Aviation equipment inspection and maintenance method based on target detection model and MBSETest

Abstract

The present invention discloses an aviation equipment inspection and maintenance method based on a target detection model and MBSETest, including: S1) creating a system model of aviation equipment in an MBSE environment, defining components, fault types and inspection requirements of aviation equipment; S2) integrating the target detection model into the MBSETest framework, receiving real-time or historical inspection image data for inspection algorithm training prediction; and associating the inspection results with the components and fault types defined in the system model; S3) automatically generating test cases based on the inspection requirements in the system model and the inspection results of the target detection model; S4) running inspection tests in the MBSETest environment to determine the inspection accuracy and consistency; S5) feeding back the test and inspection results to the system model to generate a maintenance plan. The present invention can provide a comprehensive and automated aviation equipment inspection and maintenance solution to achieve more efficient and systematic fault diagnosis and maintenance.

Description

Technical Field

[0001] The present invention relates to an aviation equipment detection and maintenance method, in particular to an aviation equipment detection and maintenance method based on a target detection model and MBSETest. Background Art

[0002] With the rapid development of aviation technology, the structure and function of aviation equipment are becoming increasingly complex, and the requirements for its inspection and maintenance are becoming more stringent. Traditional aviation equipment inspection methods usually rely on manual inspection and simple inspection equipment, which have the defects of cumbersome operation, low efficiency and easy human omission.

[0003] The current aviation equipment inspection and maintenance system has the following technical problems: 1. Incomplete detection: The existing detection system cannot cover the various complex fault types of aviation equipment, making it difficult to achieve fully automatic and comprehensive detection.

[0004] 2. Insufficient detection accuracy: Since aviation equipment contains a large number of delicate and complex components, traditional detection technology is difficult to achieve high-precision positioning and detection, which can easily lead to missed detection or misjudgment.

[0005] 3. Manual verification is inefficient: In the traditional testing process, test results need to be manually analyzed and verified, and test case generation and execution are time-consuming and inefficient.

[0006] 4. Poor system integration: The existing system detection and testing processes are often separated and lack effective integration with system models and test cases, making it difficult to conduct system-level verification and maintenance planning.

[0007] 5. Insufficient maintenance and prediction: The lack of ability to use detection data for intelligent analysis and prediction makes it difficult to carry out preventive maintenance, resulting in delayed maintenance and sudden failures. Summary of the invention

[0008] The technical problem to be solved by the present invention is to provide an aviation equipment detection and maintenance method based on a target detection model and MBSETest, which can provide a comprehensive and automated aviation equipment detection and maintenance solution to achieve more efficient and systematic fault diagnosis and maintenance.

[0009] The technical solution adopted by the present invention to solve the above technical problems is to provide an aviation equipment inspection and maintenance method based on a target detection model and MBSETest, which includes the following steps: S1) creating a system model of aviation equipment in an MBSE environment, defining the components, fault types and inspection requirements of the aviation equipment, and using this as the basis for the inspection task; S2) integrating the target detection model into the MBSETest framework, receiving real-time or historical inspection image data for inspection algorithm training and prediction; and associating the inspection results with the components and fault types defined in the system model; S3) automatically generating test cases based on the inspection requirements in the system model and the inspection results of the target detection model; S4) running the inspection test in the MBSETest environment, comparing the results of the target detection with the expected output in the system model, and judging the accuracy and consistency of the inspection; S5) feeding back the test and inspection results to the system model, updating the system's maintenance log and equipment status, and generating maintenance recommendations and repair plans.

[0010] Furthermore, the aviation equipment in step S1 is a fuselage, an engine or a landing gear of an aircraft.

[0011] Furthermore, the target detection model in step S2 is a YOLO, Faster-R-CNN or ResNet deep learning model.

[0012] Furthermore, the step S2 collects images and video data of aviation equipment in real time through a high-definition camera or infrared thermal imaging equipment, and pre-processes the collected images; then uses the target detection model to analyze the pre-processed images, detects and annotates faults or anomalies in the images; the step S3 imports the detection results into the MBSE system, associates the detection results with the system model, and generates test cases related to the faults or anomalies.

[0013] Furthermore, the step S2 pre-processes the collected image, including denoising, contrast enhancement, brightness adjustment, edge sharpening detection, image scaling and format conversion.

[0014] Furthermore, the faults or anomalies detected and marked by the target detection model in step S2 include surface cracks, wear or loose bolts, and output the fault type, fault location and severity information.

[0015] Furthermore, when detecting cracks in aircraft engine blades, a high-definition camera is used to capture high-resolution images of the engine. After image preprocessing, the images are input into the YOLOv8 target detection module, which identifies and locates the cracks on the blade surface. The detection results are imported into the MBSE system model, and test cases related to crack detection are automatically generated. By running these test cases in the MBSETest environment, it is verified whether the severity of the cracks exceeds the system safety standard. If the test results indicate that there is a risk, a maintenance plan is generated and immediate maintenance is recommended.

