A non-contact detection method for ultra-high cycle fatigue damage

Abstract

The application belongs to the technical field of structural strength test detection, and is a kind of non-contact detection method for ultrahigh cycle fatigue damage, binocular vision system and infrared thermal imaging system are respectively built to carry out non-contact detection of ultrahigh cycle fatigue damage, first, the displacement of the test piece point of the test piece is tested by the laser displacement sensor, and is converted into the field displacement / strain data of the vision system, the signal output by the laser displacement sensor is read by means of the synchronous trigger device, then the binocular camera collects the test piece cycle phase photos under different phases in several cycles, and then the thermal imaging signal sampling of the infrared thermal imager is carried out, the corresponding relationship between the vision image displacement / strain data and the thermal imaging signal is established, and the determination of the starting moment of the test piece damage and the quantitative characterization of the damage are completed. The data acquisition amount is effectively reduced, the long-time monitoring and data acquisition of the ultrahigh cycle fatigue test can be completed, and the determination of the starting moment of the ultrahigh cycle fatigue test damage and the quantitative characterization of the damage evolution are realized.

Description

Technical Field

[0001] This application belongs to the field of structural strength testing technology, and specifically relates to a non-contact detection method for ultra-high cycle fatigue damage. Background Technology

[0002] Aero engines operate under harsh environments characterized by high temperature, high pressure, high speed, and strong vibration. The ultra-high cycle fatigue induced by these severe conditions can lead to engine failure, posing a significant safety hazard. In actual operation, critical components of aero engines, including turbine blades, compressors, and bladed disks, are simultaneously subjected to large-amplitude, low-frequency centrifugal loads and small-amplitude, high-frequency vibration loads. The high-temperature alloys used in these components exhibit typical ultra-high cycle fatigue characteristics, meaning that even under relatively low stress levels of fatigue cyclic loading, cracks appear and eventually fracture occurs when the fatigue life exceeds 10⁷ cycles. Improving the ultra-high cycle fatigue performance of engines to enhance their reliability and extend their safe lifespan has garnered widespread attention. Ultra-high cycle fatigue testing is a crucial means of performance evaluation, and early damage detection is essential for understanding the initiation and propagation mechanisms of ultra-high cycle fatigue cracks. Therefore, early damage detection is a critical issue that urgently needs to be addressed in ultra-high cycle fatigue testing.

[0003] Ultra-high cycle fatigue crack initiation life accounts for more than 90% of the total lifespan. Traditional detection methods such as ultrasound, X-ray, and eddy current are not very sensitive to the early initiation of fatigue cracks and it is difficult to obtain the process state of ultra-high cycle fatigue damage generation and evolution. Therefore, it is impossible to achieve early detection of damage. Non-contact detection methods are needed to determine the initiation and propagation of damage.

[0004] Current visual inspection mainly relies on DIC testing for displacement or strain, or infrared thermal imaging for temperature data acquisition and analysis. The main problems are that ultra-high cycle accelerated fatigue testing has a high frequency, large data volume for high-frequency signal acquisition, and high computational cost, making long-term online detection impossible. It is also inaccurate in determining the damage initiation time and cannot achieve quantitative characterization of damage evolution.

[0005] Therefore, how to achieve non-contact testing for ultra-high cycle fatigue is a problem that needs to be solved. Summary of the Invention

[0006] The purpose of this application is to provide a non-contact detection method for ultra-high cycle fatigue damage, in order to solve the problems that existing ultra-high cycle accelerated fatigue tests have high frequency, large amount of high-frequency signal acquisition data, and high computational cost, making it difficult to achieve non-contact detection of ultra-high cycle fatigue tests.

[0007] The technical solution of this application is: a non-contact detection method for ultra-high cycle fatigue damage, comprising:

[0008] A binocular vision system and an infrared thermal imaging system were set up separately. The binocular vision system was controlled to acquire synchronous displacement change signals and visual images of the test piece, and the infrared thermal imaging system was controlled to acquire infrared images of the test piece. The binocular vision system includes a laser displacement sensor and a binocular camera, and the infrared thermal imaging system includes an infrared thermal imager.

