High-precision background schlieren tomography system and method for measuring jet flow of an aero-engine

Abstract

A kind of high-precision background schlieren tomography system and method for measuring aircraft engine exhaust flow.The system includes a movable lifting platform, a ring frame assembly, five camera mounts, five precision angle fine adjustment frames, five high-speed industrial cameras, five pairs of background plate mounts, five active light-emitting background plates, a synchronization controller, an image acquisition and processing computer, a test bench and an aircraft engine.The effects of the present application are as follows:1) significantly reduces the effect of astigmatism, solves the problem of shock wave structure smoothing caused by defocus blur in traditional methods from the physical level, improves the imaging clarity of the density jump region, greatly simplifies the calibration process, and can directly use the classic Zhang Zhengyou calibration method, effectively suppresses the interference of flame, fundamentally solves the interference problem of flame emission on displacement measurement, improves the structural accuracy of the system, reduces the system error, has a reasonable structure layout, and improves the stability and maintainability of the system.

Description

Technical Field

[0001] This invention belongs to the technical field of aero-engine testing equipment, specifically relating to a high-precision background schlieren tomography system and method for measuring aero-engine exhaust flow. Background Technology

[0002] Precise measurement of aero-engine exhaust flow is a key technique for evaluating engine performance, validating nozzle design, and studying exhaust noise mechanisms. Supersonic exhaust flows contain complex shock trains, expansion wave systems, and turbulent mixing layers; accurate measurement of these structures is crucial for optimal engine design.

[0003] Traditional contact measurement methods (such as probes and thermocouples) can interfere with the flow field, and single-point measurements are insufficient to obtain full-field information. Among non-contact optical measurement methods, schlieren and shadow methods can visualize the flow field structure, but they typically only provide two-dimensional information of the optical path integration, making quantitative three-dimensional measurements difficult. Background schlieren tomography combines background schlieren and computed tomography, enabling quantitative measurements of non-axisymmetric three-dimensional flow fields.

[0004] However, existing BOST technology faces the following technical bottlenecks when applied to the measurement of aero-engine exhaust flow:

[0005] 1) Defocusing and Blur Problem: Traditional BOST requires the camera to be focused on the background plate, with the flow field under test in a defocused position. When measuring supersonic jets with strong density gradients, the blurring effect caused by defocusing can "smooth" abrupt structural changes such as shock waves, leading to displacement measurement failure. This astigmatism effect blurs the imaging in areas of abrupt density changes, severely affecting measurement accuracy.

[0006] 2) Difficulty in multi-camera calibration: Since the cameras focus on their respective backgrounds, there is no common field of view between the cameras. The traditional Zhang Zhengyou calibration method cannot be used to calibrate the multi-camera system directly, which makes it difficult to accurately determine the spatial position relationship, thus introducing system errors.

[0007] 3) Flame emission interference: The exhaust flow field of aero-engines emits light over a wide wavelength range. This stray light will be superimposed on the background pattern, seriously interfering with the accuracy of displacement field calculation.

[0008] 4) Insufficient structural precision: Traditional BOST experiments often use aluminum profiles to build the frame, which has limited installation precision and makes it difficult to ensure that the optical axes of multiple cameras intersect precisely at one point. Summary of the Invention

[0009] To address the aforementioned problems, the present invention aims to provide a high-precision background schlieren tomography system and method for measuring the exhaust flow of aero-engines.

