine cluster at a plurality of candidate positions, and determining at least one target position corresponding to the sample aeroengine cluster from the plurality of candidate positions;
Determining a target sample aeroengine cluster from a plurality of sample aeroengine clusters based on size information of the aeroengine to be repaired;
and acquiring ultrasonic scanning information of the aeroengine to be maintained based on at least one target position corresponding to the target sample aeroengine cluster.
Specifically, the laser radar scans the sample aeroengine, and in the process, the laser radar emits pulse laser beams which are scattered after contacting the engine surface and returned to the radar receiver, and the time difference is measured: the laser radar determines the distance between each point on the surface of the sample aeroengine and the radar by measuring the time difference between the laser emission and the laser receiving, namely the round trip time of the laser pulse, based on the measurement of the time difference between the laser emission and the laser receiving, the three-dimensional coordinates of each point on the surface of the sample aeroengine can be calculated by combining the emission angle and the scanning mode of the laser beam, the obtained three-dimensional point cloud image needs further data processing, such as filtering, denoising, smoothing and the like to improve the accuracy and the reliability of the data, then the three-dimensional modeling technology can be utilized to reconstruct a three-dimensional model of the sample aeroengine based on the processed point cloud data, and a plurality of dimension parameters of the sample aeroengine, such as length, width, height, diameter and the like, are obtained based on the three-dimensional model of the sample aeroengine, wherein the length refers to the maximum linear distance from the front end to the rear end of the sample aeroengine, the whole length of the sample aeroengine is reflected, the width is the transverse maximum dimension of the sample aeroengine, namely the measured value of the widest place, and the height refers to the maximum vertical distance from the top to bottom to top of the sample aeroengine, and the diameter represents the diameter of the fan.
For sample aeroengines of any two types, the appearance similarity of the sample aeroengines of the two types can be determined based on the size information of the sample aeroengines of the two types, and the sample aeroengines of various types are clustered based on the appearance similarity of the sample aeroengines of any two types through a K-means clustering algorithm to determine a plurality of sample aeroengine clusters.
FIG. 6 is a flow chart of clustering multiple model sample aircraft engines, as shown in FIG. 6, according to some embodiments of the present description, e.g., multiple model sample aircraft engines may be clustered according to the following flow:
s31, initializing the number K of sample aero-engine clustering clusters, randomly selecting K sample aero-engines as initial cluster centers, and executing S32;
S32, for each sample aeroengine in the sample aeroengines with various models, calculating the distance between the sample aeroengine and each cluster center based on the appearance similarity of the sample aeroengine corresponding to each cluster center, and distributing the distance to the sample aeroengine cluster closest to the cluster center, wherein the higher the appearance similarity of the sample aeroengine corresponding to the cluster center, the closer the distance between the sample aeroengine and the cluster center, and S33 is executed;
s33, for each sample aero-engine cluster, calculating a similarity mean value corresponding to each sample aero-engine included in the sample aero-engine cluster, and taking the sample aero-engine with the largest similarity mean value as a new cluster center to execute S34;
S34, repeating the step S32 and the step S33 until the cluster center determined by the current clustering and the cluster center determined by the last clustering are not changed any more or the preset single-round iteration times are reached, and executing the step S35;
S35, calculating appearance similarity fluctuation parameters corresponding to each sample aero-engine cluster, taking the sample aero-engine cluster with the appearance similarity fluctuation parameters larger than a preset appearance similarity fluctuation parameter threshold as an abnormal sample aero-engine cluster, and executing S36;
s36, judging whether the number of abnormal sample aero-engine clusters is larger than a preset number threshold (for example, half of the number of the sample aero-engine clusters), if so, adjusting the K value, executing S32, and if not, completing clustering.
For example only, the similarity mean may be calculated based on the following formula:
，
Wherein, For the corresponding similarity mean of the i-th sample aeroengine,
For the appearance similarity of the ith sample aeroengine and the jth sample aeroengine included in the sample aeroengine cluster to which the ith sample aeroengine belongs,
The total number of sample aeroengines included for the sample aeroengine cluster to which the i-th sample aeroengine belongs.