[0009] The binocular vision system and infrared thermal imaging system are assembled at a certain position on the test piece. Feature points are sprayed on the surface of the test piece. The feature points of the test piece are detected and adjusted by the binocular camera and infrared thermal imager until both the binocular camera and infrared thermal imager can acquire high-resolution data to determine the current position and distance of the binocular vision system and infrared thermal imaging system.

[0010] A laser displacement sensor is installed at the bottom of the test piece. Ultrasonic excitation is applied to the test piece. The laser displacement sensor measures the point displacement of the test piece from the bottom of the test piece and outputs the point displacement signal of the test piece to the software system of the binocular vision system.

[0011] The software system of the binocular vision system converts the point displacement signal of the test specimen into field displacement / strain data of the vision system;

[0012] Ultra-high cycle fatigue accelerated tensile-compression test was conducted to capture high-frequency signals in the field displacement / strain data of the vision system, obtain excitation signals of different phases at high frequency, until a complete periodic signal was formed;

[0013] The phase-locked loop signal processing method is used to multiply the high-frequency complete periodic signal with the reference signal, and then filter it through an adjustable low-pass filter to obtain the low-frequency signal;

[0014] After reading the low-frequency signal output by the laser displacement sensor, the binocular camera acquires periodic phase photos of the test piece under different phases within several cycles through super-cycle sampling.

[0015] The software system of the binocular vision camera acquires the average displacement and strain change of each reference point in the periodic camera image, obtains visual image displacement / strain data, compares it with the simulated displacement signal collected by the laser displacement sensor, and judges whether the corresponding parameters of the visual image displacement / strain data and the field displacement / strain data of the vision system are within the set range. If they are, the sampling is determined to be accurate. If not, the position of the binocular vision system is adjusted and measured again until the set range requirements are met.

[0016] When the stress difference of the sampling points within the field of view of the binocular camera reaches the limit value, the excitation signal is input to the infrared thermal imaging system. The infrared thermal imager uses infrared thermal imaging signals on the test piece. The software system of the infrared thermal imaging system analyzes the infrared thermal imaging signals to determine the damage initiation time and obtain the second initiation and evolution data of the damage of the test piece.

[0017] The software system of the binocular vision system determines the first initiation and evolution data of the damage to the test specimen through the field displacement / strain data of the vision system, calculates the displacement of the test specimen through the displacement relationship of feature points, and establishes the correspondence between temperature signal and strain / displacement signal by combining the second initiation and evolution data of the infrared thermal imaging system, thus completing the determination of the damage initiation time of the test specimen and the quantitative characterization of the damage.

[0018] Preferably, the hardware system of the binocular vision system further includes a console, a synchronization triggering device, a light source system, a camera movement track, and a vision system frame; the binocular camera and the light source system are both mounted on the vision system frame, the laser displacement sensor is electrically connected to the synchronization triggering device, the synchronization triggering device is electrically connected to the console, and the vision system frame is mounted on the camera movement track, allowing the vision system frame to move horizontally on the camera movement track; the software system of the binocular vision system includes an acquisition system, a calibration system, and a full-field strain analysis system; the acquisition system can receive detection data acquired by the laser displacement sensor and the binocular camera, and send it to the calibration system and the full-field strain analysis system; the calibration system can determine the sharpness based on the detection data, and if the sharpness is insufficient, adjust the position of the binocular vision system and re-detect until the calibration system determines that it can receive detection data of a specified resolution; the full-field strain analysis system can perform full-field strain analysis on the received detection data after calibration to obtain the first initial and evolution data;

[0019] The hardware system of the infrared thermal imaging system also includes a camera bracket and a computer module. The infrared thermal imager is fixedly connected to the camera bracket and electrically connected to the computer module. The software system of the infrared thermal imager is located within the computer module. The software system of the infrared thermal imager includes an acquisition and analysis module and an image processing module. The acquisition and analysis module can receive the infrared thermal imaging signals acquired by the infrared thermal imager and analyze them to obtain image data of feature points. The image processing module can analyze and process the image data of feature points to obtain the second initiation and evolution data of the test piece damage.

