Three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging

Abstract

The invention discloses a three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging, solves the problem of poor precision of three-dimensional and multi-parameter diagnosis of a complex combustion field by the existing combustion field diagnosis technology, and belongs to the technical field of flame combustion detection. According to the application, the visible hyperspectral light field camera and the infrared hyperspectral light field camera are combined, the limitation of a single technology is broken through, and high-precision synchronous measurement of multiple parameters of a combustion field is realized; through a hyperspectral light field single-time three-dimensional imaging technology, three-dimensional space information and spectral information of a combustion field are obtained at the same time in single-time imaging, and the measurement efficiency and the spatial resolution are remarkably improved; spectral information and light field imaging are deeply fused, rich spectral data support is provided for three-dimensional reconstruction, and the problem of measurement errors caused by insufficient spectral information in a traditional method is solved; based on a multispectral radiation source term matrix and an advanced inversion algorithm, high-precision inversion of a three-dimensional temperature field, a particle concentration field and a gas concentration field is realized.

Description

Technical Field

[0001] This application relates to a three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging, belonging to the field of flame combustion detection technology. Background Technology

[0002] Combustion, as the fundamental form of converting fossil fuels into heat, propulsion, and electricity, is widely used in aerospace, energy and power, power, and transportation. Although combustion provides 85% of the world's energy and over 60% of its electricity, the process involves complex coupled phenomena such as multiphase flow, heat and mass transfer, and chemical reactions, exhibiting characteristics of three-dimensionality, high dynamics, and multi-scale processes. Therefore, in-depth research into combustion mechanisms and effective control of the combustion process are crucial for improving energy conversion efficiency, reducing pollutant emissions, and optimizing equipment design.

[0003] Accurate measurement of combustion parameters (such as temperature, pressure, velocity, particle concentration, and gaseous component concentration) is crucial for understanding the nature of combustion. Temperature affects reaction rates and energy release, component concentration helps analyze pollutant formation, and flow velocity is closely related to flow field stability and mass transport. Furthermore, optical imaging of free radicals can reveal flame structure and reaction locations, providing important information for combustion control and pollutant analysis. However, traditional invasive measurement techniques (such as thermocouples and hot-wire anemometers) suffer from drawbacks such as flow field interference and low spatial resolution, making it difficult to meet the demands for high-precision, high-dynamic measurements.

[0004] Non-invasive optical measurement techniques, especially diagnostic methods based on laser spectroscopy and emission spectroscopy, have become the preferred solution for measuring combustion flow field parameters due to their high response speed, high spatial resolution, and spectral resolution. To comprehensively monitor the temperature field, particle concentration field, gas concentration field, and intermediate products during combustion, there is an urgent need to develop an efficient measurement method capable of simultaneously retrieving three-dimensional temperature, concentration, and particle information.

[0005] Optical diagnostic techniques based on emission spectroscopy analyze the electromagnetic radiation emitted by the combustion flame itself to infer the composition, concentration distribution, and flow field characteristics of the flame. The spectral signals mainly originate from the following processes: (1) chemiluminescence generated by the transition from the excited state to the ground state during combustion; (2) wavelength-selective emission bands formed by the rotational energy level transitions of high-temperature gas molecules; and (3) continuous blackbody radiation generated by soot or carbon particles. This technique has significant advantages: (1) it eliminates the need for external probes or excitation devices, avoiding interference with the flow field and ensuring measurement accuracy; (2) combined with spectral analysis, it can simultaneously acquire two-dimensional or three-dimensional distributions of multiple parameters such as temperature and multi-component concentrations; (3) it can capture dynamic changes and transient behavior of the flow field; and (4) compared to laser spectroscopy, the equipment structure is simpler and the cost is lower. Therefore, emission spectroscopy diagnostic methods have been widely used in combustion research.

