Intelligent borescope detection system for aero-engine

Abstract

According to the intelligent borescope detection system for the aero-engine, in the system, an image acquisition module is arranged in a shell and used for acquiring image data of blade damage of the aero-engine; the image processing module is arranged in the shell and is connected with the image acquisition module so as to perform model reasoning and judge the type of blade damage based on the image data to obtain a detection result and a state; the display module is arranged on the shell and is connected with the image processing module to display the detection result and state in real time; the power supply module is arranged in the shell and is connected with and supplies power to the image processing module, the image acquisition module and the display equipment; the detection interface realizes real-time borescope inspection of the aero-engine in a man-machine cooperative interaction manner, and the specific functions comprise checking, storing, photographing, video recording, detecting and the like; and the detection unit is mainly used for reasoning and identifying the image data of the damage of the aero-engine blade acquired by the image acquisition module.

Description

Technical Field

[0001] The present invention relates to the technical field of aero-engine borescope technology, and particularly relates to an intelligent aero-engine borescope detection system. Background Art

[0002] As the heart of an aircraft, an aero-engine directly affects the flight performance and reliability of the aircraft and is the key to ensuring the safe execution of aviation flight missions. As the component with the largest number in an aero-engine, the blades work under harsh and complex conditions for a long time, and the probability of failure is extremely high. However, the internal space of the engine is narrow, and it is impossible for humans to directly observe the damage of these blades with the naked eye. Once not discovered and processed in time and maintained, it will not only reduce the performance of the engine and threaten flight safety, but more seriously, it may even cause air crash accidents and result in a large number of casualties. Therefore, it is necessary to detect various damage problems of engine blades in time, reduce safety risks, and ensure aviation flight safety.

[0003] Currently, the detection of internal blade damage in aero-engines mainly relies on borescope detection. However, the current borescope inspection mainly depends on manual operation of the borescope for visual inspection. The detection results are easily affected by the subjective judgment criteria of the inspectors, resulting in difficulties in ensuring the detection efficiency and accuracy. At the same time, since borescope detection has high requirements for the professional skills and experience of inspectors, and there are few experienced experts, the aero-engine borescope detection faces typical off-site detection problems, and it is impossible to guarantee the time cost and timeliness.

[0004] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The present invention provides an intelligent aero-engine borescope detection system, which can perform real-time, intelligent and accurate detection on the surface damage of aero-engine blades, further improve the detection efficiency, reduce the inaccuracy of manual visual inspection by inspectors, thereby improving the detection accuracy and better maintaining the flight safety of aero-engines.

[0006] An intelligent aero-engine borescope detection system includes

[0007] a housing

[0008] an image acquisition module, which is arranged in the housing to obtain image data of the damage of aero-engine blades;

[0009] an image processing module, which is arranged in the housing and connected to the image acquisition module. Among them, the image processing module includes a core board as the hardware platform of the image processing module and a baseboard for system power supply, power management and communication;

[0010] A display module, which is arranged on the housing and is used for displaying the detection results and status in real time;

[0011] A power supply module, which is arranged in the housing and connects and supplies power to the image processing module, the image acquisition module and the display device;

[0012] A GUI detection interface, which is used for realizing real-time borescope inspection of aero-engines through human-machine collaborative interaction, as well as viewing, saving, taking pictures, recording videos and detecting;

[0013] A detection unit, which is coupled to the GUI detection interface. Among them, the detection unit performs inference and recognition based on the image data of the damage of the aero-engine blade collected by the image acquisition module and cooperates with the human-machine collaborative interaction of the GUI detection interface.

[0014] In the described intelligent borescope detection system for aero-engines, the housing includes an upper screen housing, a lower screen housing, an upper handle housing, a lower handle housing and a handle bottom housing. The upper handle housing and the lower handle housing cooperate and enclose via the handle bottom housing, and the upper screen housing and the lower screen housing cooperate to assemble the display module.