[0016] Compared with the prior art, the present invention has the following beneficial effects: the aviation equipment inspection and maintenance method based on the target detection model and MBSETest provided by the present invention combines the target detection model with MBSETest to improve the comprehensiveness and accuracy of detection, realize automated system-level verification, and improve the accuracy and efficiency of maintenance through closed-loop feedback, thereby realizing more efficient and systematic fault diagnosis and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flowchart of aviation equipment inspection and maintenance based on the target detection model and MBSETest of the present invention; Figure 2 Schematic diagram of the image acquisition and preprocessing process of the present invention; Figure 3 Schematic diagram of target detection and recognition process of the present invention; Figure 4 It is a schematic diagram of the detection and maintenance process of the present invention. DETAILED DESCRIPTION

[0018] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0019] In order to improve detection efficiency and accuracy, it is necessary to introduce computer vision and deep learning-based target detection technology. The present invention can be applied in the following scenarios: 1. Real-time detection and definition: Use target detection technology (such as YOLO, Faster-R-CNN) to detect external damage, cracks, wear, etc. of aviation equipment in real time and locate problem areas.

[0020] 2. Multi-target detection: Ability to simultaneously identify multiple types of faults, such as loose bolts, surface cracks, and corroded areas.

[0021] 3. Automatic analysis: Automatically analyze images through trained deep learning models to reduce the time and errors of manual inspection.

[0022] YOLO (You Only Look Once) is a real-time target detection algorithm developed in recent years. It is widely used in the field of image detection for its high efficiency and accuracy. As the latest version of the YOLO family, YOLOv8 further improves the detection accuracy and real-time performance, and is suitable for use in complex scenarios of aviation equipment detection. At the same time, model-based system engineering test (MBSETest) has advantages in system design, requirements management and test coverage, and achieves consistency in requirements, testing and verification through system modeling.

[0023] Based on the YOLOv8 target detection model, it is technically possible to detect targets in aviation application scenarios and improve the detection range and efficiency. Combined with MBSETest, it can realize the linkage upgrade of the entire maintenance and detection system, making it more intelligent and automated.

[0024] See also Figure 1 , the aviation equipment detection and maintenance method based on target detection model and MBSETest provided by the present invention includes: S1. First, create a system model of aviation equipment in the MBSE environment, define the structure, function and performance requirements of the equipment, and model the possible fault types and detection requirements of the equipment in the system model as the basis for the detection task.

[0025] S2. Then the target detection technology is integrated into the MBSETest framework. The system receives real-time or historical detection image data and performs detection algorithm training prediction. The detection results are associated with the components and fault types defined in the system model, and the detection information is automatically marked and fed back.

[0026] S3. Automatically generate test cases based on the requirements and test results in the system model. These cases include detection of different fault types, device status verification, etc. MBSETest can ensure that the test cases cover all key components and test requirements specified in the model.

[0027] S4. Run the detection test in the MBSETest environment to verify whether the detection system can correctly identify and report equipment problems. During the test execution, the results of the target detection are compared with the expected output in the system model to determine the accuracy and consistency of the detection.

[0028] S5. Integrate the test results and detection feedback back into the MBSETest system model, update the system maintenance log and equipment status, and continuously improve the model through this closed-loop process to make it closer to actual operation and maintenance needs.

[0029] The present invention consists of the following main modules: Image acquisition module: responsible for collecting image and video data of aviation equipment in real time through high-definition cameras or infrared thermal imaging devices.

[0030] Image preprocessing module: preprocess the collected images, including denoising, contrast enhancement, image scaling and format conversion, to adapt to subsequent detection processing.

[0031] YOLOv8 target detection module: Use the YOLOv8 deep learning model to analyze the preprocessed images, detect and annotate faults or anomalies in the images.

[0032] MBSE system integration module: imports the test results into the MBSE system (such as Cameo Systems Modeler), associates the test results with the system model, and generates relevant test cases.

[0033] Test case generation and execution module: automatically generates test cases related to the detection results and executes them in the MBSETest environment to verify whether the system meets the design and performance requirements.

[0034] Feedback and maintenance decision module: Feedback test and detection results to the system model and generate maintenance recommendations and repair plans.

[0035] The working principle of the present invention is as follows: the image acquisition module acquires real-time images and improves the image quality through the image preprocessing module. The YOLO target detection module processes the image, automatically detects and marks the fault. The detection results are imported into the system model through the MBSE system integration module to generate test cases. The test case generation and execution module performs tests in the MBSEtest environment to ensure that the detected faults meet the system model requirements. The feedback and maintenance decision module generates maintenance suggestions based on the test and detection results to achieve closed-loop maintenance management.

[0036] The specific implementation steps of the present invention are: Step 1: Image acquisition and preprocessing Place high-definition cameras and necessary sensors near the detection equipment to continuously collect image data of aviation equipment. Pre-process the collected images to improve image quality and optimize detection results, such as Figure 2 shown.