The profile similarity fluctuation parameter may be calculated based on the following formula:
，
Wherein, As the appearance similarity fluctuation parameter of the e-th sample aeroengine cluster,
For the similarity in appearance between the f-th sample aeroengine included in the e-th sample aeroengine cluster and the sample aeroengine corresponding to the cluster center of the e-th sample aeroengine cluster,
The total number of sample aeroengines included for the e-th sample aeroengine cluster.
The K value may be adjusted based on the following formula:
，/>，
Wherein, For the adjusted K value for the n +1 th round,
For the K value of the nth round,
For the number of abnormal sample aeroengine clusters in the nth round,
Is a preset parameter, and/>，
To get/>Is an integer part of (c).
The profile similarity of two model sample aeroengines may be calculated based on the following formula:
，
Wherein, For the similarity in profile of the ith sample aeroengine to the jth sample aeroengine,
Is a preset parameter, and/>，
For the value of the g-th dimensional parameter of the i-th sample aeroengine,
The value of the g-th dimensional parameter of the j-th sample aeroengine,
Is the total number of size parameters.
For each sample aeroengine cluster, calculating the appearance similarity between the aeroengine to be maintained and the sample aeroengine cluster based on the size information of the aeroengine to be maintained and the size information of the sample aeroengine corresponding to the cluster center of the sample aeroengine cluster, and taking the sample aeroengine cluster with the appearance similarity larger than a preset appearance similarity threshold as a target sample aeroengine cluster.
For each candidate position, the ultrasonic scanning similarity of the ultrasonic scanning information of any two types of sample aeroengines in the candidate position, which is included in the sample aeroengine cluster, can be calculated, and the ultrasonic scanning similarity fluctuation parameter corresponding to the candidate position is calculated. And taking the candidate position with the ultrasonic scanning similarity fluctuation parameter larger than the preset ultrasonic scanning similarity fluctuation parameter threshold as the target position.
For example, the ultrasound scan similarity fluctuation parameter may be calculated based on the following formula:
，/>，
，
Wherein, For the ultrasonic scanning similarity fluctuation parameter corresponding to the xth candidate position,
In order to take the maximum value to operate,
For the local ultrasonic scanning similarity fluctuation parameter corresponding to the j-th sample aeroengine included in the sample aeroengine cluster,
For the ultrasonic scanning similarity of the ultrasonic scanning information of the jth sample aeroengine in the xth candidate position included in the sample aeroengine cluster and the ultrasonic scanning information of the jth sample aeroengine in the xth candidate position included in the sample aeroengine cluster,
Y is the total number of sample aeroengines included in the sample aeroengine cluster.
After at least one target position corresponding to the target sample aero-engine cluster is determined, controlling an ultrasonic scanning device to perform ultrasonic scanning on the aero-engine to be maintained at the at least one target position, and acquiring ultrasonic scanning information of the aero-engine to be maintained.
It can be understood that by calculating the ultrasonic scanning similarity fluctuation parameter, the sample aero-engines in the same sample aero-engine cluster are determined, and the target positions with larger structure differences exist, so that the aero-engines to be maintained are subjected to targeted scanning, the workload of ultrasonic scanning on the aero-engines to be maintained is reduced, meanwhile, the data volume of the follow-up three-dimensional model for determining the target sample aero-engines is reduced, and the modeling efficiency is improved.
And 130, acquiring a target sample aeroengine three-dimensional model from an aeroengine model library based on scanning information of the aeroengine to be maintained.
The method specifically comprises the following steps:
determining size information of the aeroengine to be maintained based on laser radar scanning information of the aeroengine to be maintained;
And acquiring a target sample aero-engine three-dimensional model from the aero-engine model library based on the size information of the aero-engine to be maintained and the ultrasonic scanning information of the aero-engine to be maintained.
In some embodiments, obtaining a target sample aircraft engine three-dimensional model from an aircraft engine model library based on dimensional information of an aircraft engine to be serviced and ultrasonic scan information of the aircraft engine to be serviced, comprises:
Determining a target sample aeroengine from the target sample aeroengine cluster based on ultrasonic scanning information of sample aeroengines of each model included in the target sample aeroengine cluster at least one target position and ultrasonic scanning information of aeroengines to be maintained;
and acquiring a sample aeroengine three-dimensional model corresponding to the target sample aeroengine from the aeroengine model library.