[0015] In the described intelligent borescope detection system for aero-engines, the power supply module, the image processing module, the image acquisition module and the detection software are arranged between the upper handle housing and the lower handle housing.

[0016] In the described intelligent borescope detection system for aero-engines, the image acquisition module includes,

[0017] An endoscope hose, which is used to penetrate into the interior of the aero-engine,

[0018] A camera module, which is connected to the front end of the endoscope hose and is used to acquire image data of the aero-engine blade frame by frame in real time, and the acquired picture data is input into the image processing module, preprocessed and then input into the detection model;

[0019] A mechanical rocker, which is connected to and controls the attitude of the camera module on the endoscope hose, and is adjusted according to the position and detection angle of the part to be detected inside the aero-engine to capture the image data of the aero-engine damage.

[0020] In the described intelligent borescope detection system for aero-engines, the image processing module includes a core board as the hardware platform of the image processing module and a base board for system power supply, power management and communication. The core board is equipped with a detection unit to perform damage detection and damage type discrimination on the image data; the base board includes a display screen output module, a USB module, a network communication module, a storage module, a custom function button and a fan.

[0021] In the described intelligent borescope detection system for aero - engines, the display output module includes three display interfaces: a mainstream HDMI interface, a MIPI DSI interface, and a full - function Type - C interface. The mainstream HDMI interface is connected to a display device based on the TMDS transmission protocol for screen mirroring and demonstration; the MIPI DSI interface is used to connect to a screen; the full - function Type - C interface is used for update and iterative debugging and simultaneously supports the DP protocol for connecting to a one - cable - through Type - C screen.

[0022] In the described intelligent borescope detection system for aero - engines, the detection unit includes an input end, a backbone network, and a head network; the input end scales the image data collected by the image acquisition module to the input size of the model, fills in black borders simultaneously, and performs a normalization operation; the backbone network includes CBS modules, C3 modules, and SPPF modules. Each CBS module consists of a 1×1 convolutional layer, a batch normalization layer, and a SiLU activation function, and CBS performs downsampling through 3×3 or 6×6 stride convolutions. Each time the feature map passes through a CBS module, the size of the feature map is halved, and then the C3 module further extracts features. The feature map output by the backbone network is first output to the SPFF module to increase the size of the receptive field; the head network includes a neck network and a Detect module. The neck network adopts the FPN + PAN structure and is used to fuse the features extracted by the backbone network to enhance the feature expression ability of the model; the Detect module consists of three 1×1 convolutional layers and predicts the features output by the neck network.

[0023] In the described intelligent borescope detection system for aero - engines, the detection unit annotates the image data of the damaged blades of the aero - engine, constructs a blade damage dataset, and the damage categories cover four damage types: chipping, surface damage, breakage, and crack. At the same time, the size of the input damaged image for training the detection model is set to 640*640, the training batch size (batch - size) is 16, the initial number of training iteration epochs is 100, the learning rate is 0.01, and the best model weight parameter file, the training fitting curve, and the training detection results of the first three batches of damaged images will be saved after the training iteration ends. If the training fitting curve does not converge, the number of training iteration epochs will be increased and training will be restarted until convergence.

[0024] Compared with the prior art, the present invention has the following advantages: The present invention realizes real - time, accurate, and intelligent detection of damaged blades of aero - engines; and integrates the system into an integrated unit, forming a handheld device that is convenient to carry, convenient to detect, and convenient to operate, thereby improving the accuracy and efficiency of detection, which has important practical significance for maintaining the flight safety of aero - engines and promoting the automation process of borescopes. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] By reading the detailed description in the following preferred specific embodiments, various other advantages and benefits of the present invention will become clear to those of ordinary skill in the art. The accompanying drawings of the specification are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts. Moreover, throughout the drawings, the same reference numerals are used to represent the same components.