[0006] Multispectral or hyperspectral imaging techniques overcome the limitations of single-wavelength diagnostics by simultaneously detecting emission spectra across multiple wavelength ranges, enabling the simultaneous acquisition of information such as flame temperature and component concentration. However, visible light hyperspectral cameras can only measure the flame temperature field, particle concentration field, and some furnace thermal radiation parameters, and cannot directly acquire gas concentration data, requiring supplementation with infrared hyperspectral images. Traditional flame imaging typically acquires only a single image or an RGB three-color image, containing only spatial dimension information and lacking detailed spectral dimension information (RGB images only reflect the flame state at three specific wavelengths). Furthermore, achieving three-dimensional imaging is technically complex, using multiple cameras is expensive, and methods such as sliding rail movement cannot guarantee real-time performance. Methods such as fiber optic splitting may reduce resolution and signal strength, and are difficult to provide a full-space window in devices such as engines. Summary of the Invention

[0007] To address the issue of poor accuracy in the three-dimensional and multi-parameter diagnosis of complex combustion fields by existing combustion field diagnostic technologies, this application provides a three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging.

[0008] The three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging in this application includes:

[0009] S1. Simultaneously capture the combustion flow field using a visible hyperspectral light field camera and an infrared hyperspectral light field camera to obtain visible light field images and infrared light field images of the flame field.

[0010] S2. Perform blackbody radiation calibration on the acquired visible light field image and infrared light field image to obtain the blackbody radiation intensity curve of the flame.

[0011] Extract the spectral grayscale values ​​of the flame image from the acquired visible light field image and infrared light field image;

[0012] The spectral radiation intensity of the flame is obtained by marking the correspondence between the spectral gray values ​​of the flame image and the blackbody radiation intensity in the blackbody radiation intensity curve.

[0013] S3. Based on the spectral radiation intensity of the flame and the radiation transfer model of the combustion flame, the three-dimensional space of the combustion field is discretized into N grids, a spatial grid model is established, and the radiation intensity received by each detection line in the N grids is determined.

[0014] S4. Based on the relationship between the radiation intensity received by the N grids of all detector lines and the radiation source terms on the detector lines, establish a light field convolution imaging model, solve the light field convolution imaging model, and obtain the radiation source terms for each wavelength.

[0015] S5. Using the relationship between the obtained radiation source term and the combustion parameters, the three-dimensional distribution map of the combustion parameters is derived.

[0016] As a preferred option, the light field convolution imaging model is:

[0017]

[0018] In the formula, It is a matrix of probe line lengths in N grids; This is the radiation intensity matrix received by the N grids of all probe lines.

[0019] As a preferred approach, the Tikhonov regularization algorithm is used to solve the optical field convolution imaging model to obtain the radiation source term for each wavelength.

[0020] As a preferred option, the radiation intensity received by a single optical field detection line is:

[0021]

[0022] in, Indicates the detection line The radiation intensity received on the corresponding N grids,

[0023] This represents the radiation source term corresponding to the i-th grid.

[0024] The point spread function represents the optical imaging system.

[0025] Indicates the first The optical thickness of each grid.

[0026] As a preferred embodiment, S5 includes:

[0027] Based on the obtained radiation source term, the temperature of the soot particles, the absorption coefficient of the soot particles, and the gas absorption coefficient are solved by using the characteristic absorption peak wavelength of the gas.

[0028] Based on the solved soot particle temperature, a three-dimensional distribution map of the combustion flow field temperature is obtained;

[0029] Based on the Mie scattering theory, the concentration of soot particles is inverted according to the absorption coefficient of soot particles, and a three-dimensional distribution map of soot particle concentration is obtained.

[0030] Using a statistical narrowband model or line-by-line integration method, the three-dimensional distribution map of gas concentration can be derived from the gas absorption coefficient.

[0031] As a preferred embodiment, the method of this application also includes:

[0032] The three-dimensional distribution maps of the combustion flow field temperature, the soot particle concentration, and the gas concentration, or the fused three-dimensional multi-parameter distribution map of the combustion field, are displayed using visualization technology.

[0033] As a preferred method, the calculation methods for the absorption coefficient of black smoke particles and the gas absorption coefficient are as follows:

[0034] In the band without gas absorption peaks:

[0035]

[0036] In the formula is the monochromatic absorption coefficient of the flame. The absorption coefficient of black smoke particles;

[0037]

[0038] In the formula The first radiation constant, For wavelength, The second radiation constant, Let the temperature be the temperature of the nth spatial grid. The radiation intensity received by the nth grid;

[0039] Within the band containing gas absorption peaks:

[0040]

[0041] In the formula The gas absorption coefficient;

[0042] .