Specifically, for each target sample aeroengine cluster, based on ultrasonic scanning information of an aeroengine to be maintained at least one target position corresponding to the target sample aeroengine cluster and ultrasonic scanning information of sample aeroengines of each model included in the target sample aeroengine cluster at least one target position, calculating ultrasonic scanning similarity between the aeroengine to be maintained and the sample aeroengine, and taking the sample aeroengine with the maximum ultrasonic scanning similarity as the target sample aeroengine.
And 140, establishing a first knowledge graph.
The first knowledge graph is used for recording the associated fault types of the sample aeroengines with various models under different sample operation parameters. The operating parameters may include operating environment (e.g., temperature, humidity, wind speed, air pressure, etc.) and equipment operating parameters (e.g., thrust, power, fuel consumption rate, engine pressure ratio, low pressure rotor speed, high pressure rotor speed, engine exhaust temperature, oil pressure, oil temperature, etc.).
The associated fault types of the sample aero-engine under the different sample operating parameters can characterize the fault types of the sample aero-engine which are easy to occur under the different sample operating parameters. Fig. 3 is a schematic diagram of a first knowledge graph, shown in some embodiments of the present disclosure, as shown in fig. 3, under a sample operating parameter 1, the associated fault types of the sample aero-engine 1 include: rotor failure, thrust component failure, turbine system failure, and cylinder system failure; under the sample operating parameter 2, the associated fault types of the sample aeroengine 1 include: air passage failure and fan and compressor failure.
And 150, establishing a second knowledge graph.
The second knowledge graph is used for recording fault characteristics corresponding to each associated fault type. FIG. 4 is a schematic diagram of a second knowledge graph illustrating a failure signature corresponding to an air path failure including intake surge, abnormal fluctuations in temperature and pressure within an engine, and compressor failure including engine surge, reduced air pressure generated by a compressor, and elevated exhaust temperature, as shown in FIG. 4, according to some embodiments of the present disclosure.
Step 160, obtaining historical operating parameter information of the aero-engine to be maintained.
The historical operating parameter information for the aircraft engine to be serviced may include operating environment (e.g., temperature, humidity, wind speed, air pressure, etc.) and equipment operating parameters (e.g., thrust, power, fuel consumption rate, engine pressure ratio, low pressure rotor speed, high pressure rotor speed, engine exhaust temperature, slip pressure, slip temperature, etc.) of the aircraft engine to be serviced over a period of time.
Step 170, determining candidate fault types corresponding to the aeroengine to be maintained based on the historical operation parameter information of the aeroengine to be maintained and the first knowledge graph.
The method specifically comprises the following steps:
for each sample operation parameter corresponding to the target sample aero-engine, calculating the similarity of the historical operation parameter information of the aero-engine to be maintained and the operation parameters of the sample operation parameters;
Determining a target sample operation parameter from a plurality of sample operation parameters corresponding to the target sample aeroengine based on the similarity of the historical operation parameter information of the aeroengine to be maintained and each sample operation parameter corresponding to the target sample aeroengine, for example, taking the sample operation parameter with the maximum operation parameter similarity as the target sample operation parameter;
And determining the associated fault type of the target sample aeroengine under the target sample operation parameters from the first knowledge graph based on the target sample operation parameters and the target sample aeroengine, and taking the associated fault type as a candidate fault type corresponding to the aeroengine to be maintained.
For example, if the target sample aeroengine is the sample aeroengine 1 and the target sample operation parameter is the sample operation parameter 1, the relevant fault type of the sample aeroengine 1 in the sample operation parameter 1 is obtained from the first knowledge graph and is used as the candidate fault type corresponding to the aeroengine to be maintained.
And 180, acquiring the running state information of the aeroengine to be maintained, and predicting the fault information of the aeroengine to be maintained based on the historical running state information, the second knowledge graph and the candidate fault type corresponding to the aeroengine to be maintained.