[0026] In the drawings:

[0027] Figure 1 is a schematic structural framework diagram of the intelligent borescope detection system for an aero-engine of the present invention;

[0028] Figure 2 is a model framework diagram of the borescope intelligent detection unit for an aero-engine of the present invention;

[0029] Figure 3 is a functional diagram of the detection software interface in an embodiment of the present invention;

[0030] Figure 4 is a program interface diagram of the detection software interface in an embodiment of the present invention;

[0031] Figure 5 is an effect diagram of the detection result in this embodiment;

[0032] Figure 6 is a schematic diagram of the integrated housing structure in this embodiment.

[0033] The present invention will be further explained below in conjunction with the accompanying drawings and embodiments. Specific Embodiments

[0034] The specific embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the specific embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present invention can be more thoroughly understood and the scope of the present invention can be fully conveyed to those skilled in the art.

[0035] It should be noted that in the description and claims, certain terms are used to refer to specific components. Those skilled in the art should understand that technicians may use different terms to refer to the same component. The description and claims in this specification do not distinguish components based on the differences in terms, but rather on the functional differences of the components. For example, the terms "comprising" or "including" mentioned throughout the specification and claims are open-ended terms and should be interpreted as "including but not limited to". The subsequent description in the specification is for the purpose of describing the preferred embodiments of implementing the present invention, but the description is based on the general principles of the specification and is not used to limit the scope of the present invention. The protection scope of the present invention shall be subject to that defined by the appended claims.

[0036] For the convenience of understanding the embodiments of the present invention, the following will further explain with specific embodiments in conjunction with the drawings, and each drawing does not constitute a limitation on the embodiments of the present invention.

[0037] As Figures 1 to 6 shown, the intelligent borescope detection system for aeroengines includes two parts: hardware and software:

[0038] The hardware part includes a housing,

[0039] an image acquisition module, which is arranged in the housing to obtain image data of the damage of aeroengine blades;

[0040] an image processing module, which is arranged in the housing and connected to the image acquisition module. Among them, the image processing module includes a core board as the hardware platform of the image processing module and a baseboard for system power supply, power management, and communication;

[0041] a display module, which is arranged on the housing and used to display the detection results and status in real time;

[0042] a power module, which is arranged in the housing and connected to and supplies power to the image processing module, the image acquisition module, and the display device.

[0043] The software part includes a GUI detection interface and a detection unit. Among them, the detection interface is used to realize the real-time borescope inspection of aeroengines through human-machine collaborative interaction. The specific functions include functions such as viewing, saving, taking pictures, recording videos, and detecting;

[0044] The detection unit is coupled to the GUI detection interface. Among them, the detection unit is used to perform reasoning and recognition based on the image data of the damage of aeroengine blades collected by the image acquisition module and cooperate with the human-machine collaborative interaction of the GUI detection interface.

[0045] In a preferred embodiment of the described intelligent borescope detection system for aero-engines, the housing includes an upper screen housing, a lower screen housing, an upper handle housing, a lower handle housing, and a bottom handle housing. The upper handle housing and the lower handle housing are fitted together and enclosed by the bottom handle housing, and the upper screen housing and the lower screen housing are fitted together to assemble a display module.

[0046] In a preferred embodiment of the described intelligent borescope detection system for aero-engines, the power module, the image processing module, the image acquisition module, and the detection software are arranged between the upper handle housing and the lower handle housing.

[0047] In a preferred embodiment of the described intelligent borescope detection system for aero-engines, the image acquisition module includes

[0048] an endoscope hose for penetrating deep into the aero-engine

[0049] a camera module connected to the front end of the endoscope hose for real-time acquisition of image data of the aero-engine blades;

[0050] a mechanical rocker connected to and controlling the posture of the camera module on the endoscope hose, and adjusted according to the position and detection angle of the part to be detected inside the aero-engine to capture image data of the aero-engine damage.

[0051] In a preferred embodiment of the described intelligent borescope detection system for aero-engines, the core board is equipped with a detection unit to perform damage detection and damage type discrimination on the image data.

[0052] In a preferred embodiment of the described intelligent borescope detection system for aero-engines, the bottom board includes a display screen output module, a USB module, a network communication module, a storage module, custom function buttons, and a fan.