[0043] The beneficial effects of this application are as follows: This application combines a visible hyperspectral light field camera and an infrared hyperspectral light field camera, overcoming the limitations of single technologies and achieving high-precision synchronous measurement of multiple parameters of the combustion field; through hyperspectral light field single-shot three-dimensional imaging technology, it simultaneously acquires three-dimensional spatial and spectral information of the combustion field in a single imaging, significantly improving measurement efficiency and spatial resolution; it deeply integrates spectral information with light field imaging, providing rich spectral data support for three-dimensional reconstruction and solving the measurement error problem caused by insufficient spectral information in traditional methods; based on the multispectral radiation source term matrix and advanced inversion algorithm, it achieves high-precision inversion of the three-dimensional temperature field, particle concentration field, and gas concentration field; through the collaborative work of the synchronous controller and data processing system, it generates a three-dimensional multi-parameter distribution map of the combustion field in real time, providing immediate data support for combustion process analysis and optimization, which has important scientific significance and engineering application value. Attached Figure Description

[0044] Figure 1This is a flowchart of the method described in this application;

[0045] Figure 2 This is a schematic diagram of the inversion system of this application;

[0046] Figure 3 A schematic diagram illustrating the working principle of a hyperspectral light field camera with adjustable wavelengths;

[0047] Figure 4 Here is a schematic diagram of a hyperspectral light field camera: (a) is a schematic diagram of the overall structure model; (b) is a cross-sectional view of the structure.

[0048] Figure 5 A schematic diagram of a hyperspectral light field camera with a filter matrix;

[0049] Figure 6 This is a schematic diagram of a hyperspectral light field camera with a grating. Detailed Implementation

[0050] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0051] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0052] The present application will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the application.

[0053] The three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging in this application includes:

[0054] Step 1: Simultaneously capture the combustion flow field using a visible hyperspectral light field camera and an infrared hyperspectral light field camera to obtain visible light field images and infrared light field images of the flame field;

[0055] The visible light field images and infrared light field images of the flame field in this application need to be acquired synchronously to ensure temporal and spatial consistency. Based on the requirements of the parameters to be measured (temperature, particle concentration, gas concentration), visible light band filter parameters are set for the visible light hyperspectral light field camera, and infrared band filter parameters are set for the infrared hyperspectral light field camera to cover the target spectral range. The two cameras are synchronously calibrated to eliminate systematic errors between the devices and ensure data consistency.

[0056] Acquire visible light field image and Infrared light field image This is used to simultaneously solve for the temperature field, particle concentration field, and gas concentration field. Visible light field images are used in common alkane combustion. and Assuming it only contains particle absorption, the temperature field and particle concentration field can be obtained. Infrared light field image. The absorption of both gas and particles can be combined to separate and solve the gas absorption coefficient field, and the gas concentration field can be obtained.

[0057] Step 2: Perform blackbody radiation calibration on the acquired visible light field image and infrared light field image to obtain the blackbody radiation intensity curve of the flame; extract the spectral gray value of the flame image from the acquired visible light field image and infrared light field image; mark the flame spectral radiation intensity according to the correspondence between the spectral gray value of the flame image and the blackbody radiation intensity in the blackbody radiation intensity curve.

[0058] Specifically, the spectral radiation intensity of the flame is obtained using the blackbody radiation law:

[0059] (1)

[0060] Where h is Planck's constant, c is the speed of light, λ is the wavelength, k is Boltzmann's constant, and T is the absolute temperature of the blackbody furnace. The intensity of blackbody radiation;

[0061] Extracting visible light field images and infrared light field images The grayscale values ​​are then labeled according to the one-to-one correspondence between blackbody radiation intensity and the grayscale values ​​of the flame image, thereby obtaining the spectral radiation intensity. ;

[0062] Step 4: Based on the spectral radiation intensity of the flame and the radiation transfer model of the combustion flame, discretize the three-dimensional space of the combustion field into N grids, establish a spatial grid model, and determine the radiation intensity received by each detection line in the N grids.

