Spectral screening method, equipment and storage medium for LIBS quantitative detection of deep-sea rare earth elements

Abstract

The present invention provides a spectral screening method, equipment and storage medium for LIBS quantitative detection of deep-sea rare earth elements. The LIBS spectra of rare earth samples under different conditions are collected, and the spectrum is mapped to the CIE L*a*b* color space by combining the spectral intensity and the CIE standard colorimetric observer spectrum tristimulus value function to achieve the characterization of the plasma color. The LIBS spectra under different conditions are quantitatively analyzed. The central area of ​​the spectral mapping point at this time is delineated based on the condition with the best quantitative effect. The standard area is used to constrain the spectral mapping points under other conditions to achieve spectral screening based on the luminescence characteristics of the plasma and improve the linear fit of the spectral calibration curve under other conditions. This method does not require complex data processing means. It only needs to be combined with mathematical formulas for calculation to achieve the mapping of the spectrum to the color space. It has a short running time, requires a small amount of data, is highly applicable, and the results have actual physical meaning and are interpretable.

Description

Technical Field

[0001] The present invention belongs to the technical field of rare earth spectral element analysis and identification, and in particular relates to a spectral screening method, equipment and storage medium for LIBS quantitative detection of deep-sea rare earth elements. Background Art

[0002] Rare earth elements (REEs), a collective term for 15 lanthanide elements and two transition metals (scandium and yttrium), are widely used in advanced technologies and manufacturing. Deep-sea rare earth resources hold enormous potential. "Deep-sea rare earth-rich deposits" are found in deep-sea basins. Their total rare earth concentration is generally greater than 700 μg / g, with known concentrations reaching as high as 8,000 μg / g, over 2,000 times the total onshore rare earth reserves. These reserves have become a major focus of international attention in recent years, with demand for rare earth resources projected to increase to 70,900 tons by 2035. As the second country in the world to conduct deep-sea rare earth surveys and research, my country urgently needs to develop new technologies and methods for deep-sea rare earth resource exploration and detection to facilitate and ensure future resource extraction.

[0003] Laser-induced breakdown spectroscopy (LIBS) is a newly emerging elemental analysis technique based on plasma radiation. It uses a high-energy laser focused by a lens to ablate the sample surface, forming a plasma. This spectral data allows for qualitative and quantitative analysis of the elements in the sample. LIBS requires no sample pretreatment, is real-time, and is pollution-free. It has been widely used in fields such as nuclear fusion, archaeology, environmental monitoring, and metallurgy. Therefore, LIBS technology is fully capable of conducting deep-sea rare earth exploration.

[0004] LIBS quantitative analysis is a method for inferring the concentration of a target element in a sample by detecting the spectral intensity of its characteristic spectral lines. Its core principle is that the intensity of the target element's characteristic spectral lines is proportional to its concentration. This method first measures the characteristic spectral intensities of a series of standard samples of known concentrations, plots the relationship between spectral intensity and concentration, known as a calibration curve, and performs a linear fit to derive a mathematical model that relates spectral intensity to element concentration. Subsequently, for unknown samples, the spectral intensity can be measured and substituted into the calibration curve to infer the element's concentration. However, LIBS quantitative analysis requires high accuracy in the calibration curve, a process that can be affected by multiple factors, such as matrix effects (interference in the measurement caused by sample composition and its physical state), fluctuations in plasma temperature, and electron density. These factors can lead to measurement errors, thereby affecting the accuracy of quantitative results. Therefore, LIBS quantitative analysis often requires the integration of signal optimization and data screening methods to minimize errors and improve analytical reliability and accuracy.

[0005] To address the problem of decreased quantitative analysis accuracy due to plasma state fluctuations, existing approaches employ machine learning or deep learning techniques, using complex algorithms to optimize the calibration curve fit and refine the relationship model between spectral intensity and elemental concentration. However, these approaches require large spectral datasets and long runtimes, consuming significant computational resources. The training process is complex, placing high demands on equipment performance and data processing capabilities. Furthermore, these algorithms fail to fully reflect the fundamental physical properties of LIBS spectra, such as plasma color, and lack universal applicability and interpretability. Summary of the Invention

[0006] To address these issues, the present invention discloses a CIE L*a*b* data screening method for deep-sea rare earth element quantification using LIBS. A formula is used to calculate the plasma emission spectrum—the coordinates of the points in the CIE L*a*b* color space corresponding to the LIBS spectrum—to characterize the plasma's color properties. Because detection conditions significantly influence quantitative analysis, the LIBS spectrum under optimal quantitative conditions is selected, and its region in the color space is designated as a standard region. Spectra under different conditions are then screened to improve the fit of the rare earth element calibration curve after fluctuations in conditions.