[0010] To achieve the above objectives, the high-precision background schlieren tomography system for measuring aero-engine exhaust jets provided by this invention includes: a movable lifting platform, a ring frame assembly, five camera mounting bases, five precision angle fine-tuning frames, five high-speed industrial cameras, five pairs of background plate mounting bases, five active-emitting background plates, a synchronization controller, an image acquisition and processing computer, a test bench, and an aero-engine; wherein, the lower end of the test bench is set on the ground, and a support frame is provided at the rear of the top surface; the movable lifting platform is located on the outer rear end of the test bench; the ring frame assembly is installed on the top of the movable lifting platform, including an inner semi-circular plate and an outer semi-circular plate. The test platform consists of a circular plate, five camera support plates, and six background plate side pillars. The inner and outer semi-circular plates are arranged parallel to each other vertically, with their upper ends forming a straight line and an arched frame protruding upwards from the middle of the upper end. The lower part of the arched frame is hollowed out, and the inner semi-circular plate is located near the rear end of the test platform. The two ends of the five camera support plates are equally spaced between the arched frames on the inner and outer semi-circular plates, and the included angle between adjacent camera support plates is 36°. Each camera support plate has an opening in the middle. The two ends of the six background plate side pillars are equally spaced between the bottom arc-shaped edges of the inner and outer semi-circular plates. The five high-speed industrial cameras are arranged alternately with the camera support plates. A precision angle adjustment bracket is installed in the opening of each camera support plate. A camera mounting base is installed on the movable platform of each precision angle adjustment bracket. The top surface of the camera mounting base is machined with mounting holes that match the bottom of the high-speed industrial camera. One high-speed industrial camera is installed in each mounting hole, and the optical axes of the five high-speed industrial cameras precisely intersect at the center O of the measurement area of ​​the flow field to be measured. A background plate mounting base is installed on the inner surface of the inner and outer semicircular plates between adjacent background plate side pillars. The two ends of each active light-emitting background plate are fixed to the inner semicircular plate. The test bench is mounted on a pair of background plate mounting bases for the plate and the outer semi-circular plate, with each side edge supported by two corresponding background plate side columns. The aero-engine is fixed on the support frame of the test bench, with its exhaust nozzle facing the center O of the flow field measurement area. The synchronous controller and the image acquisition and processing computer are placed on the test bench. The image acquisition and processing computer is a high-performance workstation with a built-in image acquisition card and GPU, used to store images and run dedicated processing software. It is electrically connected to the high-speed industrial camera, the synchronous controller, and the aero-engine, and is used to receive and store synchronously acquired image sequences, and perform subsequent image processing and three-dimensional density field reconstruction.

[0011] The active light-emitting background plate includes a substrate and multiple thin optical fibers. The substrate is made of low-reflectivity aluminum alloy material, and its surface is precisely machined with tens of thousands of micro-holes according to a preset random speckle pattern. It is also treated with black anodizing to achieve a matte black finish. One end of the thin optical fiber is inserted and fixed in the micro-hole, and the other end is gathered and connected to an external laser.

[0012] The movable lifting platform is a scissor-type electric lifting platform with casters installed at the bottom, and a lifting stroke of 180mm-540mm.

[0013] Both the inner and outer semicircular plates are made of 6061 aluminum alloy and machined by CNC machine tools, with a thickness of 6mm.

[0014] The camera support plate is made of 6061 aluminum alloy and machined by CNC machine tool. It has precision positioning stops at both ends, which are precisely matched with the positioning holes on the arched frame of the inner and outer semicircular plates and fixed by corner brackets.

[0015] The side pillars of the background panel are made of standard 30mm×30mm industrial aluminum profiles and are connected to the inner and outer semi-circular plates by angle brackets or bolts.

[0016] The precision angle fine-tuning frame adopts a two-axis precision optical adjustment frame with an adjustment accuracy of 0.01°.

[0017] The core of the synchronization controller is an FPGA development board, whose output is connected to the external trigger interface of five high-speed industrial cameras via coaxial cables.

[0018] The high-speed industrial camera uses a C-mount lens with an 8mm focal length.

[0019] The method for measuring the three-dimensional density field distribution of aero-engine exhaust jet using the aforementioned high-precision background schlieren tomography system includes the following steps performed in sequence:

[0020] 1) System calibration: Place the high-precision checkerboard calibration plate at the center O of the flow field measurement area to be measured. Under the control of the image acquisition and processing computer, use five high-speed industrial cameras to simultaneously capture images of the high-precision checkerboard calibration plate in different postures. Then, use Zhang Zhengyou calibration method to calibrate the intrinsic parameters of each high-speed industrial camera individually, and calculate the extrinsic parameters of all high-speed industrial cameras 6 in a unified world coordinate system.

[0021] 2) Acquiring Reference Images: With the aircraft engine off and no flow field, the laser is activated. The laser light is transmitted through the thin optical fiber of the active light-emitting background plate, forming high-brightness point light sources at the micro-holes on the substrate surface. This creates a random speckle pattern composed of thousands of emitting points. Then, a high-level TTL trigger signal is generated by the synchronization controller, triggering all high-speed industrial cameras to simultaneously begin exposure and synchronously acquire the aforementioned random speckle image. For the first... Each viewpoint captures 20 reference images from each high-speed industrial camera. , , ..., Then, for each pixel coordinate in the reference image The arithmetic mean of the gray values ​​at each location is calculated.