[0063] The radiative transfer model of a combustion flame is as follows:

[0064] (2)

[0065] in, For direction and location Spectral radiation intensity on For direction and location Spectral radiation intensity on For direction and location Spectral radiation intensity on For direction and The scattering phase function between them Positions The extinction coefficient, absorption coefficient, and scattering coefficient of the light.

[0066] For high-temperature flames, the flame temperature is much higher than the ambient temperature, and the radiation intensity around the medium can be ignored, i.e., the incident radiation intensity. Integrating formula (1) and discretizing it using the discrete coordinate method, we obtain the discretized calculation formula for directional radiation intensity:

[0067] (3)

[0068] (4)

[0069] In the formula, This represents the number of grid cells the probe line passes through. The total radiation intensity, For the radiation source term at the end of the detection line, Total optical thickness In the first The source item of the grid, For the first Optical thickness of the grid. For the first The scattering albedo of the grid.

[0070] For common hydrocarbon flames such as methane and ethylene, the size of the solid particles produced is smaller than the wavelength of the incident radiation, resulting in an absorption rate much greater than a scattering rate. Therefore, they can be treated as pure absorbing media, considering only the emission and absorption processes and ignoring the scattering phase.

[0071] Combustion flow field measurement based on light field imaging is an advanced non-contact optical detection technology. By using a light field camera to capture the radiation intensity emitted by the flame itself, it enables 3D tomographic reconstruction of the combustion flow field and accurate measurement of the temperature field.

[0072] In step one, during the light field imaging process, the coordinates of the intersection point between the virtual image plane and the probe line are calculated using formula (5). Using formula (6) Convert to The position of the light ray on the front optical imaging system of the main lens is obtained. Then, formula (7) is used to... Convert to This yields the coordinates of the intersection points of the light rays on the virtual image plane.

[0073] (5)

[0074] (6)

[0075] (7)

[0076] In the formula, and These represent the distances from the main lens front optical imaging system to the MLA plane and from the MLA plane to the CCD sensor, respectively. and as well as These represent the distance from the standard object plane to the main lens front optical imaging system, the distance from the main lens front optical imaging system to the virtual image plane, and the distance from the virtual image plane to the MLA plane, respectively. and These are the coordinates of the intersection of the MLA plane and the CCD sensor, respectively.

[0077] Taking into account the radiation transfer mechanism, the characteristics of the optical imaging system, and the principle of light field convolution imaging, a combustion flow field measurement model based on light field imaging is constructed to achieve accurate description and 3D tomographic reconstruction of the complex radiation transfer process inside the flame. The two-dimensional projection image received by the optical imaging system... It can be represented as:

[0078] (8)

[0079] Equation (8) represents the two-dimensional projected image received by the optical imaging system. It is an ideal image function with multiple cross sections. and response function The convolutional stacking. Indicates spatial location The brightness distribution function at that location, Let be the point source response (i.e., the point spread function) of the optical imaging system. Discretizing the continuous integral form, it can be expressed as:

[0080] (9)

[0081] Equation (9) represents the spatially continuous combustion flow field divided into: The discrete form after N grids. Combining the radiation transfer mechanism of the combustion flow field and the light field convolution imaging model, the radiation intensity received by a single probe line in N grids is:

[0082] (10)

[0083] Formula (9) combines the radiation transfer mechanism of the combustion flow field and the light field convolution imaging model to calculate the radiation intensity received on a single pixel. .in and Let represent the radiation source term and point spread function on the detection line, respectively; This indicates the optical thickness along the detection path.

[0084] Step 4: Establish a light field convolution imaging model based on the relationship between the radiation intensity received by the N grids of all detector lines and the radiation source terms on the detector lines. Solve the light field convolution imaging model to obtain the radiation source terms for each wavelength.

[0085] Radiation source term The matrix equation is:

[0086] (11)

[0087] The mesh was calculated based on Wien's law. Monochromatic blackbody radiation intensity :

[0088] (12)

[0089] The equation for the energy of monochromatic blackbody radiation is transformed into a system of equations:

[0090] (13)

[0091] In the formula, It is a probe line In the grid The length of the probe line in the middle, It is a grid absorption coefficient, It is a grid The monochromatic blackbody radiation intensity, Given the total number of discrete directions in three-dimensional space, the system of equations can be transformed into matrix equation form as follows:

[0092] (14)

[0093] In the formula, It is a matrix of probe line lengths in N grids; This is the radiation intensity matrix received by the N grids of all probe lines.