[0053] In a preferred embodiment of the described intelligent borescope detection system for aero-engines, the display screen output module includes three display interfaces: a mainstream HDMI interface, a MIPI DSI interface, and a full-function Type-C interface. The mainstream HDMI interface is connected to a display device based on the TMDS transmission protocol for screen mirroring demonstration; the MIPI DSI interface is used to connect to the screen; the full-function Type-C interface is used for update and iterative debugging and simultaneously supports the DP protocol.

[0054] In a preferred embodiment of the described intelligent borescope detection system for aero-engines, the full-function Type-C interface is used for connecting a one-line-through Type-C screen.

[0055] The detection unit model structure consists of an input, a backbone network, and a head network. The input end scales the image data collected by the image acquisition module to the input size of the model, fills in black borders, and performs a normalization operation at the same time; the backbone network includes CBS modules, C3 modules, and SPPF modules; each CBS module consists of a 1x1 convolutional layer, a batch normalization layer, and a SiLU activation function, and CBS performs downsampling through 3x3 or 6x6 stride convolutions; every time the feature map passes through a CBS module, the size of the feature map is halved, and then the C3 module further extracts the features; the feature map output by the backbone network is first output to the SPFF module, which can increase the size of the receptive field and enable the model to better adapt to targets of different sizes and shapes; the head network includes a neck network and a Detect module, and the neck network adopts an FPN+PAN structure, where: the neck network can fuse the features extracted by the backbone network and enhance the feature expression ability of the model; the Detect module, composed of three 1x1 convolutional layers, is mainly used to predict the features output by the neck network.

[0056] The detection unit is labeled based on the collected image data of aero-engine blade damage to construct a blade damage dataset, and the damage categories mainly cover four damage types: chipping, surface damage, breakage, and crack. At the same time, the size of the input damage image for the detection model training is set to 640*640, the training batch size is 16, the initial training iteration number epochs is 100, the learning rate is 0.01, and the best model weight parameter file, the training fitting curve, and the detection results of the first three batches of damage images will be saved after the training iteration ends. If the training fitting curve does not converge, the number of training iterations needs to be increased and retraining is required.

[0057] In one embodiment, the aero-engine intelligent borescope detection system includes

[0058] The image acquisition module is used to obtain blade damage image data. The basic structure of the module includes a camera module, an endoscope hose and a mechanical rocker. The module communicates with the USB module of the image processing module base plate to transmit the collected image data of the damage of the aircraft engine blade to the image processing module of the system. The camera module has a dimming function, which can adjust the light according to the internal environment of the aircraft engine. The mechanical rocker can control the posture of the camera module at the front end of the endoscope hose. With the endoscope hose, it can adapt to the complex structure inside the aircraft engine and go deep into the damaged part to be detected. At the same time, with the camera module, it can realize image acquisition from multiple perspectives, making the detection more comprehensive and detailed. Exemplarily, the diameter of the endoscope hose is divided into two specifications, such as 6mm and 4mm diameters, to meet the requirements of the aircraft engine peephole. In the present invention, the intelligent detection of the aircraft engine borescope requires real-time performance, and the detection unit is used to realize real-time processing of each frame of the image and display the detection results. It can be understood that this has certain requirements for computing power. Therefore, exemplary, the image processing module selects a solution combining the RK3588S core board and the base plate. RK3588S uses an eight-core 64-bit processor, a quad-core GPU, and an NPU computing power of up to 6Tops. It supports 8K video decoding and has excellent image processing capabilities. Using the RK3588S core board as the hardware platform of the image processing module can well meet the performance requirements of the intelligent detection of aircraft engine borescopes. In the image processing module, the baseboard is used for the system power module to supply power and power management to the various components of the system, and it is also a bridge for the display module to communicate with the image acquisition module and the image processing module. The circuit of the baseboard mainly includes a display output module, a USB module, a network communication module, a storage module, a function button, and a fan. The display output module contains three display interfaces: the mainstream HDMI interface, based on the TMDS transmission protocol, can be connected to the display device, and is used for the software projection screen demonstration of the aircraft engine borescope intelligent detection instrument; the MIPI DSI interface is used for the connection between the aircraft engine borescope intelligent detection instrument and the system display module screen; the full-function Type C interface is mainly used for the update and iteration debugging of the aircraft engine intelligent detection unit, and supports the DP protocol, and can also be used for the connection of the Type C screen.