In some embodiments, obtaining operational status information of an aircraft engine to be serviced includes:
And setting an operation state acquisition device at least one monitoring position of the aeroengine to be maintained, wherein the operation state acquisition device is used for acquiring operation state information of the aeroengine to be maintained, and the operation state information of the aeroengine to be maintained at least comprises temperature information, vibration information and sound information and can also comprise other types of information, such as pipeline air pressure information, air flow information and the like.
In some embodiments, predicting the fault information of the aircraft engine to be serviced based on the historical operating state information, the second knowledge-graph, and the candidate fault type corresponding to the aircraft engine to be serviced includes:
For each candidate fault type, determining the state characteristics of the aircraft engine to be maintained corresponding to the candidate fault type based on the historical operation parameter information of the aircraft engine to be maintained, determining the fault characteristics of the candidate fault type based on the second knowledge graph, and predicting the fault information of the aircraft engine to be maintained based on the state characteristics of the aircraft engine to be maintained corresponding to the candidate fault type and the fault characteristics of the candidate fault type.
For example, the fault signature corresponding to candidate fault type 1 includes engine surge, air pressure drop by the compressor, and exhaust gas temperature rise, and the status signature corresponding to candidate fault type 1 for the aircraft engine to be serviced may include engine vibration information, air pressure information by the compressor, and exhaust gas temperature information.
The matching degree of the fault of the aeroengine to be maintained and the candidate fault type can be judged through the fault judging model based on the state characteristics of the candidate fault type corresponding to the aeroengine to be maintained and the fault characteristics corresponding to the candidate fault type.
And 190, establishing an AR model corresponding to the aeroengine to be maintained based on the predicted fault information of the aeroengine to be maintained and the target sample aeroengine three-dimensional model.
Fig. 5 is a schematic diagram of building an AR model corresponding to an aeroengine to be repaired according to some embodiments of the present disclosure, where, as shown in fig. 5, building an AR model corresponding to an aeroengine to be repaired based on predicted fault information of the aeroengine to be repaired and a target sample three-dimensional model of the aeroengine, includes:
generating fault prompt information of different components of the three-dimensional model of the aeroengine corresponding to the target sample based on the predicted fault information of the aeroengine to be maintained, wherein the fault prompt information can comprise information such as names of the components with faults, fault types, possibility of faults and the like;
Fusing fault prompt information of different components of the three-dimensional model of the aeroengine corresponding to the target sample with the three-dimensional model of the aeroengine corresponding to the target sample, and establishing an AR model corresponding to the aeroengine to be maintained;
Generating suggested maintenance schemes of different components of the three-dimensional model of the target sample aeroengine based on the predicted fault information of the aeroengine to be maintained, wherein the suggested maintenance schemes of the different components under different fault types can be pre-stored in a maintenance scheme database, and the suggested maintenance schemes of the different components of the three-dimensional model of the target sample aeroengine can be directly extracted from the maintenance scheme database;
Generating a maintenance demonstration model corresponding to the different components of the three-dimensional model of the target sample aeroengine based on the suggested maintenance schemes corresponding to the different components of the three-dimensional model of the target sample aeroengine, wherein the maintenance demonstration model can be used for demonstrating a maintenance flow corresponding to the suggested maintenance schemes;
And fusing the suggested maintenance schemes of the different components of the three-dimensional model of the corresponding target sample aeroengine, the maintenance demonstration models of the different components of the three-dimensional model of the corresponding target sample aeroengine and the three-dimensional model of the target sample aeroengine, and establishing an AR model corresponding to the aeroengine to be maintained.
It can be understood that, based on the predicted fault information of the aeroengine to be maintained, fault prompt information corresponding to different components of the three-dimensional model of the target sample aeroengine is generated, fault auxiliary judgment data support can be performed on maintenance personnel in an AR mode, meanwhile, maintenance demonstration models corresponding to different components of the three-dimensional model of the target sample aeroengine are generated, and fault maintenance auxiliary data support can be performed on the maintenance personnel in an AR mode.
Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments of this specification. Other variations are possible within the scope of this description. Thus, by way of example, and not limitation, alternative configurations of embodiments of the present specification may be considered as consistent with the teachings of the present specification. Accordingly, the embodiments of the present specification are not limited to only the embodiments explicitly described and depicted in the present specification.