[0037] Step 2: Target Detection and Recognition The preprocessed image is input into the YOLO object detection module to automatically identify and locate anomalies or faults in the image, such as cracks, wear, loose bolts, etc. The output includes information about the fault type, location, and severity, such as Figure 3 shown.

[0038] Step 3: Integration and analysis of test results The output results of the target detection module are imported into the MBSE system model and data is transmitted through interfaces or APIs. The MBSE system automatically generates fault-related test cases based on the detection results to verify whether the detected anomalies meet the equipment design and performance standards.

[0039] Step 4: Test case execution and verification The output results of the target detection module are imported into the MBSE system model and data is transmitted through the interface or API. The test module determines the device status based on the execution results and outputs a test report.

[0040] Step 5: Result feedback and decision support: The test and inspection results are fed back to the MBSE system model, automatically updating the status and maintenance log of the equipment. Based on model analysis and historical inspection data, the system generates intelligent maintenance suggestions and repair plans for engineers to refer to and implement, such as Figure 4 shown.

[0041] The present invention can realize efficient, automated and systematic detection and maintenance of aviation equipment, ensuring that the equipment is always in the best condition and meets aviation safety standards; the present invention can be widely used in the detection and maintenance of aircraft fuselages, engines, landing gear and other key components. Due to the scalability of the technology, more sensor data and deep learning models can be combined in the future to further enhance the detection and maintenance capabilities. When detecting cracks in aircraft engine blades, high-definition cameras capture high-resolution images of the engine, which are input into the YOLOv8 target detection module after preprocessing. The module identifies and locates cracks on the surface of the blades. The detection results are imported into the MBSE system model, and test cases related to crack detection are automatically generated. By running these cases in the MBSETest environment, it is verified whether the severity of the cracks exceeds the system safety standards. If the test results indicate that there is a risk, the system will generate a maintenance plan and recommend immediate maintenance.

[0042] Although the present invention has been disclosed as above with preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art may make some modifications and improvements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be based on the definition of the claims.

Claims

1. An aviation equipment detection and maintenance method based on target detection model and MBSETest, characterized in that: The steps include: S1) Create a system model of aviation equipment in the MBSE environment, define the components, fault types and detection requirements of aviation equipment, and use this as the basis for detection tasks; S2) Integrate the target detection model into the MBSETest framework, receive real-time or historical detection image data for detection algorithm training and prediction; and associate the detection results with the components and fault types defined in the system model; S3) Automatically generate test cases based on the detection requirements in the system model and the detection results of the target detection model; S4) Run the detection test in the MBSETest environment, compare the target detection results with the expected output in the system model, and judge the accuracy and consistency of the detection; S5) Feedback the test and detection results to the system model, update the system's maintenance log and equipment status, and generate maintenance recommendations and repair plans.

2. The aviation equipment detection and maintenance method based on the target detection model and MBSETest according to claim 1, characterized in that: The aviation equipment in step S1 is the fuselage, engine or landing gear of the aircraft.

3. The aviation equipment detection and maintenance method based on target detection model and MBSETest as claimed in claim 1, characterized in that: The target detection model in step S2 is a YOLO, Faster-R-CNN or ResNet deep learning model.

4. The aviation equipment detection and maintenance method based on target detection model and MBSETest as claimed in claim 1, characterized in that: The step S2 collects images and video data of aviation equipment in real time through a high-definition camera or infrared thermal imaging equipment, and pre-processes the collected images; then uses the target detection model to analyze the pre-processed images, detects and annotates faults or anomalies in the images; The step S3 imports the detection results into the MBSE system, associates the detection results with the system model, and generates test cases related to the fault or anomaly.

5. The aviation equipment detection and maintenance method based on target detection model and MBSETest as claimed in claim 4, characterized in that: The step S2 pre-processes the collected image, including denoising, contrast enhancement, brightness adjustment, edge sharpening detection, image scaling and format conversion.

6. The aviation equipment detection and maintenance method based on target detection model and MBSETest as claimed in claim 4, characterized in that: The faults or anomalies detected and marked by the target detection model in step S2 include surface cracks, wear or loose bolts, and output the fault type, fault location and severity information.

7. The aviation equipment detection and maintenance method based on target detection model and MBSETest as claimed in claim 1, characterized in that: When detecting cracks in aircraft engine blades, a high-definition camera is used to capture high-resolution images of the engine. After image preprocessing, the images are input into the YOLOv8 target detection module, which identifies and locates cracks on the blade surface. The detection results are imported into the MBSE system model to automatically generate test cases related to crack detection. By running these test cases in the MBSETest environment, it is verified whether the severity of the crack exceeds the system safety standard; If the test results indicate a risk, a maintenance plan is generated and immediate maintenance is recommended.

Images