[0094] Equation (14) is used as the light field convolution imaging model;

[0095] The Tikhonov regularization algorithm is used to solve the matrix equation to obtain the radiation source term for each wavelength. .

[0096] Step 5: Using the relationship between the obtained radiation source term and combustion parameters, the three-dimensional distribution map of combustion parameters is derived.

[0097] Based on the obtained radiation source term, the temperature of the soot particles, the absorption coefficient of the soot particles, and the gas absorption coefficient are solved by using the characteristic absorption peak wavelength of the gas.

[0098] Based on the solved soot particle temperature, a three-dimensional distribution map of the combustion flow field temperature is obtained;

[0099] Based on Mie scattering theory, the concentration of black smoke particles is inverted according to the absorption coefficient of black smoke particles, and then corrected by combining the temperature field to obtain a three-dimensional distribution map of the concentration of black smoke particles.

[0100] Using the statistical narrow-band model (SNB) or the line-by-line integration method (LBL), the three-dimensional distribution map of gas concentration (such as CO2, H2O, etc.) can be derived from the gas absorption coefficient.

[0101] This application uses visualization technology to display the three-dimensional distribution map of the combustion flow field temperature, the three-dimensional distribution map of the soot particle concentration, and the three-dimensional distribution map of the gas concentration, or the three-dimensional multi-parameter distribution map of the fused combustion field. The visualization technology displays the spatial distribution of temperature, particle concentration, and gas concentration in the combustion field, providing data support for combustion process analysis and optimization.

[0102] In this application, the calculation methods for the absorption coefficient of black smoke particles and the gas absorption coefficient are as follows:

[0103] In the band without gas absorption peaks:

[0104] (15)

[0105] In the formula is the monochromatic absorption coefficient of the flame. The absorption coefficient of black smoke particles;

[0106] (16)

[0107] In the formula The first radiation constant, For wavelength, The second radiation constant, Let the temperature be the temperature of the nth spatial grid. The radiation intensity received by the nth grid;

[0108] Within the band containing gas absorption peaks:

[0109] (17)

[0110] In the formula The gas absorption coefficient;

[0111] (18)

[0112] The combustion field three-dimensional multi-parameter field inversion system of this application includes: a visible hyperspectral light field camera 1, an infrared hyperspectral light field camera 2, a synchronous controller 3, and a data processing system 4;

[0113] The visible hyperspectral light field camera 1 needs to be placed in a location that can clearly capture the visible light radiation of the combustion field, maintaining a suitable distance and angle from the combustion field, with its height roughly level with the center of the combustion field to avoid obstruction. A suitable visible light filter should be selected based on the measurement parameters, and the exposure time and gain should be adjusted according to the brightness of the combustion field and the camera sensitivity to ensure that the image is neither too dark nor too overexposed, clearly capturing the details of the combustion field while minimizing noise.

[0114] The infrared hyperspectral light field camera 2 should be placed in conjunction with the visible hyperspectral light field camera 1 to avoid mutual interference. At the same time, the position should be determined by considering the infrared radiation distribution of the combustion field and avoiding thermal interference from the surrounding environment. It is preferable that the optical axes of the visible hyperspectral light field camera and the infrared hyperspectral light field camera are placed perpendicularly.

[0115] Traditional single-lens designs often fail to meet the requirements of multispectral sampling, and their inability to guarantee parallel light incidence on the filter can introduce additional aberrations, affecting image quality. Tunable filters also have limitations on the angle of incidence; an excessively large angle can negatively impact filter performance. To overcome these issues, this application designs a wavelength-tunable hyperspectral imager employing a dual-lens design. The front lens group uses a retrograde design, with the tunable filter strategically positioned between the front and rear lens groups. This design effectively avoids the aforementioned problems while ensuring compliance with the filter's angle of incidence and the optical system's F-number design requirements. A schematic diagram of the wavelength-tunable visible and infrared hyperspectral light field camera is shown below. Figure 3 As shown, the main components include a front optical system, a tunable filter, a rear optical system, and a camera. A schematic diagram of the visible and infrared hyperspectral light field camera, along with its overall structural model and cross-sectional view, are shown below. Figure 4 As shown. The light emitted from the object surface passes through the front optical system, where the wide field of view light is compressed into a collimated beam within the incident range of the filter, ensuring the normal operation of the filter; after passing through the tunable filter, specific wavelengths of light are filtered out, and then imaged onto the target surface of the camera by the rear optical system.