[0022] ;

[0023] in, Here are the pixel coordinates, and k is the frame number. For the first Reference images from various perspectives in pixel coordinates The average gray value at that location;

[0024] 3) Image Acquisition and Measurement: After starting the aircraft engine and waiting for the exhaust flow field to stabilize, the five high-speed industrial cameras are triggered again by the synchronous controller to synchronously acquire random speckle images. Each high-speed industrial camera acquires a continuous sequence of multiple frames. As a measurement image, to capture the dynamic changes of the flow field under test;

[0025] 4) Calculate the displacement field: for each viewpoint Every moment The measured image was compared with the aforementioned reference image, and the two-dimensional displacement field of the random speckle pattern was calculated using a cross-correlation algorithm. ;

[0026] 5) Three-dimensional density field reconstruction: The three-dimensional spatial region of the flow field to be measured is discretized into a voxel mesh. Based on the calibration results of step 1), the ray path from each voxel to each pixel of the high-speed industrial camera is determined, thereby constructing a tomographic projection matrix. Then, the ray deflection angle is calculated based on the two-dimensional displacement field, and the projection matrix equation between the ray deflection angle and the three-dimensional density field gradient is established. Finally, the algebraic reconstruction algorithm or its improved algorithm is used to solve the projection matrix equation to obtain the three-dimensional density field distribution of the flow field measurement area. .

[0027] The key innovation of this invention lies in:

[0028] First, the camera's focusing plane is changed. Traditional BOST cameras focus on the background plate, while this invention sets the focusing plane of the high-speed industrial camera at the center of the measurement area of ​​the flow field to be measured. This change brings two direct effects: First, the flow field to be measured is within the focal depth range of the high-speed industrial camera, fundamentally reducing the astigmatism effect caused by abrupt changes in the density of the dense flow field, making the imaging of structures such as shock waves and expansion waves clearer; second, all high-speed industrial cameras focus on the same point at the center of the measurement area of ​​the flow field to be measured, giving the five high-speed industrial cameras a common field of view. This allows the calibration plate to be easily placed at the center, and the endo- and extrinsic parameters of all high-speed industrial cameras can be calibrated in one go using the classic Zhang Zhengyou calibration method.

[0029] Second, an active-emitting background plate is used. Traditional background plates are passive reflective and easily affected by flame light interference. This invention uses an active-emitting background plate, which consists of a substrate and densely, irregularly distributed fine optical fibers. One end of the fiber is embedded in the substrate surface as the light emission point, and the other end is converged and connected to an external laser of a specific wavelength, which avoids the main radiation spectrum range of aviation kerosene combustion flames. The advantages of this design are: 1) Active emission provides high-intensity illumination, allowing for shorter exposure times and smaller apertures, further reducing defocus blur; 2) The specific wavelength can effectively filter out stray light interference from the flame; 3) By controlling the diameter of the fine optical fibers, the distance M, and the aperture value, the size of the imaging blur spot can be precisely controlled to achieve optimal measurement conditions.

[0030] Third, the high-precision composite frame structure and reasonable installation sequence. The ring frame assembly adopts a composite structure of "CNC precision machined parts + aluminum profiles". The precision angle fine-tuning bracket is directly mounted on the CNC precision machined camera support plate, ensuring the accuracy of the adjustment reference; the high-speed industrial camera is mounted on the precision angle fine-tuning bracket via a camera mounting base, facilitating disassembly and maintenance. The background plate mounting base is directly fixed to the inner and outer semi-circular plates, forming a stable support structure independent of the camera side. This layout ensures the accuracy of optical axis convergence while avoiding mutual interference between camera adjustment and background plate installation.

[0031] The high-precision background schlieren tomography system and method for measuring aero-engine exhaust flow provided by this invention have the following beneficial effects:

[0032] 1) Significantly reduce astigmatism effect: By moving the focal plane to the center of the flow field, the flow field to be measured is within the focal depth range of the camera, which solves the problem of smoothing the shock wave structure caused by defocusing in traditional methods from a physical perspective, and improves the imaging clarity in areas of abrupt density change.

