A method for reconstructing radiation spectra of metal surfaces

Abstract

This invention discloses a method for processing radiation spectral reconstruction data of a metal surface, comprising the following steps: First, spectral data of the molten metal surface temperature is acquired using a spectral chip camera and a fiber optic spectrometer. The acquisition process includes: adjusting the optical path to ensure all devices are coaxial and at the same height during acquisition, while ensuring the location of the spectral camera's detection point is on the axis of symmetry of the molten metal. Then, the spectral data is processed to reduce thermal noise using the spectral camera, and spectral images of the molten metal are captured for computer acquisition. This invention utilizes a spectral image reconstruction method for spectral reconstruction and an integrated spectral camera temperature measurement model to achieve temperature inversion. Compared to various current temperature field reconstruction algorithms, the multispectral camera radiation thermometry method proposed in this invention can achieve high stability while ensuring the accuracy of temperature inversion, and can better reflect the actual temperature distribution.

Description

Technical Field

[0001] This invention relates to the technical field of spectral reconstruction data processing, and more particularly to a method for processing radiation spectral reconstruction data of metal surfaces. Background Technology

[0002] In furnaces such as converters and AOD furnaces for refining molten metal, the temperature of the molten metal is continuously and accurately measured. Using this information as operational data is extremely useful for improving refining efficiency, quality, yield, and reducing resource consumption per unit. Continuously monitoring the molten metal temperature and controlling it as a temperature progression profile for each steel grade is crucial for effective refining.

[0003] Currently, many temperature measurement technologies in the industry are still limited to single-point or limited multi-point measurements. For example, while infrared thermal imagers can quickly obtain temperature distribution images of solid targets, traditional fiber optic spectrometers can only measure the spectrum of a limited number of points, either single-point or line-by-line, to retrieve the temperature. This is no longer sufficient to meet current practical needs. Furthermore, during the acquisition and processing of spectral data, baseline tilting and drift problems arise due to sample size, instrument background, and other related factors. Additionally, interference from molten metal can cause overlapping spectral lines. To address these issues, we propose a data processing method for reconstructing the radiation spectrum of metal surfaces. Summary of the Invention

[0004] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0005] In view of the problems existing in the above-mentioned methods for processing data for reconstructing radiation spectra of metal surfaces, this invention is proposed.

[0006] Therefore, the purpose of this invention is to provide a data processing method for reconstructing radiation spectra of metal surfaces. This method utilizes spectral image reconstruction to perform spectral reconstruction and uses an integrated spectral camera temperature measurement model to achieve temperature inversion. Compared with various current temperature field reconstruction algorithms, the multispectral camera radiation temperature measurement method proposed in this invention can achieve higher stability while ensuring the accuracy of temperature inversion, and can better reflect the actual temperature distribution.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for processing radiation spectrum reconstruction data of metal surfaces, comprising the following steps:

[0008] S1. First, spectral data of the surface temperature of the molten metal is acquired using a spectral chip camera and a fiber optic spectrometer. The acquisition process includes: adjusting the optical path to ensure that all devices are coaxial and at the same height during the acquisition process, while ensuring that the position of the spectral camera detection point is on the axis of symmetry of the molten metal. Then, the spectral data is processed to reduce thermal noise and the spectral image of the molten metal is captured and captured for computer acquisition.

[0009] S2. Subsequently, the collected spectral data is preprocessed to improve measurement accuracy;

[0010] S3. Background correction is applied to the obtained spectral data, and spectral reconstruction is achieved by using the spectral raw image reconstruction method.

[0011] S4. Spectral data is acquired again using the fiber optic probe of the spectrometer, and temperature inversion is performed to verify the accuracy of the temperature inverted by the spectrometer camera and determine the error.

[0012] In a preferred embodiment of the metal surface radiation spectrum reconstruction data processing method of the present invention, wherein: in step S2, the derivative algorithm is used to process the overlap of the acquired spectral lines during the spectral data preprocessing process.

[0013]

[0014]

[0015] Among them, S i S represents the spectral value at the i-th point of the spectrum in the original sample. i ′ With S i "" represents the first and second derivatives with respect to the original spectral value, respectively; w is the width of the current spectral region.

[0016] As a preferred embodiment of the metal surface radiation spectrum reconstruction data processing method of the present invention, the derivative algorithm is followed by a multivariate scattering correction process to average the obtained spectra at each position; then, the spectral data at each position is subjected to a univariate linear regression operation with the average spectrum to obtain its regression coefficient and regression constant; then, the regression constants of each initial spectrum are subtracted and divided by their regression coefficients.

[0017] As a preferred embodiment of the metal surface radiation spectrum reconstruction data processing method of the present invention, the method of spectral raw image reconstruction in step S3 includes: extending the image gradient correlation parameters contained in the neighborhood sampling points to the unsampled points in the spectral image. Let the location of the unsampled spectral image point be b, and the corresponding sampling point contained in its neighborhood be a. Based on extending the known gradient information from point a to point b, the unsampled points are estimated.