[0007] The first aspect of the present invention provides a spectral screening method for LIBS quantitative detection of deep-sea rare earth elements, comprising the following steps:

[0008] Step 1: Based on the established LIBS detection system, LIBS spectral data of multiple groups of rare earth samples with target element concentration gradients under different detection delays of the spectrometer are obtained, and each delay condition includes a total of K LIBS spectral data of multiple groups of rare earth samples;

[0009] Step 2: Based on the acquired LIBS spectral data, establish an element calibration curve to find a total of K LIBS spectral data of multiple groups of rare earth samples under the detection delay condition with the best quantitative effect;

[0010] Step 3: Based on the K pieces of LIBS spectral data under the detection delay condition with the best quantitative effect, calculate the CIE L*a*b* value corresponding to each piece of spectral data, thereby obtaining the coordinate set of the spectral mapping point;

[0011] Step 4: Based on the spectral mapping point coordinate set obtained in step 3, a standard area is defined according to the center of the point set;

[0012] In step 5, according to the standard area, the LIBS spectral data under other delay conditions are converted to the CIEL*a*b* color space according to step 3 to obtain the coordinates of the spectral data mapping points under other delay conditions. The spectral data falling outside the standard area are screened out.

[0013] Preferably, the constructed LIBS detection system includes a pulsed laser, a sample platform, a spectral signal receiving system, a spectral detection system, a timing control system and a computer control system;

[0014] A reflector is installed at the light outlet of the pulse laser to guide the pulse laser into the subsequent light path;

[0015] The sample platform is a three-dimensional movable platform on which the sample pressed into a sheet is placed. After the laser passes through the dichroic mirror, it is focused on the sample surface through the microscope objective lens.

[0016] The spectral signal receiving system includes a lens and a fiber optic probe. The spectral signal receiving system adopts backward collection. The light emitted by the plasma is reflected by the dichroic mirror. The lens converges the LIBS spectral signal into the fiber optic probe and connects to the spectral detection system through an optical fiber. The spectral detection system is controlled by a timing control system and a computer control system.

[0017] The spectrum detection system is realized by a spectrometer;

[0018] The timing control system is implemented by a PIN tube; a small portion of the laser light passes through the reflector and reaches the PIN tube, which then responds and triggers the spectrometer to collect data;

[0019] The computer control system is connected to the spectrum detection system and is used to set the spectrometer parameters and display the LIBS spectrum of the collected samples.

[0020] Preferably, the acquisition of LIBS spectral data of multiple groups of rare earth samples under different detection delays of the spectrometer is to continuously collect LIBS spectra of p laser pulses at each point of each group of samples under different detection delays of the spectrometer, collect m points in total, and obtain multiple LIBS spectra for each group of samples in total, thereby realizing the acquisition of LIBS spectra of samples under different delays.

[0021] Preferably, the calibration curve in step 2 is a regression curve of characteristic line intensity and element content established by using a series of standard samples. First, according to the acquired spectral data, the characteristic line of the rare earth element is selected, and the element content is used as the horizontal axis and the intensity of the characteristic line of the element at different contents is used as the vertical axis. A linear fit is performed to establish a calibration curve of a total of K LIBS spectral data of multiple groups of samples under each detection delay and calculate the determination coefficient R 2 The maximum value is 1 and the minimum value is 0. The closer the value is to 1, the better the model is. K pieces of LIBS spectral data under the detection delay condition closest to 1 are selected.