[0020] Preferably, the high-frequency signal emitted by the laser sensor is captured by a synchronous triggering device. The specific method is as follows: a step size is set, a reference low-frequency signal is given first, the synchronous triggering device captures the reference low-frequency signal, and after the capture is completed, the frequency of the reference low-frequency signal is gradually increased according to the step size. Each time the frequency is increased, the signal is captured once, until the high-frequency signal of the specified frequency is captured.

[0021] Preferably, the method for performing super-periodic sampling with the binocular camera is as follows: setting a specific phase, automatically acquiring photos within the specific phase, and then reconstructing the photos of different specific phases within several periods to form a complete periodic phase photo.

[0022] Preferably, irregular feature points are formed on the surface of the test piece by spraying a patchy matte paint, and feature points are collected by selecting one feature point at each of the three positions of the test piece (top, middle, and bottom).

[0023] Preferably, the method for obtaining the visual image displacement / strain data is as follows: images of the ROI regions of three feature points are acquired by a binocular camera, and the average displacement and strain change in the tensile loading direction of the three ROI regions of the test piece are analyzed and calculated to obtain the visual image displacement / strain data.

[0024] Preferably, the method for determining the damage initiation time of the infrared thermal imaging signal is as follows: first, the infrared thermal imaging signal is subjected to five-point cubic smoothing, and then the initial time of the average temperature rise is determined by taking the derivative of the fitting function based on the least squares method. This time is the damage initiation position. The second initiation and evolution data are selected based on the energy dissipation method to calculate the average temperature rise of the infrared thermal imaging signal, and then the data is processed by the Gaussian filtering method. Finally, the energy dissipation is obtained by combining the two-dimensional energy dissipation method, and the changes in temperature dissipation at different times are analyzed and determined to obtain the second initiation and evolution data.

[0025] Preferably, the method for obtaining the first initiation and evolution data is as follows: acquiring periodic phase photos of the test specimen captured by a binocular camera, first performing grayscale and brightness processing, then performing feature matching based on SIFT or SURF algorithms, finally calculating the displacement of the test specimen through the displacement relationship of pixel points, calculating the strain of the test specimen through differential calculation, and finally obtaining the state of the test specimen at different times to form visual system field displacement / strain data; setting a stress change threshold, the position in the periodic phase photo of the test specimen that meets the stress change threshold is the damage initiation time, and then obtaining the visual system field displacement / strain data after the damage initiation time to obtain the first initiation and evolution data.

[0026] This application presents a non-contact detection method for ultra-high cycle fatigue damage. It employs both a binocular vision system and an infrared thermal imaging system for non-contact detection. First, a laser displacement sensor captures the point displacement of the test specimen, converting it into field displacement / strain data for the vision system. A synchronous triggering device reads the signal output from the mechanical displacement sensor. Then, a binocular camera acquires periodic phase images of the test specimen at different phases within several cycles. Finally, the thermal imaging signal is sampled by an infrared thermal imager. By establishing a correspondence between the visual image displacement / strain data and the thermal imaging signal, the method determines the damage initiation time and quantifies the damage. This method effectively reduces the amount of data acquisition without compromising accuracy, enabling long-term monitoring and data acquisition for ultra-high cycle fatigue tests. It achieves the determination of the damage initiation time and the quantitative characterization of damage evolution in ultra-high cycle fatigue tests, providing an effective technical means for non-contact detection of ultra-high cycle fatigue tests. Attached Figure Description

[0027] To more clearly illustrate the technical solutions provided in this application, the accompanying drawings will be briefly described below. Obviously, the drawings described below are merely some embodiments of this application.