[0018]

[0019] In the formula: f n (x) is the nth derivative of the current function expression; R n (x) represents the Lagrange remainder term.

[0020] As a preferred embodiment of the metal surface radiation spectrum reconstruction data processing method of the present invention, the two-dimensional spectral image can be reconstructed using the gradient-related information of the sampled spectral image points, and the direction vector from a to b is defined as:

[0021]

[0022] In the formula: |ab| represents the distance between a and b in the current two-dimensional space, from which its directional derivative can be obtained as:

[0023]

[0024] It is possible to estimate b based on the directional derivative and gradient, given a:

[0025]

[0026] Reconstructing unsampled points b in the neighborhood of a known sampled point a is represented as follows:

[0027]

[0028] In the formula: a i ∈W s For window W in the spectral image s Sample points already collected; For unsampled point a i The binary mask at the location and For a i The kernel weight value at its location; For from a i The contribution value to b, where the kernel weight Represented as:

[0029]

[0030] Geometric distance is selected as the weight for calculation.

[0031] As a preferred embodiment of the metal surface radiation spectrum reconstruction data processing method of the present invention, in step S4, the energy radiated outward from the surface of the molten metal to be tested is captured as an image signal after passing through filters with different center wavelengths, and the radiation information at a specific wavelength is obtained. The obtained signal is then transmitted to a computer for calibration and calculation to obtain the temperature distribution of the surface of the molten metal to be tested. At the same time, the spectrum of the molten metal is directly measured using a fiber optic spectrometer, and the actual temperature data is obtained by temperature inversion, thus verifying the temperature measurement accuracy of the temperature field reconstruction method of the spectroscopic camera.

[0032] The beneficial effects of this invention are as follows: This invention utilizes a spectral image reconstruction method to perform spectral reconstruction and uses an integrated spectral camera temperature measurement model to achieve temperature inversion; compared with various current temperature field reconstruction algorithms, the multispectral camera radiation temperature measurement method proposed in this invention can achieve higher stability while ensuring the accuracy of temperature inversion, and can better reflect the actual temperature distribution. Attached Figure Description

[0033] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0034] Figure 1 This is a schematic diagram of the method steps for processing data of metal surface radiation spectrum reconstruction according to the present invention. Detailed Implementation

[0035] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0036] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0037] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0038] Secondly, the present invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of the present invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not according to the usual scale. Furthermore, the schematic diagrams are merely examples and should not limit the scope of protection of the present invention. In addition, actual fabrication should include three-dimensional spatial dimensions of length, width, and depth.

[0039] Reference Figure 1 A method for processing radiation spectrum reconstruction data of metal surfaces is provided, including the following steps:

[0040] S1. First, spectral data of the surface temperature of the molten metal is acquired using a spectral chip camera and a fiber optic spectrometer. The acquisition process includes: adjusting the optical path to ensure that all devices are coaxial and at the same height during the acquisition process, while ensuring that the position of the spectral camera detection point is on the axis of symmetry of the molten metal. Then, the spectral data is processed to reduce thermal noise and the spectral image of the molten metal is captured and captured for computer acquisition.

[0041] S2. Subsequently, the collected spectral data is preprocessed to improve measurement accuracy;

[0042] S3. Background correction is applied to the obtained spectral data, and spectral reconstruction is achieved by using the spectral raw image reconstruction method.

[0043] S4. Spectral data is acquired again using the fiber optic probe of the spectrometer, and temperature inversion is performed to verify the accuracy of the temperature inverted by the spectrometer camera and determine the error.

[0044] In step S2, the derivative algorithm is used to process the overlap of the acquired spectral lines during the spectral data preprocessing.

[0045]

[0046]

[0047] Among them, S i S′ represents the spectral value at the i-th point of the spectrum in the original sample. i With S″ i These are the first and second derivatives with respect to the original spectral value, respectively; w is the width of the current spectral region at wavelength.

[0048] The derivative algorithm is followed by a multivariate scattering correction process, averaging the obtained spectra at each location. Then, a univariate linear regression operation is performed on the spectral data at each location and the average spectrum to obtain the regression coefficients and regression constants. Finally, the regression constants of each initial spectrum are subtracted and divided by their regression coefficients.

[0049] Furthermore, the method for reconstructing the spectral raw image in step S3 includes: extending the image gradient correlation parameters contained in the neighborhood sampling points to the unsampled points in the spectral image. Let the location of the unsampled spectral image point be b, and the corresponding sampling point in its neighborhood be a. Based on extending the known gradient information from point a to point b, the unsampled points are estimated.