Claims (10)
1. An AR-based aeroengine repair modeling method, comprising:
establishing an aero-engine model library, wherein the aero-engine model library is used for storing sample aero-engine three-dimensional models of various models;
Acquiring scanning information of an aeroengine to be maintained;
acquiring a target sample aeroengine three-dimensional model from the aeroengine model library based on the scanning information of the aeroengine to be maintained;
establishing a first knowledge graph, wherein the first knowledge graph is used for recording the associated fault types of various types of sample aeroengines under different sample operation parameters;
Establishing a second knowledge graph, wherein the second knowledge graph is used for recording fault characteristics corresponding to each associated fault type;
acquiring historical operation parameter information of the aeroengine to be maintained;
Determining a candidate fault type corresponding to an aeroengine to be maintained based on historical operation parameter information of the aeroengine to be maintained and a first knowledge graph;
Acquiring the running state information of the aeroengine to be maintained, and predicting the fault information of the aeroengine to be maintained based on the historical running state information, the second knowledge graph and the candidate fault type corresponding to the aeroengine to be maintained;
and establishing an AR model corresponding to the aeroengine to be maintained based on the predicted fault information of the aeroengine to be maintained and the target sample three-dimensional model of the aeroengine.
2. The AR-based aircraft engine repair modeling method of claim 1, wherein the scan information of the aircraft engine to be repaired includes at least laser radar scan information and ultrasonic scan information of the aircraft engine to be repaired;
Based on the scanning information of the aero-engine to be maintained, acquiring a target sample aero-engine three-dimensional model from the aero-engine model library, wherein the method comprises the following steps:
determining size information of the aeroengine to be maintained based on laser radar scanning information of the aeroengine to be maintained;
And acquiring a target sample aeroengine three-dimensional model from the aeroengine model library based on the size information of the aeroengine to be maintained and the ultrasonic scanning information of the aeroengine to be maintained.
3. The AR-based aeroengine repair modeling method of claim 2, further comprising:
Acquiring laser radar scanning information of sample aeroengines of various models;
for each model of sample aeroengine, determining size information of the sample aeroengine based on laser radar scanning information of the sample aeroengine;
Based on the size information of the sample aeroengines of each model, clustering the sample aeroengines of various models, and determining a plurality of sample aeroengine clustering clusters;
And for each sample aeroengine cluster, determining ultrasonic scanning information of each model of sample aeroengine included in the sample aeroengine cluster at a plurality of candidate positions, and determining at least one target position corresponding to the sample aeroengine cluster from the plurality of candidate positions.
4. The AR-based aircraft engine repair modeling method of claim 3, wherein the obtaining scan information of the aircraft engine to be repaired comprises:
determining a target sample aeroengine cluster from the plurality of sample aeroengine clusters based on the size information of the aeroengine to be repaired;
And acquiring ultrasonic scanning information of the aeroengine to be maintained based on at least one target position corresponding to the target sample aeroengine cluster.
5. The AR-based aircraft engine repair modeling method of claim 4, wherein obtaining a target sample aircraft engine three-dimensional model from the aircraft engine model library based on dimensional information of the aircraft engine to be repaired and ultrasonic scan information of the aircraft engine to be repaired comprises:
Determining a target sample aeroengine from the target sample aeroengine cluster based on ultrasonic scanning information of sample aeroengines of each model included in the target sample aeroengine cluster at least at one target position and ultrasonic scanning information of the aeroengine to be maintained;
and acquiring a sample aeroengine three-dimensional model corresponding to the target sample aeroengine from the aeroengine model library.
6. The AR-based aircraft engine repair modeling method of claim 5, wherein determining the candidate fault type for the aircraft engine to be repaired based on historical operating parameter information of the aircraft engine to be repaired and the first knowledge graph comprises:
For each sample operation parameter corresponding to the target sample aero-engine, calculating the similarity of the historical operation parameter information of the aero-engine to be maintained and the operation parameters of the sample operation parameters;
determining a target sample operation parameter from a plurality of sample operation parameters corresponding to the target sample aeroengine based on the similarity of the historical operation parameter information of the aeroengine to be maintained and the operation parameter of each sample operation parameter corresponding to the target sample aeroengine;
and determining the associated fault type of the target sample aeroengine under the target sample operation parameters from the first knowledge graph based on the target sample operation parameters and the target sample aeroengine, and taking the associated fault type as the candidate fault type corresponding to the aeroengine to be maintained.