[0033] 2) The calibration process is greatly simplified: all cameras have a common field of view, which makes the calibration of multi-camera systems change from "difficult joint calibration" to "simple and accurate calibration", and the classic Zhang Zhengyou calibration method can be directly used.

[0034] 3) Effectively suppresses flame interference: The active light-emitting background plate with a specific wavelength fundamentally solves the problem of flame light interfering with displacement measurement.

[0035] 4) Improve system structural accuracy: The CNC integrated frame, combined with the precision fine-tuning bracket directly mounted on the column, ensures the spatial alignment accuracy of the camera optical axis and reduces system errors.

[0036] 5) Reasonable structural layout: The camera adjustment system and the background support system are separated, avoiding mutual interference and improving the stability and maintainability of the system. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of the overall structure of the high-precision background schlieren tomography system for measuring the exhaust flow of aero-engines provided by the present invention.

[0038] Figure 2 This is a schematic diagram of the ring frame assembly structure in this invention.

[0039] Figure 3 This is an assembly diagram of the camera support plate, precision angle fine-tuning frame, camera mounting base, and high-speed industrial camera in this invention.

[0040] Figure 4 This is a schematic diagram showing the connection relationship between the background plate mounting base and the inner and outer semi-circular plates and the active light-emitting background plate in an embodiment of the present invention.

[0041] Figure 5 This is a schematic diagram of the geometric relationship between the high-speed industrial camera and the center of the measurement area of ​​the flow field to be measured in this invention.

[0042] Figure 6 This is a schematic diagram of the active light-emitting background plate structure in this invention. Detailed Implementation

[0043] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings.

[0044] like Figures 1 to 6As shown, the high-precision background schlieren tomography system for measuring the exhaust flow of an aero-engine provided by this invention includes: a movable lifting platform 1, a ring frame assembly 2, five camera mounting bases 4, five precision angle fine-tuning frames 5, five high-speed industrial cameras 6, five pairs of background plate mounting bases 7, five active light-emitting background plates 8, a synchronization controller 10, an image acquisition and processing computer 11, a test bench 12, and an aero-engine 13; wherein, the lower end of the test bench 12 is set on the ground, and a support frame is provided at the rear of the top surface; the movable lifting platform 1 is set on the outer rear end of the test bench 12; the ring frame assembly 2 is installed on the top of the movable lifting platform 1, including an inner semi-circular plate 21 and an outer semi-circular plate 22. Five camera support plates 3 and six background plate side pillars 9; the inner semi-circular plate 21 and the outer semi-circular plate 22 are arranged parallel to each other in the vertical direction, with the upper end being straight and the upper middle part protruding upward to form an arched frame. The lower part of the arched frame is hollow, and the inner semi-circular plate 21 is close to the rear end of the test platform 12; the two ends of the five camera support plates 3 are equally spaced between the arched frames on the inner semi-circular plate 21 and the outer semi-circular plate 22, and the included angle between adjacent camera support plates 3 is 36°. Each camera support plate 3 has an opening in the middle; the two ends of the six background plate side pillars 9 are equally spaced between the bottom arc-shaped edges of the inner semi-circular plate 21 and the outer semi-circular plate 22, and are connected to... Camera support plates 3 are alternately arranged; a precision angle fine-tuning bracket 5 is installed in the opening on each camera support plate 3; a camera mounting base 4 is installed on the movable platform of each precision angle fine-tuning bracket 5, and the top surface of the camera mounting base 4 is machined with mounting holes that match the bottom of the high-speed industrial camera 6. One high-speed industrial camera 6 is installed in each mounting hole, and the optical axes of the five high-speed industrial cameras 6 are precisely intersected at the center O of the measurement area of ​​the flow field to be measured; a background plate mounting base 7 is installed on the inner side of the inner semicircular plate 21 and the outer semicircular plate 22 between adjacent background plate side columns 9; the two ends of each active light-emitting background plate 8 are respectively fixed to one of the inner semicircular plate 21 and the outer semicircular plate 22. The background plate is mounted on the mounting base 7, with two corresponding background plate side columns 9 supporting each side edge. The aero-engine 13 is fixed on the support frame of the test bench 12, with its tail nozzle facing the center O of the measurement area of ​​the flow field to be measured, to ensure that the tail jet flows through the center of the measurement area. The synchronous controller 10 and the image acquisition and processing computer 11 are placed on the test bench 12. The image acquisition and processing computer 11 is a high-performance workstation with a built-in image acquisition card and GPU, used to store images and run dedicated processing software. It is electrically connected to the high-speed industrial camera 6, the synchronous controller 10 and the aero-engine 13, respectively, to receive and store the synchronously acquired image sequence, and to perform subsequent image processing and three-dimensional density field reconstruction.

[0045] The active light-emitting background plate 8 includes a substrate 81 and multiple thin optical fibers 83. The substrate 81 is made of low-reflectivity aluminum alloy material, and its surface is precisely machined with tens of thousands of micro-holes according to a preset random speckle pattern. In this invention, the diameter of the micro-holes is 0.1 mm, and they are treated with black anodizing to be matte black, so as to minimize the interference of ambient light reflection on the random speckle pattern and ensure that only the light spots emitted from the optical fibers are captured by the high-speed industrial camera 6. One end of the thin optical fiber 83 is inserted and fixed in the micro-hole, and the other end is gathered and connected to an external laser 82.

[0046] The movable lifting platform 1 is a scissor-type electric lifting platform with casters installed at the bottom for easy movement. The lifting stroke is 180mm-540mm to accommodate the tail nozzle positions of different types of aircraft engines 13.

[0047] Both the inner semicircular plate 21 and the outer semicircular plate 22 are made of 6061 aluminum alloy by CNC machine tool and have a thickness of 6mm.

[0048] The camera support plate 3 is made of 6061 aluminum alloy by CNC machine tool. It has precision positioning stops at both ends, which are precisely matched with the positioning holes on the arched frame of the inner semi-circular plate 21 and the outer semi-circular plate 22, and are fixed by corner brackets.

[0049] The background panel side pillar 9 is made of standard 30mm×30mm industrial aluminum profile and is connected to the inner semi-circular plate 21 and the outer semi-circular plate 22 by angle brackets or bolts.

[0050] The precision angle fine-tuning frame 5 adopts a two-axis precision optical adjustment frame, which can realize pitch and yaw adjustment with an adjustment accuracy of 0.01°.

[0051] The core of the synchronization controller 10 is an FPGA development board, and its output is connected to the external trigger interface of five high-speed industrial cameras 6 via a coaxial cable.

[0052] The high-speed industrial camera 6 uses a C-mount lens with a focal length of 8mm.

[0053] The distance from the active luminescent background plate 8 to the center O of the measurement area of ​​the flow field to be measured The parameters are adjustable and can be achieved through fixed adjustment holes. During installation, it must be ensured that the surface of the active light-emitting background plate 8 is strictly perpendicular to the optical axis of its corresponding high-speed industrial camera 6.

[0054] The distance L from the center O of the measurement area of ​​the flow field to be measured to the lens plane of the high-speed industrial camera 6 is determined according to a preset measurement field of view. In this invention, the measurement area is a diameter The circular region, take the diameter =150mm. Based on the sensor size (1 / 4 inch, sensor width W is 3.2mm) and lens focal length of the high-speed industrial camera 6. =8mm, the vertical field of view can be calculated. for:

[0055] ;

[0056] To ensure complete coverage of the measurement area and allow sufficient margin, a larger field of view is adopted. =21°. To make the diameter The circular area is entirely within the horizontal field of view of the high-speed industrial camera 6. Based on geometric relationships, the distance... Should meet:

[0057] ;

[0058] Known Substituting into the calculation, we get:

[0059] ;

[0060] Therefore, distance It is set to approximately 405mm.

[0061] The laser 83 emits green light with a wavelength of 532 nm. This wavelength was chosen based on the following considerations: the exhaust flame of the aero-engine 13 is mainly generated by kerosene combustion, and its radiation spectrum includes characteristic spectral lines of combustion intermediate products (such as OH-based, CH-based, and C2-based) and the blackbody radiation continuous spectrum of high-temperature carbon particles. Specifically, the OH-based characteristic spectral lines are mainly concentrated around 309 nm, the CH-based around 431 nm, and the C2-based around 473 nm and 516 nm. 532 nm is located in the intervening region of these main characteristic spectral lines and is far from the infrared band where flame radiation is strongest. Furthermore, the radiation intensity of the flame continuous spectrum at 532 nm is relatively weak. Therefore, selecting 532 nm as the background light source, combined with a corresponding narrowband filter (center wavelength 532 nm, half-width 10 nm) installed in front of the lens of the high-speed industrial camera 6, can effectively suppress stray flame light interference, allowing the high-speed industrial camera 6 to primarily receive the active emission signal emitted by the active luminous background plate 8, thereby significantly improving the signal-to-noise ratio of the displacement field calculation.

[0062] Another advantage of active-emitting design is the ability to precisely control the size of the imaging blur spot. According to geometric optics, the blur spot diameter... Determined by the formula. In this system, the high-speed industrial camera 6 focuses on the center O of the measurement area of ​​the flow field to be measured, therefore the image at the center O is clear. For a light-emitting point on the active light-emitting background plate 8 (located at a distance of [distance] from the center O), (At this location), its image on the sensor of the high-speed industrial camera 6 is a blur spot. At this point, the diameter of the blur spot... for:

[0063] ;

[0064] in, For the lens focal length, This refers to the aperture value. By selecting an appropriate aperture value... ,For example and adjust The diameter of the dispersion spot can be controlled. In this invention, the diameter of the dispersion spot is controlled through calculation and adjustment. The optimal size for subpixel displacement measurement is approximately 3 pixels, which ensures the sensitivity of the displacement measurement while avoiding the loss of accuracy caused by excessively large speckles.

[0065] The method for measuring the three-dimensional density field distribution of aero-engine exhaust jet using the aforementioned high-precision background schlieren tomography system includes the following steps performed in sequence:

[0066] 1) System calibration: The high-precision checkerboard calibration plate is placed at the center O of the flow field measurement area to be measured. Under the control of the image acquisition and processing computer 11, five high-speed industrial cameras 6 are used to simultaneously capture images of the high-precision checkerboard calibration plate in different postures. Then, the intrinsic parameters of each high-speed industrial camera 6 are individually calibrated using the Zhang Zhengyou calibration method, and the extrinsic parameters of all high-speed industrial cameras 6 in a unified world coordinate system are calculated. The average reprojection error of the calibration in this invention is less than 0.15 pixels.

[0067] 2) Acquiring Reference Images: With the aero-engine 13 not running and in a state of no flow field, the laser (82) is turned on. After the laser is transmitted through the thin optical fiber 83 of the active light-emitting background plate 8, a high-brightness point light source is formed at the micro-holes on the surface of the substrate 81, forming a random speckle pattern composed of thousands of light-emitting points. Then, the synchronization controller 10 is controlled to generate a high-level TTL trigger signal, which triggers all high-speed industrial cameras 6 to simultaneously start exposure and synchronously acquire the above-mentioned random speckle image. The time synchronization accuracy is better than 1 microsecond. Each high-speed industrial camera captures 20 reference images from 6 different perspectives. , , ..., Then, for each pixel coordinate in the reference image The gray values ​​at each location are averaged arithmetically to reduce random noise.

[0068] ;

[0069] in, Here are the pixel coordinates, and k is the frame number. For the first Reference images from various perspectives in pixel coordinates The average gray value at that location;

[0070] 3) Image Acquisition and Measurement: Start the aero-engine 13. After the tail jet flow field stabilizes, trigger the five high-speed industrial cameras 6 again via the synchronous controller 10 to synchronously acquire random speckle images. Each high-speed industrial camera 6 acquires a continuous sequence of multiple frames. As a measurement image, to capture the dynamic changes of the flow field under test;

[0071] 4) Calculate the displacement field: for each viewpoint Every moment The measured image was compared with the aforementioned reference image, and the two-dimensional displacement field of the random speckle pattern was calculated using a cross-correlation algorithm. The magnitude and direction of the displacement field directly reflect the degree of deflection of light rays after passing through the flow field;

[0072] 5) Three-dimensional density field reconstruction: The three-dimensional spatial region of the flow field to be measured is discretized into a voxel mesh. Based on the calibration results of step 1), the ray path from each voxel to each 6-pixel high-speed industrial camera is determined, thereby constructing a tomographic projection matrix. Then, the ray deflection angle is calculated based on the two-dimensional displacement field, and the projection matrix equation between the ray deflection angle and the three-dimensional density field gradient is established. Finally, an algebraic reconstruction algorithm or its improved algorithm (such as the SART algorithm) is used to solve the projection matrix equation to obtain the three-dimensional density field distribution of the flow field measurement area. .

Claims

1. A high-precision background schlieren tomography system for measuring the exhaust flow of an aero-engine, characterized in that: The system includes: a movable lifting platform (1), a ring frame assembly (2), five camera mounting bases (4), five precision angle fine-tuning frames (5), five high-speed industrial cameras (6), five pairs of background plate mounting bases (7), five active light-emitting background plates (8), a synchronization controller (10), an image acquisition and processing computer (11), a test bench (12), and an aero-engine (13); wherein, the lower end of the test bench (12) is set on the ground, and a support frame is provided on the rear of the top surface; the movable lifting platform (1) is set on the outer rear end of the test bench (12); the ring frame assembly (2) is installed on the top of the movable lifting platform (1), including the inner side The test bench consists of a semi-circular plate (21), an outer semi-circular plate (22), five camera support plates (3), and six background plate side pillars (9). The inner semi-circular plate (21) and the outer semi-circular plate (22) are arranged parallel to each other in the vertical direction. The upper end is straight and the middle of the upper end protrudes upward to form an arched frame. The lower part of the arched frame is hollow. The inner semi-circular plate (21) is close to the rear end of the test bench (12). The two ends of the five camera support plates (3) are connected at equal intervals between the arched frames on the inner semi-circular plate (21) and the outer semi-circular plate (22). The included angle between adjacent camera support plates (3) is 36°. The middle of each camera support plate (3) forms a... There is an opening; the two ends of the six background plate side columns (9) are connected at equal intervals between the bottom arc edges of the inner semicircular plate (21) and the outer semicircular plate (22), and are alternately set with the camera support plate (3); a precision angle fine-tuning frame (5) is installed in the opening on each camera support plate (3); a camera mounting base (4) is installed on the movable platform of each precision angle fine-tuning frame (5), and the top surface of the camera mounting base (4) is machined with mounting holes that match the bottom of the high-speed industrial camera (6). A high-speed industrial camera (6) is installed in each mounting hole, and the optical axes of the five high-speed industrial cameras (6) are precisely intersected at the center O of the measurement area of ​​the flow field to be measured. A background plate mounting seat (7) is installed on the inner side of the inner semicircular plate (21) and the outer semicircular plate (22) between the adjacent background plate side pillars (9); the two ends of each active light-emitting background plate (8) are fixed on a pair of background plate mounting seats (7) of the inner semicircular plate (21) and the outer semicircular plate (22), and the two sides are supported by two corresponding background plate side pillars (9); the aero-engine (13) is fixed on the support frame of the test bench (12), and its tail nozzle faces the center O of the flow field measurement area to be measured; the synchronous controller (10) and the image acquisition and processing computer (11) are placed on the test bench (12);The image acquisition and processing computer (11) is a high-performance workstation with a built-in image acquisition card and GPU for storing images and running dedicated processing software. It is electrically connected to the high-speed industrial camera (6), the synchronization controller (10), and the aircraft engine (13) to receive and store synchronously acquired image sequences and perform subsequent image processing and three-dimensional density field reconstruction.

2. The high-precision background schlieren tomography system for measuring aero-engine exhaust flow according to claim 1, characterized in that: The active light-emitting background plate (8) includes a substrate (81) and multiple thin optical fibers (83); the substrate (81) is made of aluminum alloy with low reflectivity, and the surface is precisely machined with tens of thousands of micro holes according to a preset random speckle pattern, and is treated with black anodizing to be matte black; one end of the thin optical fiber (83) is inserted and fixed in the micro hole, and the other end is gathered and connected to an external laser (82).

3. The high-precision background schlieren tomography system for measuring aero-engine exhaust flow according to claim 1, characterized in that: The movable lifting platform (1) is a scissor-type electric lifting platform with casters installed at the bottom and a lifting stroke of 180mm-540mm.

4. The high-precision background schlieren tomography system for measuring aero-engine exhaust flow according to claim 1, characterized in that: The inner semicircular plate (21) and the outer semicircular plate (22) are both made of 6061 aluminum alloy by CNC machine tool and have a thickness of 6mm.

5. The high-precision background schlieren tomography system for measuring aero-engine exhaust flow according to claim 1, characterized in that: The camera support plate (3) is made of 6061 aluminum alloy by CNC machine tool. It has precision positioning stops at both ends, which are precisely matched with the positioning holes on the arched frame of the inner semi-circular plate (21) and the outer semi-circular plate (22), and fixed by corner brackets.

6. The high-precision background schlieren tomography system for measuring aero-engine exhaust flow according to claim 1, characterized in that: The background panel side pillars (9) are made of standard 30mm×30mm industrial aluminum profiles and are connected to the inner semi-circular plate (21) and the outer semi-circular plate (22) by angle brackets or bolts.

7. The high-precision background schlieren tomography system for measuring aero-engine exhaust flow according to claim 1, characterized in that: The precision angle fine-tuning frame (5) adopts a two-axis precision optical adjustment frame with an adjustment accuracy of 0.01°.

8. The high-precision background schlieren tomography system for measuring aero-engine exhaust flow according to claim 1, characterized in that: The core of the synchronous controller (10) is an FPGA development board, and its output is connected to the external trigger interface of five high-speed industrial cameras (6) via a coaxial cable.

9. The high-precision background schlieren tomography system for measuring aero-engine exhaust flow according to claim 1, characterized in that: The high-speed industrial camera (6) uses a C-mount lens with a focal length of 8mm.

10. A method for measuring the three-dimensional density field distribution of an aero-engine exhaust jet using a high-precision background schlieren tomography system as described in any one of claims 1 to 9, characterized in that: The method includes the following steps performed in sequence: 1) System calibration: The high-precision checkerboard calibration plate is placed at the center O of the flow field measurement area to be measured. Under the control of the image acquisition and processing computer (11), five high-speed industrial cameras (6) are used to simultaneously capture images of the high-precision checkerboard calibration plate in different postures. Then, the Zhang Zhengyou calibration method is used to calibrate the intrinsic parameters of each high-speed industrial camera (6) separately, and the extrinsic parameters of all high-speed industrial cameras (6) in a unified world coordinate system are calculated. 2) Acquiring reference images: With the aero-engine (13) not running and in a state of no flow field, the laser (82) is turned on. After the laser is transmitted through the thin optical fiber (83) of the active light-emitting background plate (8), a high-brightness point light source is formed at the micro-holes on the surface of the substrate (81), forming a random speckle pattern composed of thousands of light-emitting points. Then, the synchronous controller (10) is controlled to generate a high-level TTL trigger signal, which triggers all high-speed industrial cameras (6) to simultaneously start exposure and synchronously acquire the above random speckle image. For the first reference image... Each high-speed industrial camera (6) captures 20 reference images from a single perspective. , , ..., Then, for each pixel coordinate in the reference image The arithmetic mean of the gray values ​​at each location is calculated. ; in, Here are the pixel coordinates, and k is the frame number. For the first Reference images from various perspectives in pixel coordinates The average gray value at that location; 3) Acquiring and measuring images: Start the aero-engine (13), and after the tail jet flow field stabilizes, trigger the five high-speed industrial cameras (6) again through the synchronous controller (10) to synchronously acquire random speckle images. Each high-speed industrial camera (6) acquires a continuous multi-frame sequence of images. As a measurement image, to capture the dynamic changes of the flow field under test; 4) Calculate the displacement field: for each viewpoint Every moment The measured image was compared with the aforementioned reference image, and the two-dimensional displacement field of the random speckle pattern was calculated using a cross-correlation algorithm. ; 5) Three-dimensional density field reconstruction: The three-dimensional spatial region of the flow field to be measured is discretized into a voxel grid. Based on the calibration results of step 1), the ray path from each voxel to each pixel of the high-speed industrial camera (6) is determined, thereby constructing a tomographic projection matrix. Then, the ray deflection angle is calculated based on the two-dimensional displacement field, and the projection matrix equation between the ray deflection angle and the three-dimensional density field gradient is established. Finally, the projection matrix equation is solved by the algebraic reconstruction algorithm or its improved algorithm to obtain the three-dimensional density field distribution of the flow field measurement area to be measured. .

Images