[0050]

[0051] In the formula: f n (x) is the nth derivative of the current function expression; R n (x) represents the Lagrange remainder term. The two-dimensional spectral image can be reconstructed using the gradient-related information of the sampled spectral image points. The direction vector from a to b is defined as:

[0052]

[0053] In the formula: |ab| represents the distance between a and b in the current two-dimensional space, from which its directional derivative can be obtained as:

[0054]

[0055] It is possible to estimate b based on the directional derivative and gradient, given a:

[0056]

[0057] Reconstructing unsampled points b in the neighborhood of a known sampled point a is represented as follows:

[0058]

[0059] In the formula: a i ∈W s For window W in the spectral image s Sample points already collected; For unsampled point a i The binary mask at the location and For a i The kernel weight value at its location; For from a i The contribution value to b, where the kernel weight Represented as:

[0060]

[0061] Geometric distance is selected as the weight for calculation.

[0062] In step S4, the energy radiated outward from the surface of the molten metal to be tested is captured as it passes through filters with different center wavelengths to form image signals. This yields radiation information at specific wavelengths, which is then transmitted to a computer for calibration and calculation to determine the temperature distribution of the molten metal surface. Simultaneously, a fiber optic spectrometer is used to directly measure the spectrum of the molten metal, and temperature inversion is performed to obtain actual temperature data, verifying the accuracy of the temperature field reconstruction method using the spectral camera. Specific implementation examples:

[0064] Four heights (61.20 mm, 47.34 mm, 38.51 mm, and 16.30 mm) were selected to correspond to temperature points for multiple comparative experiments on radiation temperature measurement under multispectral conditions. The highest flame center temperature was observed at a height of 38.51 mm. In terms of relative temperature measurement error, compared to the temperature measurement results obtained by the spectrometer inversion method, the overall error of the temperature measured by this method is less than 0.6%. This scheme utilizes spectral image reconstruction to perform spectral reconstruction and uses an integrated spectral camera temperature measurement model to achieve temperature inversion. Compared to various current temperature field reconstruction algorithms, the multispectral camera radiation temperature measurement method proposed in this scheme can achieve high stability while ensuring the accuracy of temperature inversion, and can also better reflect the actual temperature distribution.

[0065] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for processing radiation spectrum reconstruction data of metal surfaces, characterized in that, Includes the following steps: S1. First, spectral data of the surface temperature of the molten metal is acquired using a spectral chip camera and a fiber optic spectrometer. The acquisition process includes: adjusting the optical path to ensure that all devices are coaxial and at the same height during the acquisition process, while ensuring that the position of the spectral camera detection point is on the axis of symmetry of the molten metal. Then, the spectral data is processed to reduce thermal noise and the spectral image of the molten metal is captured and captured for computer acquisition. S2. Subsequently, the collected spectral data is preprocessed to improve measurement accuracy; S3. Background correction is applied to the obtained spectral data, and spectral reconstruction is achieved by using the spectral raw image reconstruction method. S4. Use the fiber optic probe of the spectrometer to collect spectral data again and perform temperature inversion to verify the accuracy of the temperature inverted by the spectrophotometer and determine the error. In step S2, the derivative algorithm is used to process the overlap of the acquired spectral lines during the spectral data preprocessing. ； ； in, The corresponding spectrum in the collected original sample The magnitude of the spectral value at the point; and These are the first and second derivatives with respect to the original spectral values, respectively. This represents the width of the wavelength in the current spectral region. After processing by the derivative algorithm, a multivariate scattering correction process is performed to average the obtained spectra at each position. Then, a univariate linear regression operation is performed on the spectral data at each position and the average spectrum to obtain its regression coefficient and regression constant. Finally, the regression constants of each initial spectrum are subtracted and divided by their regression coefficients. The method for reconstructing the spectral raw image in step S3 includes: extending the image gradient correlation parameters contained in the neighborhood sampling points to the unsampled points in the spectral image. Let the location of the unsampled spectral image point be b, and the corresponding sampling point in its neighborhood be a. Based on extending the known gradient information from point a to point b, the unsampled points are estimated. ； In the formula: For the current function expression First derivative; For Lagrange type remainder; Two-dimensional spectral images can be reconstructed using gradient-related information from sampled spectral image points. The direction vector from a to b is defined as: ； In the formula: This represents the distance between a and b in the current two-dimensional space, from which we can obtain its directional derivative: ； It is possible to estimate b based on the directional derivative and gradient, given a: ； Reconstructing unsampled points b in the neighborhood of a known sampled point a is represented as follows: ； In the formula: A window in a spectral image Sample points already collected; Unsampled points The binary mask at the location and ; for The kernel weight value at its location; From arrive The contribution value, of which the core weight Represented as: ； Geometric distance is selected as the weight for calculation.

2. The method for processing data for reconstructing the radiation spectrum of a metal surface according to claim 1, characterized in that: In step S4, the energy radiated outward from the surface of the molten metal to be tested is captured as an image signal after passing through filters with different center wavelengths. The radiation information at a specific wavelength is obtained, and the obtained signal is transmitted to a computer for calibration and calculation to obtain the temperature distribution of the surface of the molten metal to be tested. At the same time, the spectrum of the molten metal is directly measured using a fiber optic spectrometer, and the actual temperature data is obtained by temperature inversion to verify the temperature measurement accuracy of the temperature field reconstruction method of the spectroscopic camera.

Images