[0028] Figure 1 This is a schematic diagram of the overall process of this application;

[0029] Figure 2 This is a schematic diagram of the hardware device structure for the present application;

[0030] Figure 3 The graph shows the change curve of the displacement sensor signal and the strain change curve of the test piece obtained by the vision system in this application. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] A non-contact detection method for ultra-high cycle fatigue damage is disclosed, wherein the ultra-high cycle fatigue test is an accelerated ultra-high cycle tensile-compressive fatigue test based on ultrasonic loading. The test employs an ultrasonic axial tensile-compressive fatigue testing system to achieve high-frequency excitation of the round bar specimen. This system includes an ultrasonic signal generator, transducer, amplitude modulator, and amplitude transformer, etc., with a test loading frequency of 20 kHz. The specimen end diameter is 10 mm, the gauge length is 3 mm, and the gauge diameter is 3 mm. The specimen is bolted to the amplitude transformer via pin holes. When the stress ratio R = -1, the lower end of the specimen is used as a free end for excitation.

[0033] like Figure 1 As shown, it includes the following steps:

[0034] Step S100: Set up a binocular vision system and an infrared thermal imaging system respectively. Control the binocular vision system to acquire synchronous displacement change signals and visual images of the test piece, and control the infrared thermal imaging system to acquire infrared images of the test piece. The binocular vision system includes a laser displacement sensor and a binocular camera, and the infrared thermal imaging system includes an infrared thermal imager.

[0035] Both the binocular vision system and the infrared thermal imaging system include hardware and software systems. The hardware system of the binocular vision system further includes a control console, a synchronization triggering device, a light source system, a camera movement track, and a vision system frame. The binocular camera and light source system are both mounted on the vision system frame. The laser displacement sensor is electrically connected to the synchronization triggering device, which is electrically connected to the control console. The vision system frame is mounted on the camera movement track and can move horizontally along the track. The software system of the binocular vision system includes an acquisition system, a calibration system, and a full-field strain analysis system. The acquisition system receives detection data from the laser displacement sensor and the binocular camera and sends it to the calibration system and the full-field strain analysis system. The calibration system judges the sharpness based on the detection data; if the sharpness is insufficient, it adjusts the position of the binocular vision system and re-detects until the calibration system determines that it can receive detection data at a specified resolution. The full-field strain analysis system performs full-field strain analysis on the received detection data after calibration to obtain initial and evolutionary data. The synchronization triggering device is used to trigger data acquisition from the binocular camera and the infrared thermal imaging system.

[0036] The hardware system of the infrared thermal imaging system also includes a camera bracket and a computer module. The infrared thermal imager is fixedly connected to the camera bracket and electrically connected to the computer module. The software system of the infrared thermal imager is located within the computer module. The software system of the infrared thermal imager includes an acquisition and analysis module and an image processing module. The acquisition and analysis module can receive the infrared thermal imaging signals acquired by the infrared thermal imager and analyze them to obtain image data of feature points. The image processing module can analyze and process the image data of feature points to obtain the second initiation and evolution data of the test piece damage.

[0037] Step S200: Assemble the binocular vision system and infrared thermal imaging system at a certain position on the test piece. The requirements for distance, field of view, and illumination must be met. The distance should be controlled within 1 meter, with good illumination conditions and the field of view should cover the entire test piece. Spray feature points onto the surface of the test piece. Specifically, irregular feature points are formed on the surface of the test piece by spraying a patchy matte paint. Feature points are collected by selecting one feature point at each of the three positions on the test piece (top, middle, and bottom), forming a total of three feature points for subsequent visual matching calculations.

[0038] The feature points of the test piece are detected and adjusted by using a binocular camera and an infrared thermal imager until both the binocular camera and the infrared thermal imager can acquire high-resolution data. The current position and distance of the binocular vision system and the infrared thermal imaging system are determined, so as to realize high-resolution acquisition of images and temperature data of the test piece under in-situ test conditions.