7. The AR-based aeroengine repair modeling method of any of claims 1-6, wherein obtaining operational status information of the aeroengine to be repaired comprises:
And setting an operation state acquisition device at least one monitoring position of the aeroengine to be maintained, wherein the operation state acquisition device is used for acquiring operation state information of the aeroengine to be maintained, and the operation state information of the aeroengine to be maintained at least comprises temperature information, vibration information and sound information.
8. The AR-based aeroengine repair modeling method of any of claims 1-6, wherein predicting fault information for the aeroengine to be repaired based on the historical operating state information, the second knowledge-graph, and candidate fault types for the aeroengine to be repaired comprises:
For each candidate fault type, determining a state characteristic of the aircraft engine to be maintained corresponding to the candidate fault type based on historical operation parameter information of the aircraft engine to be maintained, determining a fault characteristic of the candidate fault type based on the second knowledge graph, and predicting fault information of the aircraft engine to be maintained based on the state characteristic of the aircraft engine to be maintained corresponding to the candidate fault type and the fault characteristic of the candidate fault type.
9. The AR-based aeroengine repair modeling method of any of claims 1-6, wherein establishing the corresponding AR model of the aeroengine to be repaired based on the predicted fault information of the aeroengine to be repaired and the target sample three-dimensional model of the aeroengine comprises:
Generating fault prompt information of different components of the three-dimensional model of the aircraft engine corresponding to the target sample based on the predicted fault information of the aircraft engine to be maintained;
And fusing fault prompt information of different components corresponding to the three-dimensional model of the target sample aeroengine with the three-dimensional model of the target sample aeroengine, and establishing an AR model corresponding to the aeroengine to be maintained.
10. The AR-based aeroengine repair modeling method of any of claims 1-6, wherein establishing the corresponding AR model of the aeroengine to be repaired based on the predicted fault information of the aeroengine to be repaired and the target sample three-dimensional model of the aeroengine comprises:
Generating a suggested maintenance scheme corresponding to different components of the target sample aeroengine three-dimensional model based on the predicted fault information of the aeroengine to be maintained;
generating a maintenance demonstration model corresponding to different components of the target sample aircraft engine three-dimensional model based on suggested maintenance schemes corresponding to different components of the target sample aircraft engine three-dimensional model;
And fusing a suggested maintenance scheme corresponding to different components of the three-dimensional model of the target sample aeroengine, a maintenance demonstration model corresponding to different components of the three-dimensional model of the target sample aeroengine and the three-dimensional model of the target sample aeroengine, and establishing an AR model corresponding to the aeroengine to be maintained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410555406.XA CN118135143A (en) | 2024-05-07 | 2024-05-07 | AR-based aeroengine maintenance modeling method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410555406.XA CN118135143A (en) | 2024-05-07 | 2024-05-07 | AR-based aeroengine maintenance modeling method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118135143A true CN118135143A (en) | 2024-06-04 |
Family
ID=91234084
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410555406.XA Pending CN118135143A (en) | 2024-05-07 | 2024-05-07 | AR-based aeroengine maintenance modeling method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118135143A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190339688A1 (en) * | 2016-05-09 | 2019-11-07 | Strong Force Iot Portfolio 2016, Llc | Methods and systems for data collection, learning, and streaming of machine signals for analytics and maintenance using the industrial internet of things |
| CN115510253A (en) * | 2022-08-26 | 2022-12-23 | 国网山东省电力公司枣庄供电公司 | Transformer substation equipment AR auxiliary maintenance method and terminal based on three-dimensional knowledge graph |