[0059] In addition, the USB module may include USB3.0, USB2.0, and the above-mentioned Type C. Exemplarily, USB2.0 includes two channels, one of which is used for connecting the system camera, and can transmit the collected image data to the image processing module through the USB communication protocol, which adopts the XH2.54 interface form. USB3.0 is used for system external debugging and expansion, and can develop, debug and maintain the aircraft engine intelligent boresight instrument and system by connecting peripherals such as mouse and keyboard. In addition, it can also connect to a U disk device to transfer the saved photos and videos of the aircraft engine blade damage detection results.

[0060] The network communication module includes an Ethernet interface and a WIFI module. The storage module uses an SD card to store images and video data of blade damage detected by the intelligent borescope equipment of aircraft engines. The custom function buttons include photo taking, video recording, power switch, dimming, stop / end, and detection functions. When performing borescope detection, if the system detects damage, the inspector can use the photo taking and video recording buttons to take photos or record the images of damaged blades detected in real time; the power switch controls the power supply of the entire system; the dimming button supports the inspector to control the LED light of the front lens of the camera module according to the internal lighting conditions of the aircraft engine to collect clearer damage images and avoid missed detection and false detection. The stop / end button can realize the pause and end operation of the detection result recording, and also realize the shutdown of the system detection status. The detection button can realize the call of the intelligent borescope detection instrument of the aircraft engine and the camera module of the system image acquisition module with one click, realize real-time detection of damage, and display the results on the screen.

[0061] Since aircraft engine borescope inspection is often not conducted indoors, it will be affected by high temperatures. At the same time, the image processing module will also generate a certain amount of heat when processing images. If the heat is not dissipated, it is likely to cause the system to crash due to high temperature and the processor performance to decline. Therefore, fans must be used to dissipate heat to maintain the service life of aircraft engine borescope intelligent inspection instruments and systems.

[0062] In addition, the power module can use a 12V DC battery to power the image processing module, image acquisition module camera module and display device in the aircraft engine intelligent boresight detection instrument and system. The battery capacity is 3400mAh, which can ensure that the aircraft engine boresight system can work continuously for more than 2 hours in the detection state, meeting the requirements of field engine boresight detection. The display module adopts a 1080P 7-inch capacitive touch screen in consideration of the size of the aircraft engine intelligent boresight detection instrument and system. In order to facilitate assembly of the system integration, the touch screen display mode selects the MIPI display solution with the simplest wiring in the baseboard and can fully meet the volume requirements of the assembly.

[0063] Figure 3 This is a functional diagram of the detection software interface of the system of the present invention. For the convenience and interactivity of the operation of the system, a software application is encapsulated by QT software to implement the detection unit and GUI detection interface. Among them, the functions of the GUI detection interface cover the detection of aircraft engine blade damage, taking pictures, saving and recording blade damage detection results, and viewing and playing pictures and videos of blade damage detection results. When the detection personnel perform borescope detection operations, they can click the corresponding buttons and select the appropriate functions to further improve the detection efficiency.

[0064] Figure 6 The shown housing includes an upper screen housing, a lower screen housing, an upper handle housing, a lower handle housing and a handle bottom housing. For the operation of aero-engine field borescope inspection, one hand needs to hold the camera, and the other hand needs to control the joystick to control the camera. At the same time, the operator also needs to closely watch the screen. The design of the handle is convenient for the inspectors to operate the mechanical joystick on one hand, and on the other hand, it is also convenient for the inspectors to hold the borescope device with one hand while closely watching the screen and operating the camera coordinately to jointly complete the inspection work. The lower screen housing is mainly used to place the image processing module circuit board and cooperate with the upper screen housing to install the screen. The handle bottom housing is used as the base for the battery and is provided with a charging interface to charge the system when the power of the borescope device is low.