[0039] In step S300, the laser displacement sensor is installed on the bottom of the test piece. Since the testing machine cannot be directly connected to the synchronous triggering device through the interface, the displacement signal of the test piece collected by the laser displacement sensor is connected to the synchronous triggering device to determine whether to perform synchronous acquisition of image and temperature data.

[0040] The test piece is a cylindrical rod, thinner in the middle and thicker at both ends, with a smooth flat bottom. Based on the shape characteristics of the test piece itself, the LK-G5000 laser displacement sensor is placed at the bottom of the test piece to capture the displacement changes at the bottom of the test piece and output the displacement signal.

[0041] After debugging, the ultra-high cycle fatigue testing machine applies ultrasonic excitation to the test piece, measures the point displacement of the test piece from the bottom of the test piece through a laser displacement sensor, outputs the point displacement signal of the test piece to the synchronous triggering device, and then inputs the signal processed by the synchronous triggering device into the software system of the binocular vision system.

[0042] It is important to note that the external signal voltage received by the synchronization triggering device needs to be controlled within a certain range; in this system, the voltage must be less than 5V. The synchronization triggering hardware device includes... Figure 2 As shown.

[0043] In step S400, the software system of the binocular vision system converts the point displacement of the test specimen into field displacement / strain data of the vision system.

[0044] Step S500: Perform an ultra-high cycle fatigue accelerated tensile-compression test at a frequency of 20 kHz. Capture high-frequency signals from the field displacement / strain data of the vision system, obtain excitation signals of different phases at high frequency, until a complete periodic signal is formed.

[0045] When the frequency is high and the voltage is low, the synchronous triggering device has difficulty effectively capturing high-frequency signals. The high-frequency signal emitted by the laser sensor is captured by the synchronous triggering device. The specific method is as follows: set a step size, first give a reference low-frequency signal, the synchronous triggering device captures the reference low-frequency signal, after the capture is completed, the frequency of the reference low-frequency signal is gradually increased according to the step size, and the signal is captured once for each increase, until the high-frequency signal of the specified frequency is captured.

[0046] In step S600, the received high-frequency laser displacement signal is subjected to Fourier transform and Butterworth filtering. With the help of a synchronous triggering device, a phase-locked signal processing method is adopted to multiply the complete periodic signal based on the high frequency with the reference signal, and then filter it through an adjustable low-pass filter to obtain a low-frequency signal. The high-frequency signal is converted into a low-frequency signal, which is convenient for the digital-to-analog converter to process.

[0047] In step S700, after reading the low-frequency signal output by the laser displacement sensor, the binocular camera acquires periodic phase photos of the test piece under different phases within several cycles through super-cycle sampling.

[0048] The method of super-periodic sampling for binocular cameras is as follows: a specific phase is set, such as a peak or trough, and photos within the specific phase are automatically acquired. Then, the photos of different specific phases within several periods are reconstructed to form a complete periodic phase photo, thus realizing super-periodic sampling.

[0049] It enables the acquisition of high-frequency data from ultra-high cycle fatigue accelerated tests using ordinary industrial cameras, reducing the cost of image acquisition and the amount of image data stored, and effectively capturing the state changes of test pieces over a long period of time.

[0050] Step S800: The software system of the binocular vision camera acquires the average displacement and strain change of each reference point within the periodic camera image, obtaining visual image displacement / strain data. This data is compared with the simulated displacement signal acquired by the laser displacement sensor. The system determines whether the corresponding parameters of the visual image displacement / strain data and the vision system field displacement / strain data are within the set range. If yes, the sampling is considered accurate; otherwise, the position of the binocular vision system is adjusted and the measurement is repeated until the set range requirement is met. A comparative analysis of the two sampling methods is performed. If the corresponding parameters differ significantly, it indicates a large sampling error; otherwise, it indicates accurate sampling. The displacement sensor signal change curve is compared with the strain change curve of the test piece acquired by the vision system as follows: Figure 3 As shown.