[0022] Preferably, the specific process of step 3 is:

[0023] Convert the spectral data to the CIE L*a*b* color space using the following formula:

[0024]

[0025] Where X, Y, and Z are the tristimulus values ​​of the CIE1931 standard colorimetric system. n 、Y n , Z n is the tristimulus value of the CIE standard illuminant irradiated on a completely diffuse reflector and then reflected into the observer's eyes, I(λ) is the relative spectral power distribution of the light source radiation, k is the normalization coefficient, is the CIE1931 standard colorimetric observer spectrum tristimulus value, Δλ is the wavelength interval; "L*" represents lightness, indicating color from black to white; "a*" represents green-red color saturation, "+a*" represents red, and "-a*" represents green; "b*" represents blue-yellow color saturation, "+b*" represents yellow, and "-b*" represents blue;

[0026] According to the calculation formula, the 380-780nm visible light band of the LIBS spectrum was selected, and MATLAB was used for interpolation to obtain the spectral line intensity values ​​at a certain wavelength interval. Combined with the spectral tristimulus values, the XYZ tristimulus values ​​were solved to obtain the L*a*b* values ​​in the CIE L*a*b* color space. The mapping of LIBS spectral data to the CIE L*a*b* color space was realized, and the high-dimensional spectral data was converted into three-dimensional color features, and the coordinate set of the spectral data mapping points was obtained.

[0027] Preferably, the specific process of step 4 is:

[0028] Calculate the center point position of the spectral mapping points of samples with different element concentrations, that is, solve the average value of the coordinate points according to the three dimensions respectively, and calculate the Euclidean distance between each point and the center point; sort the distances from small to large, and retain the spectral mapping points corresponding to the first 1 / e distances. The area covered by this part of the point set is the standard area.

[0029] A second aspect of the present invention provides a spectral screening device for LIBS quantitative detection of deep-sea rare earth elements, the device comprising at least one processor and at least one memory, the processor and memory being coupled; a computer executable program being stored in the memory; when the processor executes the computer executable program stored in the memory, the processor executes the spectral screening method for LIBS quantitative detection of deep-sea rare earth elements as described in the first aspect.

[0030] A third aspect of the present invention provides a computer-readable storage medium, which stores a computer program or instruction. When the program or instruction is executed by a processor, the processor executes the spectral screening method for LIBS quantitative detection of deep-sea rare earth elements as described in the first aspect.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] This paper proposes a method for improving the accuracy of deep-sea rare earth element quantitative analysis using LIBS. By introducing the CIE L*a*b* color space to characterize plasma color, and using standard regions for screening, the linear fit of the calibration curve is improved. This method requires minimal data and eliminates complex computations. It can screen spectra based on the physical properties of plasma luminescence, thereby enhancing the quantitative analysis capabilities of LIBS.

[0033] This method collects LIBS spectra of rare earth samples under different conditions, combines the spectral intensity with the CIE standard colorimetric observer spectral tristimulus value function, and maps the spectra to the CIE L*a*b* color space to characterize the plasma color. LIBS spectra under different conditions are quantitatively analyzed. The LIBS spectrum obtained under the optimal quantitative condition is used as a standard to delineate the central region of the spectral mapping point at that time. This standard region is used to constrain other spectral mapping points, enabling spectral screening based on the plasma's luminescence characteristics and improving the linear fit of the spectral calibration curve under other conditions.

[0034] The method of the present invention does not require complex data processing means, and can realize the mapping of spectrum to color space by only combining mathematical formulas for calculation. It has short running time, small amount of data required, strong applicability, and the results have actual physical meaning and are interpretable. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A flow chart for implementing the method of the present invention;

[0036] Figure 2 This is a schematic diagram of the structure of the deep-sea rare earth LIBS spectrum detection device required by the present invention;

[0037] Figure 3 Schematic diagram of the LIBS spectrum conversion process according to an embodiment of the present invention;

[0038] Figure 4 Schematic diagram of the distribution of the LIBS spectrum in the CIE L*a*b* color space according to an embodiment of the present invention;

[0039] Figure 5 Schematic diagram of the screening process according to an embodiment of the present invention;

[0040] Figure 6 4 is a comparison chart of the quantitative results before and after screening of an embodiment of the present invention.

[0041] Figure 7 A simplified diagram of the spectral screening device of the present invention.

[0042] Figure 2 Middle: 1- pulse laser, 2- reflecting mirror, 3- dichroic mirror, 4- microscope objective, 5- sample platform, 6- lens, 7- optical fiber, 8- spectrometer, 9- PIN tube, 10- computer control system. DETAILED DESCRIPTION

[0043] Principle of the invention: The present invention aims to provide a CIE L*a*b* data screening method for the quantitative analysis of deep-sea rare earth element laser-induced breakdown spectroscopy. Since the detection conditions have a significant impact on the quantitative analysis, a series of LIBS spectra under different conditions are collected. The spectrum under the optimal quantitative conditions is selected, and the corresponding color space coordinates are solved according to the calculation formula to achieve the mapping of the LIBS spectrum to the CIE L*a*b* color space. The area of ​​the color space is delineated as the standard area, and the plasma state reflected in this area is considered to be suitable for quantification. By judging whether the spectral mapping points under other conditions are in this area, the spectrum is screened, thereby improving the linear fit of the spectral calibration curve after the condition fluctuations.