[0116] Select a suitable tunable filter for measuring the infrared radiation of the combustion field to obtain specific wavelength information. Temperature calibration with a standard blackbody is required before use to ensure measurement accuracy. Specifically, the wavelength of the tunable filter can be adjusted using a wavelength modulator, filter matrix, or grating.

[0117] Figure 5This diagram illustrates a visible and infrared hyperspectral light field camera with a filter matrix. It addresses the issue of slow data acquisition due to multiple acquisitions required by liquid crystal tunable filters (LCTFs), enabling the simultaneous acquisition of multiple wavelengths. By integrating multiple specific wavelength filters arranged in a matrix in front of the camera sensor, this camera achieves simultaneous acquisition of multi-wavelength information, overcoming the slow acquisition speed and multiple tuning requirements of traditional LCTFs. This technology significantly improves data acquisition efficiency, meets real-time requirements, and avoids errors introduced by dynamic changes in the combustion field, thus enhancing measurement accuracy. Furthermore, the filter matrix is ​​directly integrated into the camera, eliminating the need for additional tuning equipment, simplifying the system structure and reducing costs. Through careful design of the filter wavelength range and bandwidth, this technology achieves high spectral resolution, meeting the needs of multi-parameter combustion field measurements. In combustion field measurements, this camera can quickly acquire key spectral data of temperature, particle concentration, and gas concentration fields; in industrial process monitoring and environmental monitoring, it can also be used for real-time monitoring of temperature distribution, gas concentration distribution, and pollutant emissions. Although the technology still faces challenges in filter design optimization, sensor performance improvement, and data processing algorithm development, its broad application prospects in combustion field measurement, industrial process monitoring, and environmental monitoring provide efficient and accurate technical support for multi-parameter field inversion, and have important scientific significance and engineering application value.

[0118] Figure 6 This diagram illustrates a visible and infrared hyperspectral light field camera with a grating. It effectively addresses the slow speed and multiple acquisitions associated with traditional liquid crystal tunable filters (LCTFs), enabling simultaneous acquisition of multiple wavelengths. The grating-equipped camera integrates grating-based spectral splitting technology to separate and focus incident light onto the sensor according to wavelength, achieving simultaneous acquisition of multi-wavelength information. This design significantly improves data acquisition efficiency, meets real-time requirements, and avoids errors introduced by multiple acquisitions in dynamic scenes, thus enhancing measurement accuracy. Furthermore, grating-based spectral splitting technology offers wide spectral coverage and high spectral resolution, accurately acquiring rich spectral information from the visible to infrared bands. Compared to traditional filter matrices, the grating-equipped camera structure is simpler, lower in cost, and requires no complex mechanical tuning, resulting in higher stability and reliability. In fields such as combustion field measurement, industrial process monitoring, and environmental monitoring, this camera can rapidly acquire multi-parameter spectral data such as temperature, particle concentration, and gas concentration fields, providing efficient and accurate technical support for multi-parameter field inversion in complex scenarios. Overall, the visible and infrared hyperspectral light field camera with grating exhibits significant advantages in multi-wavelength acquisition efficiency, spectral resolution, system stability, and wide applicability, and has important scientific significance and engineering application value.

[0119] Synchronization controller 3 should be placed close to the visible and infrared hyperspectral light field cameras, avoiding proximity to strong electromagnetic interference sources. Select the synchronization mode (hardware or software synchronization) according to the camera's operating mode and measurement requirements, and adjust the synchronization accuracy to keep the acquisition time error between the two cameras within acceptable limits, ensuring coordinated operation of the two cameras.

[0120] The data processing system 4 is used to implement the three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging of this application through a computer program.