[0051] Preferably, the method for obtaining visual image displacement / strain data is as follows: images of the ROI regions of three feature points are acquired by a binocular camera, and the average displacement and strain change in the tensile loading direction of the three ROI regions of the test piece are analyzed and calculated to obtain visual image displacement / strain data.

[0052] In step S900, when the stress difference of the sampling points within the field of view of the binocular camera reaches the limit value, it indicates that the test piece is at the initial moment of damage. The initial moment of damage also has obvious color change. Therefore, the initial moment of damage can be determined by color change or stress change.

[0053] When the stress at a certain point within the field of view shows a significant change compared to other parts, an excitation signal is input to the infrared thermal imaging system. The infrared thermal imager can accept TTL pulse signals and has a universal BNC interface, enabling synchronous acquisition by the binocular vision and infrared thermal imaging systems. The infrared thermal imager captures infrared thermal imaging signals of the test piece, and the software system of the infrared thermal imaging system analyzes these signals to determine the damage initiation time and acquire secondary initiation and evolution data of the damage to the test piece.

[0054] Damage is more likely to occur at locations with high stress. When significant stress concentration occurs, the infrared thermal imaging system is triggered, and visual signals are captured, providing multiple data support methods for determining the moment of damage initiation.

[0055] Preferably, the method for determining the damage initiation time of the infrared thermal imaging signal is as follows: first, the infrared thermal imaging signal is subjected to five-point cubic smoothing; then, the initial time of the average temperature rise is determined by differentiating the fitting function using the least squares method, and this time is the damage initiation location; the second initiation and evolution data are calculated based on the average temperature of the infrared thermal imaging signal using an energy dissipation method, then the data is processed using a Gaussian filtering method, and finally, the energy dissipation is obtained by combining a two-dimensional energy dissipation method, analyzing and determining the changes in temperature dissipation at different times to obtain the second initiation and evolution data. The highest failure temperature of the TC17 test piece obtained from the ultra-high cycle fatigue test is close to about 200 degrees Celsius.

[0056] Step S1000, binocular vision inspection mainly includes image acquisition, image matching, and deformation calculation. The software system of the binocular vision system determines the first initiation and evolution data of the test piece damage through the displacement / strain data of the vision system field, calculates the displacement of the test piece through the displacement relationship of feature points, and establishes the correspondence between temperature signals and strain / displacement signals by combining the second initiation and evolution data of the infrared thermal imaging system, thus completing the determination of the damage initiation time of the test piece and the quantitative characterization of the damage.

[0057] Preferably, the method for obtaining the first initiation and evolution data is as follows: acquiring periodic phase photos of the test specimen captured by a binocular camera, first performing grayscale and brightness processing, then performing feature matching based on SIFT or SURF algorithms, finally calculating the displacement of the test specimen through the displacement relationship of pixel points, calculating the strain of the test specimen through differential calculation, and finally obtaining the state of the test specimen at different times to form visual system field displacement / strain data; setting a stress change threshold, the position in the periodic phase photo of the test specimen that meets the stress change threshold is the damage initiation time, and then obtaining the visual system field displacement / strain data after the damage initiation time to obtain the first initiation and evolution data.

[0058] This application utilizes a binocular vision system and an infrared thermal imaging system for non-contact detection of ultra-high cycle fatigue damage. First, a laser displacement sensor captures the point displacement of the test specimen. The software system of the binocular vision system converts the point displacement signal into field displacement / strain data. A synchronous triggering device reads the signal output from the mechanical displacement sensor. Then, the binocular camera acquires periodic phase images of the test specimen at different phases within several cycles using ultra-cycle sampling. Finally, the thermal imaging signal from the infrared thermal imager is sampled. By establishing a correspondence between the visual image displacement / strain data and the thermal imaging signal, the damage initiation time of the test specimen is determined, and the damage is quantitatively characterized. This approach effectively reduces the amount of data acquisition without compromising accuracy, enabling long-term monitoring and data acquisition in ultra-high cycle fatigue tests. It achieves the determination of the damage initiation time and the quantitative characterization of damage evolution in ultra-high cycle fatigue tests, providing an effective technical means for non-contact detection of ultra-high cycle fatigue tests.