| CN117367807A (en) * | 2022-06-28 | 2024-01-09 | 中国航发商用航空发动机有限责任公司 | Method, system, equipment and medium for diagnosing faults of aero-engine |
| CN117988988A (en) * | 2024-02-01 | 2024-05-07 | 华夏航空设备(北京)有限公司 | Aeroengine fault diagnosis method and system based on feature amplification |
-
2024
- 2024-05-07 CN CN202410555406.XA patent/CN118135143A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190339688A1 (en) * | 2016-05-09 | 2019-11-07 | Strong Force Iot Portfolio 2016, Llc | Methods and systems for data collection, learning, and streaming of machine signals for analytics and maintenance using the industrial internet of things |
| CN117367807A (en) * | 2022-06-28 | 2024-01-09 | 中国航发商用航空发动机有限责任公司 | Method, system, equipment and medium for diagnosing faults of aero-engine |
| CN115510253A (en) * | 2022-08-26 | 2022-12-23 | 国网山东省电力公司枣庄供电公司 | Transformer substation equipment AR auxiliary maintenance method and terminal based on three-dimensional knowledge graph |
| CN117988988A (en) * | 2024-02-01 | 2024-05-07 | 华夏航空设备(北京)有限公司 | Aeroengine fault diagnosis method and system based on feature amplification |
Non-Patent Citations (2)
| Title |
|---|
| 李荣强等: "航空装备虚拟维修训练仿真模型快速开发技术", 《系统仿真学报》, vol. 33, no. 05, 31 May 2021 (2021-05-31), pages 1167 - 1176 * |
| 王英杰;徐璐;王宏伟;乔木;: "大数据技术在特高压变电站运维中的应用", 电力大数据, no. 11, 21 November 2018 (2018-11-21) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bruno et al. | Historic Building Information Modelling: performance assessment for diagnosis-aided information modelling and management | |
| CN110832351B (en) | Method for capturing an incident wind field and modeling the same by means of a LIDAR sensor | |
| Lange et al. | Probabilistic CFD simulation of a high-pressure compressor stage taking manufacturing variability into account | |
| Lange et al. | Introduction of a parameter based compressor blade model for considering measured geometry uncertainties in numerical simulation | |
| CN106779200A (en) | Based on the Wind turbines trend prediction method for carrying out similarity in the historical data | |
| CN109297713B (en) | Steam turbine fault diagnosis method based on stable and non-stable vibration signal characteristic selection | |
| JP5292477B2 (en) | Diagnostic device and diagnostic method | |
| CN108804850B (en) | Method for predicting parameters of aircraft engine in transient acceleration process based on spatial reconstruction | |
| CN109033178B (en) | Method for mining Glandum causal relationship among visibility multi-dimensional space-time data | |
| WO2020090767A1 (en) | Abnormality diagnostic device, abnormality diagnostic method, and program | |
| KR20210018397A (en) | System and method for automatic 3-dimensional scanning | |
| CN116861544A (en) | Building abnormal vibration source positioning method based on edge cloud cooperation and related equipment | |
| US20200103846A1 (en) | Blade sentencing | |
| CN118135143A (en) | AR-based aeroengine maintenance modeling method | |
| CN116977553A (en) | 3 DMAX-based three-dimensional parametric modeling method and device for plant pipeline | |
| Kaszynski et al. | Experimental validation of an optically measured geometric mistuning model using a system id approach | |
| Safari et al. | A novel combination of adaptive tools for turbomachinery airfoil shape optimization using a real-coded genetic algorithm | |
| Marx et al. | A machine learning based approach of performance estimation for high-pressure compressor airfoils | |
| CN116310194A (en) | Three-dimensional model reconstruction method, system, equipment and storage medium for power distribution station room | |
| CN115329493A (en) | Impeller mechanical fault detection method based on centrifugal pump digital twin model | |
| Kaszynski et al. | Automated meshing algorithm for generating as-manufactured finite element models directly from as-measured fan blades and integrally bladed disks | |
| Meckstroth et al. | Point Cloud to Parameter: An Inverse Geometric Approach to Probabilistic Design | |
| Kaszynski et al. | Uncertainties of an automated optical 3D geometry measurement, modeling, and analysis process for mistuned ibr reverse engineering | |
| Schlüter et al. | Probabilistic CFD Analysis of a High-Pressure Compressor Under Consideration of Manufacturing and In-Service Variability | |
| Majgaonkar et al. | Assessing LiDAR training data quantities for classification models |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |