Spray combustion mixing test system based on bimodal CT system

Abstract

The invention provides a spray combustion mixing test system based on a bimodal CT (Computed Tomography) system. The spray combustion mixing test system comprises an X-ray source system, a background schlieren system, a high-speed imaging system and three-dimensional reconstruction algorithm software, the X-ray source system serves as a physical excitation source and is used for generating a high-energy X-ray beam to ensure that the X-ray beam penetrates through a liquid film with a certain thickness; the background schlieren system is used for carrying out clear imaging on the gasified mixing area, the background schlieren system is coupled with the high-speed imaging system and provides a high-power and high-beam-quality standardized optical signal for the high-speed imaging system, and the high-speed imaging system captures a transient projection image at an ultrahigh frame rate through a multi-angle detector array. Providing original data input; and the three-dimensional reconstruction algorithm software processes data in real time and reconstructs a dynamic three-dimensional concentration field by applying an intelligent algorithm. The method is used for accurately capturing a key physical mechanism in the fuel spray forming process and high-precision three-dimensional reconstruction of liquid and gaseous components in spray combustion.

Description

Technical Field

[0001] This invention relates to the field of spray combustion mixing test technology, and more specifically, to a spray combustion mixing test system based on a dual-modal CT system. Background Technology

[0002] X-ray computed tomography (CT) technology utilizes the absorption characteristics of X-rays by matter to reconstruct three-dimensional density distribution through multi-angle projection data. By rotating the sample or X-ray source to acquire multi-view projection data and combining it with reconstruction algorithms such as filtered back projection or algebraic reconstruction techniques, the three-dimensional structure of complex flow fields can be quantitatively analyzed. This technology is renowned for its strong penetration capability and non-invasive characteristics, making it particularly suitable for the diagnosis of optically opaque multiphase flows.

[0003] Background-Oriented Schlieren Computed Tomography (OCT) is a non-contact diagnostic technique based on image displacement to reconstruct the three-dimensional distribution of the gas refractive index field. This technique acquires offset images of the background pattern before and after a gas spray from multiple angles, and combines this multi-view data to calculate the displacement field, enabling accurate inversion of the gas density perturbation caused by the spray and its evolution in three-dimensional space. With technological advancements, this technique has gradually expanded into the field of spray combustion diagnostics.

[0004] Existing spray combustion testing technologies have the following limitations:

[0005] 1. Existing optical diagnostic technologies (such as structured illumination, ballistic imaging, and optical connectivity) each have their unique advantages. However, under conditions of high atomization concentration (OD>5), strong ultraviolet radiation interference, and coexistence of gas and liquid phases, these technologies are difficult to meet the stringent requirements of full-field, quantitative, and dynamic visualization of fuel spray in terms of penetration capability, three-dimensional spatial resolution, or temporal resolution. They are particularly limited in accurately capturing the spatial distribution of specific gaseous components.

[0006] 2. X-ray CT imaging technology still has significant problems in the field of high-speed dynamic CT: the time resolution of mainstream technologies is still at the second level, making it difficult to capture millisecond-level mixing processes; desktop systems are limited by light source intensity, and the frame rate is usually below 100Hz at sub-millimeter resolution, which conflicts with the needs of transient process observation.

[0007] 3. Background schlieren CT imaging technology is still in its early stages of development in spray CT imaging. The main challenges include background pattern optimization, field synchronization error, and scattering interference caused by multiphase interfaces. Summary of the Invention

[0008] To address the shortcomings of existing technologies, the present invention aims to provide a spray combustion mixing test system based on a dual-modal CT system, particularly a three-dimensional spray combustion field test technology based on X-ray computed tomography and background schlieren tomography, for accurately capturing the key physical mechanisms in the fuel spray formation process and high-precision three-dimensional reconstruction of liquid and gaseous components in spray combustion.

[0009] To solve the above problems, the technical solution of the present invention is as follows:

[0010] A spray combustion mixing test system based on a dual-modal CT system includes an X-ray source system, a background schlieren system, a high-speed imaging system, and three-dimensional reconstruction algorithm software. The X-ray source system serves as the physical excitation source, generating a high-energy X-ray beam to ensure penetration of a liquid film of a certain thickness. The background schlieren system is used to clearly image the vaporized mixing region. Coupled with the high-speed imaging system, the background schlieren system provides a standardized optical signal with high power and high beam quality. The high-speed imaging system uses a multi-angle detector array to capture transient projection images at an ultra-high frame rate, providing raw data input. The three-dimensional reconstruction algorithm software employs intelligent algorithms to process data in real time and reconstruct a dynamic three-dimensional concentration field.

[0011] Preferably, the X-ray source system includes a high-voltage generator, a pulse controller, and a collimation device to ensure stable X-ray emission and directional control.

[0012] Preferably, the background schlieren system is perfectly matched with the optical parameters of the high-speed imaging device to achieve high-precision dynamic flow field measurement.

[0013] Preferably, the high-speed imaging system is the core sensing unit, employing a high-performance detector array, which can capture X-ray projection images after penetrating the sample under test at a high frame rate in a very short time; through a precise synchronous triggering mechanism, it realizes continuous recording of the dynamic mixing process, providing raw data support for three-dimensional reconstruction.

[0014] Preferably, the three-dimensional reconstruction algorithm software is used to process and reconstruct three-dimensional visualization data acquired by the high-speed imaging system. Based on mathematical modeling and image processing algorithms, the software transforms the two-dimensional projection sequence into a dynamic three-dimensional image that reflects the evolution of the internal structure of the sample over time, thereby realizing quantitative analysis and in-depth research on the mixing process.

[0015] Preferably, the functions of the 3D reconstruction algorithm software include data preprocessing, fault reconstruction, motion compensation, and 3D visualization.

[0016] Compared with the prior art, the present invention has the following beneficial effects:

[0017] 1. This invention utilizes the powerful penetrating ability of high-energy X-rays to penetrate complex high-atomization concentration spray fields and high-temperature combustion environments without interference, enabling real-time and quantitative observation of the entire process of fuel spray formation and gaseous mixing.

[0018] 2. This invention transforms the complex process of spray combustion into a calculable and designable transparent space, providing an indispensable engineering verification and optimization platform for the safe, efficient, and reliable operation of aero-engines.

[0019] 3. This invention designs an integrated background illumination laser and a high-speed image acquisition module to construct an ultra-high-speed background schlieren imaging system, and couples it with efficient three-dimensional dynamic reconstruction technology, thereby achieving high-precision and high-speed capture of the pure gas phase mixing process.

[0020] 4. The technology of this invention can not only serve combustion research, but also provide a brand-new technical route and paradigm reference for three-dimensional in-situ observation of other extremely complex transient processes in aerospace, energy and power, and even biomedicine, which has profound value for leading the discipline and driving the industry. Attached Figure Description

[0021] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0022] Figure 1 This is a diagram of the spray combustion mixing test system based on a dual-modal CT system according to the present invention.

[0023] Figure 2 a is a schematic diagram of X-ray CT;

[0024] Figure 2 b is a schematic diagram of the single-angle background schlieren principle;

[0025] Figure 2 c is a schematic diagram of the three-angle background schlieren CT principle;

[0026] Figure 3 This is a diagram of a fast 3D reconstruction algorithm based on deep learning. Detailed Implementation

[0027] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0028] Specifically, this invention proposes a spray combustion mixing test system based on a dual-modal CT system, such as... Figure 1and Figure 2 a, Figure 2 b、 Figure 2 As shown in Figure c, the system includes an X-ray source system, a background schlieren system, a high-speed imaging system, and 3D reconstruction algorithm software. The X-ray source system serves as the physical excitation source, generating a high-energy X-ray beam to ensure penetration of a liquid film of a certain thickness. The background schlieren system provides continuous illumination for clear imaging of the vaporized mixing region. Coupled with the high-speed imaging system, the background schlieren method achieves high-precision capture of the mixing process of hydrogen in ambient gases (such as nitrogen), effectively revealing gas diffusion and mixing behavior. The high-speed imaging system uses a multi-angle detector array to capture transient projection images at an ultra-high frame rate, providing raw data input. The 3D reconstruction algorithm software employs intelligent algorithms to process massive amounts of data in real time and reconstruct a dynamic 3D concentration field. These modules work collaboratively through a precise synchronization mechanism (such as a signal synchronization generator), forming a closed-loop "excitation-acquisition-processing-protection" technology chain.

[0029] The X-ray source system is responsible for generating a high-energy, continuous, and controllable X-ray beam. Its output characteristics determine the spatial resolution and temporal response capability of the entire imaging process, and are the foundation for achieving high-speed dynamic imaging. The X-ray source system typically includes key components such as a high-voltage generator, a pulse controller, and a collimation device to ensure stable X-ray emission and directional control.

[0030] The background schlieren system provides a high-power, high-beam-quality standardized optical signal for the high-speed imaging system. The system's output characteristics are optimized to exhibit excellent beam uniformity and stability, perfectly matching the optical parameters of the high-speed imaging equipment, making it a key component for achieving high-precision dynamic flow field measurement.

[0031] The high-speed imaging system is the core sensing unit, employing a high-performance detector array capable of capturing X-ray projection images of the sample after penetration at a high frame rate in an extremely short time. Through a precise synchronous triggering mechanism, it achieves continuous recording of the dynamic mixing process, providing high-quality raw data support for subsequent 3D reconstruction. Its key technologies include high-speed readout circuitry, image intensifiers, and time-resolved imaging elements.

[0032] The described 3D reconstruction algorithm software focuses on the efficient processing and 3D visualization reconstruction of large amounts of temporal projection data acquired by high-speed imaging systems. Based on advanced mathematical modeling and image processing algorithms, the software can transform 2D projection sequences into dynamic 3D images reflecting the evolution of the sample's internal structure over time, thereby enabling quantitative analysis and in-depth research on the mixing process. The functions of the 3D reconstruction algorithm software include data preprocessing, tomographic reconstruction, motion compensation, and 3D visualization.

[0033] The technical principle of the system of this invention is as follows:

[0034] 1. X-ray spray imaging

[0035] In quantitative X-ray imaging of sprays, the measurement of mass distribution is based on the X-ray attenuation mechanism. When multicolor X-rays penetrate the spray, photons interact with the liquid medium, primarily through photoelectric absorption and Compton scattering. Photoelectric absorption dominates (accounting for approximately 80%–95% of the total attenuation), while scattered photons mainly propagate in the forward direction and have a relatively small impact on spatial resolution. This process follows a modified Beer-Lambert law:

[0036]

[0037] Where I and I0 are the intensity and initial intensity of the X-ray, respectively, and α is the X-ray attenuation coefficient. Since X-rays can pass through multiple liquid structures (droplets, ligaments, etc.) from the anode to the scintillator, the equivalent path length EPL is the sum of the individual liquid path lengths traversed by the X-ray through each structure. The attenuation coefficient α is calculated using the normalized intensity I / I0, source voltage V, and potassium iodide concentration C. KI The attenuation coefficient is a function of the normalized intensity, a consequence of beam hardening, where harder (higher energy) X-rays preferentially propagate through thicker sprays; therefore, the attenuation coefficient is actually higher over shorter path lengths than over longer path lengths. The dependence on source voltage is a result of the generated X-ray spectrum, with higher source voltages producing harder X-rays and correspondingly lower attenuation coefficients. The dependence on contrast enhancer concentration is a material property of KI, which has a higher molecular weight and higher X-ray attenuation than pure water.

[0038] Image preprocessing is performed, the detector response function is calculated, the data is normalized to a flat field, source fluctuations are corrected, background signals are removed, partial spatial blur is corrected, the transmitted data is converted to EPL, and a threshold is applied to achieve zero background for tomographic reconstruction. EPL represents the integral liquid mass along each line of sight. Therefore, the mass distribution is determined by the following formula:

[0039] EPL×ρ l ×A pixel =∫ Pat hmdl (2)

[0040] Where m is the droplet mass fraction, ρ l Let A be the density of the liquid. pixel The pixel area.

[0041] 2. Background Schlieren Imaging

[0042] Background-Oriented Schlieren Tomography (BOST) is an optical flow visualization method that combines schlieren technology and tomographic reconstruction for the quantitative measurement of three-dimensional density or refractive index fields. Its core principle is to invert the refractive index gradient distribution in the flow field by analyzing the distortion of the background pattern, thereby reconstructing the three-dimensional density field. The imaging device includes a continuous background light source and a characteristic background plate. The characteristic background plate is placed in front of the continuous light source to provide a reference for light deflection. Its calculation principle is as follows: Figure 2 As shown.

[0043]

[0044] Where n is the three-dimensional refractive index and s is the path length of light propagation. The final light deflection is obtained by integrating the refractive index field gradient along the path length, and the three-dimensional refractive index field is reconstructed based on the reconstruction algorithm.

[0045] The local distortion of the background pattern directly reflects the integral effect of the refractive index perturbation along the optical path, i.e., the "refractive index projection" on the line of sight. When the background pattern passes through a high-density droplet cloud region, the refractive index change caused by local temperature, pressure, and vapor concentration differences leads to light deflection, thus displacing the original pattern on the image plane. By comparing the displacement fields of the reference image and the perturbation image, the optical path perturbation information caused by the spray can be extracted, and tomographic inversion of the refractive index perturbation field can be achieved by combining it with a reconstruction algorithm. Since the gas refractive index n and density ρ have an approximately linear relationship under normal pressure:

[0046] n-1=K GD ·ρ (4)

[0047] Where K GD Since the refractive index is constant, BOST tomography can be further used for the three-dimensional quantitative reconstruction of density or temperature fields. Notably, unlike X-ray or optical extinction, BOST imaging does not rely on light intensity attenuation but is based on minute deflections of the light direction (typically on the order of milliradians), making it extremely sensitive to refractive index perturbations. Because the refractive index of the liquid fuel itself is much higher than that of the surrounding gas, the density gradient induced by the droplet swarms in the spray can create significant distortions in the background pattern, making it possible to visualize fuel sprays without dye labeling or photoluminescence.

[0048] 3. Multi-camera calibration

[0049] To construct X-ray CT images and background schlieren CT images, it is necessary to establish the relationship between the coordinates of points in the world coordinate system and the coordinates of points in the image coordinate system. It is important to note that to describe the position of a region of interest (VOI) in space, in addition to the relative distance from the origin of the coordinate system, the pose of the VOI is also essential. Here, the VOI pose refers to the rotational relationship between the position and the three coordinate axes of the world coordinate system, usually described by a rotation matrix. Assuming a moving coordinate system is fixed on top, the relative positional relationship between this coordinate system and the fixed reference coordinate system can be described by a displacement operator and a rotation operator. There are generally two methods for describing the pose: fixed-angle method and Euler angle method. The fixed-angle method, as the name suggests, means that the coordinate axes around which the rotation is performed are the same as the coordinate axes in the fixed reference coordinate system; while in the Euler angle method, the orientation of the coordinate axes around which each rotation is performed depends on the previous rotation.

[0050] The three coordinate systems involved in the imaging process: X w -Y w -Z w X c -Y c -Z c and X p -O p -Y p The origins are defined as the upper left corner of VOI, the optical center of the lens, and the upper left corner of the image plane, respectively. A fixed-angle approach is used to describe the relative rotation between the three coordinate systems, with the world coordinate system revolving around the three coordinate axes X and Y. w Y w Z w After rotating the pitch angle γ, yaw angle β, and gyration angle α, we obtain the coordinate system X. A -Y A -Z A X B -Y B -Z B and X D -Y D -Z D Here, all three rotations are relative to a fixed coordinate system, i.e., the initial world coordinate system. Afterwards, a translation is performed to obtain the camera coordinate system X. c -Y c -Z c This process reflects the relationship between the world coordinate system and the camera coordinate system, which can be described mathematically as follows:

[0051]

[0052] Among them, X c 'Represents the coordinates of a point in the camera coordinate system, X w' is the coordinate of a point in the world coordinate system. and T c These are the rotation and translation matrices, respectively, and are external parameters of the imaging system. Therefore, the X coordinates of each camera in the world coordinate system are... w_c That is, it is calculated using the following formula:

[0053]

[0054] in, This represents the inverse of the rotation matrix. Based on this, the minimum distance d between any two cameras in the imaging system can be obtained. min Displacement matrix T c This actually represents the coordinate system X. c -Y c -Z c In the observation coordinate system X D -Y D -Z D The relative position at that time.

[0055] When using the method described here for relative rotation, the order of the three rotations cannot be changed. This is explained mathematically. In X w -Y w Projection on the plane and X w The angle between the positive directions of the axes is defined as θ; while With Z w The angle between the positive directions of the axes is defined as Hypothesis vector Around Z w If the rotation angle of the axis is α, then the vector after rotation is... It can be represented as:

[0056]

[0057] The simplified expression above is:

[0058]

[0059] Similarly, suppose vector Around Y respectively w Rotation angle β around the axis, w If the axis is rotated by an angle γ, then the resulting vector is... and They can be represented as:

[0060]

[0061] With rotation matrix R z R y and R xRepresent the third-order matrices on the right-hand side of equations (8)-(10), as mentioned above, vectors First around the coordinate axis X w Rotate, then around the Y-axis w Finally, around the coordinate axis Z w Rotation, resulting in the final vector Expressed as a formula:

[0062]

[0063] Substituting formulas (8)-(10) into formula (11) yields formula (7). As can be seen from the above derivation process, the order of the three rotations cannot be arbitrarily changed.

[0064] The above explains the physical meaning of the imaging system's external parameters, while the calculation of the coordinates of the corresponding projection points of points within the VOI onto the image plane involves internal parameters. Based on the thin lens imaging formula and the principle of similar triangles, the coordinates (x') of any internal point transformed to the camera coordinate system are... c ,y' c ,z' c Calculate using the following formula:

[0065]

[0066] Among them, z c Represents the ideal focal plane (x) in the camera coordinate system. c ,y c ,z c The third component of the coordinates of the point on the right, f c This represents the focal length. The point coordinates (x') c ,y' c ,z ' c After performing normalization, the normalized coordinates (x, y) of the projection points on the image plane are derived. n ,y n ):

[0067]

[0068] The optical characteristics of the camera also need to be considered. For example, distortion effects caused by manufacturing errors in optical components and imaging chips can lead to deviations from the theoretical position during actual imaging. A pinhole camera model that accounts for distortion is used to further correct the coordinates of the projected points on the image plane, making them closer to their true positions. This process is expressed mathematically as follows:

[0069]

[0070]

[0071] In equation (14), r = x n 2 +y n 2 k c The distortion coefficients are a five-element vector containing radial and tangential distortion coefficients; α c is the skewness coefficient, representing the angle between the horizontal and vertical axes in the image plane coordinate system. The two terms on the right-hand side of equation (14) represent the radial distortion k. r and tangential distortion k t The second term is calculated as shown in equation (15). The projection coordinates of the point in the image plane are (x... p ,y p The 3D matrix on the right-hand side of Equation (16) consists of the camera's intrinsic parameters, including the skew coefficient α. c Focal length f c The parameters, including the principal point cc, together characterize the inherent properties of the detection perspective.

[0072] At this point, the connection between points in the world coordinate system and points on the image plane has been established. As can be seen from the above, the basis for establishing this connection is the external parameters of each viewpoint in the imaging system (i.e., the rotation matrix R). C and displacement matrix T C ) and internal parameters (i.e., skewness coefficient α) c Focal length f c Principal point cc and distortion coefficient k c These parameters are determined through the viewpoint calibration process. Viewpoint calibration is an essential and crucial step in image processing and computer vision. Through viewpoint calibration, the spatial positions of geometric feature points on the surface of an object in space and the coordinates of the corresponding projection points in the image recorded by the camera can be obtained. The prerequisite for obtaining this correspondence is establishing a correct imaging model and accurate viewpoint calibration parameters. This system uses a checkerboard-shaped calibration board for calibration and a camera calibration toolbox for viewpoint calibration, obtaining the internal and external parameters of the detection viewpoint. During the experiment, multiple projections of the calibration board at different poses need to be acquired for each viewpoint, and these projections are input into the calibration algorithm for processing. The pose of the calibration board should be diversified when acquiring projections to obtain more information.

[0073] 4. Fast reconstruction algorithm based on deep learning

[0074] like Figure 3As shown, the system of this invention can construct an end-to-end pre-trained model of a projected image to a three-dimensional flow field using a convolutional neural network. Three-dimensional mass fraction voxel models of the simulated flow field are collected, and projected images are obtained using a differentiable ray tracing algorithm. The projected image-voxel model pairs are used as the training set, and a validation set is also defined. Gradient descent algorithm combined with an Adam optimizer is used for training, and the error is observed on the validation set. Training stops when the error on the validation set tends to stabilize. The loss function is as follows:

[0075]

[0076] in, For the prediction of the three-dimensional spray mass fraction field of the forward process, V gt This is a three-dimensional spray mass fraction field for simulating the flow field.

[0077] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A spray combustion mixing test system based on a dual-modal CT system, characterized in that, The system includes an X-ray source system, a background schlieren system, a high-speed imaging system, and three-dimensional reconstruction algorithm software; the X-ray source system serves as a physical excitation source to generate a high-energy X-ray beam, ensuring penetration of a liquid film with a certain thickness. The background schlieren system is used to clearly image the mixed region after vaporization. The background schlieren system is coupled with the high-speed imaging system to provide the high-power, high-beam-quality standardized optical signal to the high-speed imaging system. The high-speed imaging system captures transient projection images at an ultra-high frame rate through a multi-angle detector array, providing raw data input. The three-dimensional reconstruction algorithm software uses intelligent algorithms to process data in real time and reconstruct the dynamic three-dimensional concentration field.

2. The spray combustion mixing test system based on a dual-modal CT system according to claim 1, characterized in that, The X-ray source system includes a high-voltage generator, a pulse controller, and a collimation device to ensure stable X-ray emission and directional control.

3. The spray combustion mixing test system based on a dual-modal CT system according to claim 1, characterized in that, The background schlieren system is perfectly matched with the optical parameters of the high-speed imaging equipment, and is used to achieve high-precision dynamic flow field measurement.

4. The spray combustion mixing test system based on a dual-modal CT system according to claim 1, characterized in that, The high-speed imaging system is the core sensing unit, employing a high-performance detector array that can capture X-ray projection images of the sample after it has penetrated the sample in a very short time at a high frame rate. Through a precise synchronous triggering mechanism, it enables continuous recording of the dynamic mixing process, providing raw data support for three-dimensional reconstruction.

5. The spray combustion mixing test system based on a dual-modal CT system according to claim 1, characterized in that, The three-dimensional reconstruction algorithm software is used to process and reconstruct three-dimensional visualization data acquired by the high-speed imaging system. Based on mathematical modeling and image processing algorithms, the software transforms the two-dimensional projection sequence into a dynamic three-dimensional image that reflects the evolution of the internal structure of the sample over time, thereby enabling quantitative analysis and in-depth study of the mixing process.

6. The spray combustion mixing test system based on a dual-modal CT system according to claim 1, characterized in that, The functions of the 3D reconstruction algorithm software include data preprocessing, fault reconstruction, motion compensation, and 3D visualization.

Images