[0059] Finally, the following points should be noted: First, in the description of this application, it should be noted that, unless otherwise specified and limited, the terms "installation", "connection", and "linkage" should be interpreted broadly, and can be mechanical or electrical connections, or internal connections between two components, or direct connections. "Up", "down", "left", "right", etc. are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may change.

[0060] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other.

[0061] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A non-contact detection method for ultra-high cycle fatigue damage, characterized in that, include: A binocular vision system and an infrared thermal imaging system were set up separately. The binocular vision system was controlled to acquire synchronous displacement change signals and visual images of the test piece, and the infrared thermal imaging system was controlled to acquire infrared images of the test piece. The binocular vision system includes a laser displacement sensor and a binocular camera, and the infrared thermal imaging system includes an infrared thermal imager. The binocular vision system and infrared thermal imaging system are assembled at a certain position on the test piece. Feature points are sprayed on the surface of the test piece. The feature points of the test piece are detected and adjusted by the binocular camera and infrared thermal imager until both the binocular camera and infrared thermal imager can acquire high-resolution data to determine the current position and distance of the binocular vision system and infrared thermal imaging system. A laser displacement sensor is installed at the bottom of the test piece. Ultrasonic excitation is applied to the test piece. The laser displacement sensor measures the point displacement of the test piece from the bottom of the test piece and outputs the point displacement signal of the test piece to the software system of the binocular vision system. The software system of the binocular vision system converts the point displacement signal of the test specimen into field displacement / strain data of the vision system; Ultra-high cycle fatigue accelerated tensile-compression test was conducted to capture high-frequency signals in the field displacement / strain data of the vision system, obtain excitation signals of different phases at high frequency, until a complete periodic signal was formed; The phase-locked loop signal processing method is used to multiply the high-frequency complete periodic signal with the reference signal, and then filter it through an adjustable low-pass filter to obtain the low-frequency signal; After reading the low-frequency signal output by the laser displacement sensor, the binocular camera acquires periodic phase photos of the test piece under different phases within several cycles through super-cycle sampling. The software system of the binocular vision camera acquires the average displacement and strain change of each reference point in the periodic camera image, obtains visual image displacement / strain data, compares it with the simulated displacement signal collected by the laser displacement sensor, and judges whether the corresponding parameters of the visual image displacement / strain data and the field displacement / strain data of the vision system are within the set range. If they are, the sampling is determined to be accurate. If not, the position of the binocular vision system is adjusted and measured again until the set range requirements are met. When the stress difference of the sampling points within the field of view of the binocular camera reaches the limit value, the excitation signal is input to the infrared thermal imaging system. The infrared thermal imager uses infrared thermal imaging signals on the test piece. The software system of the infrared thermal imaging system analyzes the infrared thermal imaging signals to determine the damage initiation time and obtain the second initiation and evolution data of the damage of the test piece. The software system of the binocular vision system determines the first initiation and evolution data of the damage to the test specimen through the field displacement / strain data of the vision system, calculates the displacement of the test specimen through the displacement relationship of feature points, and establishes the correspondence between temperature signal and strain / displacement signal by combining the second initiation and evolution data of the infrared thermal imaging system, thus completing the determination of the damage initiation time of the test specimen and the quantitative characterization of the damage.