[0121] While this application has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of this application. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of this application as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging, characterized in that, include: S1. Simultaneously capture the combustion flow field using a visible hyperspectral light field camera and an infrared hyperspectral light field camera to obtain visible light field images and infrared light field images of the flame field. S2. Perform blackbody radiation calibration on the acquired visible light field image and infrared light field image to obtain the blackbody radiation intensity curve of the flame. Extract the spectral grayscale values ​​of the flame image from the acquired visible light field image and infrared light field image; The spectral radiation intensity of the flame is obtained by marking the correspondence between the spectral gray values ​​of the flame image and the blackbody radiation intensity in the blackbody radiation intensity curve. S3. Based on the spectral radiation intensity of the flame and the radiation transfer model of the combustion flame, the three-dimensional space of the combustion field is discretized into N grids, a spatial grid model is established, and the radiation intensity received by each detection line in the N grids is determined. S4. Based on the relationship between the radiation intensity received by the N grids of all detector lines and the radiation source terms on the detector lines, establish a light field convolution imaging model, solve the light field convolution imaging model, and obtain the radiation source terms for each wavelength. S5. Using the relationship between the obtained radiation source term and the combustion parameters, the three-dimensional distribution map of the combustion parameters is derived.

2. The three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging according to claim 1, characterized in that, The light field convolution imaging model is as follows: In the formula, It is a matrix of probe line lengths in N grids; This is the radiation intensity matrix received by the N grids of all probe lines.

3. The three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging according to claim 2, characterized in that, The Tikhonov regularization algorithm is used to solve the optical field convolution imaging model to obtain the radiation source term at each wavelength.

4. The three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging according to claim 2, characterized in that, The radiation intensity received by a single optical field detection line is: in, Indicates the detection line The radiation intensity received on the corresponding N grids, This represents the radiation source term corresponding to the i-th grid. The point spread function represents the optical imaging system. Indicates the first The optical thickness of each grid.

5. The three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging according to claim 1, characterized in that, S5 include: Based on the obtained radiation source term, the temperature of the soot particles, the absorption coefficient of the soot particles, and the gas absorption coefficient are solved by using the characteristic absorption peak wavelength of the gas. Based on the solved soot particle temperature, a three-dimensional distribution map of the combustion flow field temperature is obtained; Based on the Mie scattering theory, the concentration of soot particles is inverted according to the absorption coefficient of soot particles, and a three-dimensional distribution map of soot particle concentration is obtained. Using a statistical narrowband model or line-by-line integration method, the three-dimensional distribution map of gas concentration can be derived from the gas absorption coefficient.

6. The three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging according to claim 5, characterized in that, The method further includes: The three-dimensional distribution maps of the combustion flow field temperature, the soot particle concentration, and the gas concentration, or the fused three-dimensional multi-parameter distribution map of the combustion field, are displayed using visualization technology.

7. The three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging according to claim 5, characterized in that, The calculation methods for the absorption coefficient of black smoke particles and the gas absorption coefficient are as follows: In the band without gas absorption peaks: In the formula is the monochromatic absorption coefficient of the flame. The absorption coefficient of black smoke particles; In the formula The first radiation constant, For wavelength, The second radiation constant, Let the temperature be the temperature of the nth spatial grid. The radiation intensity received by the nth grid; Within the band containing gas absorption peaks: In the formula The gas absorption coefficient; 。 8. The three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging according to claim 1, characterized in that, The radiative transfer model of a combustion flame is as follows: in, For direction and location Spectral radiation intensity on For direction and location The spectral radiation intensity of the blackbody on the surface, For direction and location Spectral radiation intensity on For direction and The scattering phase function between them Positions The extinction coefficient, absorption coefficient, and scattering coefficient of the light.

9. The three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging according to claim 1, characterized in that, The visible hyperspectral light field camera and the infrared hyperspectral light field camera use wavelength modulators, filter matrices or gratings to adjust the wavelength.

10. The three-dimensional combustion multi-parameter field inversion method based on visible and infrared hyperspectral light field imaging according to claim 1, characterized in that, The optical axes of the visible hyperspectral light field camera and the infrared hyperspectral light field camera are perpendicular.

Images