[0065] Exemplarily, the right side of the lower screen housing is provided with slots for expanding USB, HDMI and SD cards. The bottom is provided with a battery charging port to charge the DC battery.

[0066] Although the embodiments of the present invention have been described above in conjunction with the accompanying drawings, the present invention is not limited to the above specific embodiments and application fields. The above specific embodiments are merely illustrative and guiding, rather than restrictive. Those of ordinary skill in the art can also make many forms under the inspiration of this specification and without departing from the scope protected by the claims of the present invention, and all of these fall within the scope of protection of the present invention.

Claims

1. An intelligent borescope detection system for aircraft engines, characterized in that: These include, case, An image acquisition module, which is arranged in the housing and used to acquire image data of damage to the aeroengine blade; An image processing module, which is arranged in the housing and connected to the image acquisition module, wherein the image processing module includes a core board as a hardware platform of the image processing module and a baseboard for system power supply, power management and communication; A display module, which is arranged on the housing and is used to display the detection result and status in real time; A power module, which is disposed in the housing and is connected to and supplies power to the image processing module, the image acquisition module and the display device; GUI inspection interface, which is used for human-machine collaborative interaction to achieve real-time borescope inspection of aircraft engines, as well as viewing, saving, taking photos, recording and testing; A detection unit is coupled to the GUI detection interface, wherein the detection unit is used to perform reasoning and recognition based on the image data of the damaged aircraft engine blades collected by the image acquisition module and cooperate with the human-computer collaborative interaction of the GUI detection interface.

2. The aircraft engine intelligent borescope detection system according to claim 1, characterized in that: Preferably, the shell includes a screen upper shell, a screen lower shell, a handle upper shell, a handle lower shell and a handle bottom shell. The handle upper shell and the handle lower shell cooperate and are enclosed by the handle bottom shell. The screen upper shell cooperates with the screen lower shell to assemble the display module.

3. The aircraft engine intelligent borescope detection system according to claim 2, characterized in that: The power module, the image processing module, the image acquisition module and the detection software are arranged between the handle upper shell and the handle lower shell.

4. The aircraft engine intelligent borescope detection system according to claim 1, characterized in that: The image acquisition module comprises: Endoscope hose, which is used to penetrate into the interior of aircraft engines. A camera module connected to the front end of the endoscope hose for acquiring image data of the aeroengine blades frame by frame in real time, and the acquired image data is input into the image processing module for preprocessing and then input into the detection model; A mechanical rocker is connected to and controls the posture of the camera module on the endoscope hose, and is adjusted according to the position and detection angle of the part to be detected inside the aircraft engine to capture image data of aircraft engine damage.

5. The aircraft engine intelligent borescope detection system according to claim 1, characterized in that: The core board is equipped with the detection unit to perform damage detection and damage type identification on the image data; the base board includes a display output module, a USB module, a network communication module, a storage module, a custom function button and a fan.

6. The aircraft engine intelligent borescope detection system according to claim 5, characterized in that: The display output module includes three display interfaces: mainstream HDMI interface, MIPI DSI interface and full-function Type C interface. The mainstream HDMI interface is based on the TMDS transmission protocol to connect the display device for screen projection demonstration; The MIPI DSI interface is used to connect to the screen; the full-function Type C interface is used for update and iteration debugging and also supports the DP protocol for connecting to a Type C screen.

7. The aircraft engine intelligent borescope detection system according to claim 1, characterized in that: The detection unit includes an input end, a backbone network and a head network, and the input end is an image acquisition module.

8. The intelligent borescope detection system for aircraft engines according to claim 7, characterized in that: The detection unit annotates the collected image data of aircraft engine blade damage and constructs a blade damage dataset. The damage categories include four types of damage: chipping, surface damage, breakage and cracks.

Images