2. The non-contact detection method for ultra-high cycle fatigue damage as described in claim 1, characterized in that: The hardware system of the binocular vision system also includes a control console, a synchronization triggering device, a light source system, a camera movement track, and a vision system frame. The binocular camera and light source system are both mounted on the vision system frame. The laser displacement sensor is electrically connected to the synchronization triggering device, which is electrically connected to the control console. The vision system frame is mounted on the camera movement track and can move horizontally along the track. The software system of the binocular vision system includes an acquisition system, a calibration system, and a full-field strain analysis system. The acquisition system receives detection data from the laser displacement sensor and the binocular camera and sends it to the calibration system and the full-field strain analysis system. The calibration system determines the sharpness based on the detection data; if the sharpness is insufficient, it adjusts the position of the binocular vision system and re-detects until the calibration system determines that it can receive detection data at a specified resolution. The full-field strain analysis system performs full-field strain analysis on the received detection data after calibration to obtain initial and evolutionary data. The hardware system of the infrared thermal imaging system also includes a camera bracket and a computer module. The infrared thermal imager is fixedly connected to the camera bracket and electrically connected to the computer module. The software system of the infrared thermal imager is located within the computer module. The software system of the infrared thermal imager includes an acquisition and analysis module and an image processing module. The acquisition and analysis module can receive the infrared thermal imaging signals acquired by the infrared thermal imager and analyze them to obtain image data of feature points. The image processing module can analyze and process the image data of feature points to obtain the second initiation and evolution data of the test piece damage.

3. The non-contact detection method for ultra-high cycle fatigue damage as described in claim 2, characterized in that: The high-frequency signal emitted by the laser displacement sensor is captured by a synchronous triggering device. The specific method is as follows: a step size is set, a reference low-frequency signal is given first, the synchronous triggering device captures the reference low-frequency signal, and after the capture is completed, the frequency of the reference low-frequency signal is gradually increased according to the step size. The signal is captured once for each increase until the high-frequency signal of the specified frequency is captured.

4. The non-contact detection method for ultra-high cycle fatigue damage as described in claim 1, characterized in that, The method for performing super-periodic sampling with the binocular camera is as follows: a specific phase is set, and photos within the specific phase are automatically acquired. Then, the photos of different specific phases within several acquired periods are reconstructed to form a complete periodic phase photo.

5. The non-contact detection method for ultra-high cycle fatigue damage as described in claim 1, characterized in that: The test specimen surface is coated with a matte paint to form irregular feature points. Feature points are collected by selecting one feature point at each of the three positions of the test specimen: top, middle, and bottom.

6. The non-contact detection method for ultra-high cycle fatigue damage as described in claim 5, characterized in that, The method for obtaining the visual image displacement / strain data is as follows: images of the ROI regions of three feature points are acquired by a binocular camera, and the average displacement and strain change of the three ROI regions of the test piece in the tensile loading direction are analyzed and calculated to obtain the visual image displacement / strain data.

7. The non-contact detection method for ultra-high cycle fatigue damage as described in claim 1, characterized in that, The method for determining the damage initiation time of the infrared thermal imaging signal is as follows: First, the infrared thermal imaging signal is subjected to five-point cubic smoothing. Then, the initial time of the average temperature rise is determined by taking the derivative of the fitting function based on the least squares method. This time is the damage initiation position. The second initiation and evolution data are calculated by selecting the average temperature of the infrared thermal imaging signal based on the energy dissipation method. Then, the data is processed by the Gaussian filtering method. Finally, the energy dissipation is obtained by combining the two-dimensional energy dissipation method. The changes in temperature dissipation at different times are analyzed and determined to obtain the second initiation and evolution data.

8. The non-contact detection method for ultra-high cycle fatigue damage as described in claim 1, characterized in that, The method for obtaining the first initiation and evolution data is as follows: acquire periodic phase photos of the test specimen captured by a binocular camera, perform grayscale and brightness processing first, then perform feature matching based on SIFT or SURF algorithms, finally calculate the displacement of the test specimen through the displacement relationship of pixel points, calculate the strain of the test specimen through differential calculation, and finally obtain the state of the test specimen at different times to form visual system field displacement / strain data; set a stress change threshold, and the position in the periodic phase photo of the test specimen that meets the stress change threshold is the damage initiation time, and then obtain the first initiation and evolution data by acquiring the visual system field displacement / strain data after the damage initiation time.

Images