[0044] To achieve the above object, the present invention adopts the following technical solutions:

[0045] 1. Based on the established LIBS experimental system, LIBS spectral data of multiple groups of samples with target element concentration gradients are obtained under different detection delay conditions of the spectrometer.

[0046] The deep-sea rare earth laser-induced breakdown spectroscopy detection system disclosed in the present invention includes a pulse laser 1, a reflector 2, a dichroic mirror 3, a microscope objective 4, a sample platform 5, a lens 6, an optical fiber 7, a spectrometer 8, a PIN tube 9 and a computer control system 10. Figure 1 shown.

[0047] The pulse laser is a Q-switched Nd:YAG laser with an excitation wavelength of 1064 nm and a pulse repetition frequency of 1 Hz. A reflector is installed at the laser light outlet to guide the pulse laser into the subsequent optical path.

[0048] The sample platform is a three-dimensional moving platform on which the pressed sample is placed. After passing through the dichroic mirror, the laser is focused on the sample surface by the microscope objective lens.

[0049] The spectral signal receiving system includes a lens and a fiber optic probe. It uses backward collection. The light emitted by the plasma is reflected by a dichroic mirror, and the lens converges the LIBS spectral signal into the fiber optic probe. This is then connected to a spectrometer via an optical fiber. The spectrometer is controlled by a timing control system and a computer control system.

[0050] The spectrum detection system is realized by a spectrometer.

[0051] The timing control system is implemented through a PIN diode. A small portion of the laser light passes through a reflector and reaches the PIN diode, which then responds and triggers the spectrometer to collect data.

[0052] The computer control system is connected to the spectrum detection system and is used to set parameters such as the spectrometer delay and integration time, and to display the LIBS spectrum of the collected samples.

[0053] Multiple spectrometer detection delays are set. Under each spectrometer detection delay, K LIBS spectral data are collected for multiple groups of samples to achieve the acquisition of sample LIBS spectral data under different detection delays.

[0054] 2. Based on the acquired LIBS spectral data, establish element calibration curves and find a total of K LIBS spectral data for multiple groups of samples under the detection delay conditions with the optimal quantitative effect.

[0055] LIBS quantitative analysis requires the use of a series of standard samples to establish a regression curve between the intensity of characteristic spectral lines and the content of elements, namely the calibration curve. This curve reflects the quantitative relationship between the intensity of the element emission line and its content under specific experimental conditions. In order to establish the calibration curve, first, based on the acquired spectrum, the characteristic spectral line of the target element is selected, and the element content is used as the horizontal coordinate and the intensity of the characteristic spectral line of the element at different contents is used as the vertical coordinate. A linear fit is performed to establish the calibration curve. For a total of K LIBS spectral data of multiple groups of samples under each detection delay, a calibration curve is established and R is calculated. 2The value (coefficient of determination, used to quantify the degree of fit of the model to the data) is calculated as follows:

[0056]

[0057] Among them, y i represents the actual observation value, represents the mean of the true observations, Represents the predicted value.

[0058] R 2 The maximum value is 1 and the minimum is 0. The closer the value is to 1, the better the model is, and the closer it is to 0, the worse the model is. 2 A total of K LIBS spectral data of multiple groups of samples under the detection delay value closest to 1 are considered to have the best quantitative effect.

[0059] 3. Based on the LIBS spectral data with the best quantitative effect, calculate the corresponding CIE L*a*b* values ​​to obtain the spectral mapping point coordinate set with the best quantitative effect.

[0060] LIBS spectra based on plasma radiation have certain color characteristics, so the spectral color is characterized using the CIE L*a*b* color space. The calculation formula for converting the spectrum to the CIE L*a*b* color space is as follows:

[0061]

[0062] Where X, Y, and Z are the three stimulus values ​​of the CIE1931 standard colorimetric system, Xn, Yn, and Zn are the three stimulus values ​​of the CIE standard illuminant irradiated on a perfect diffuse reflector and then reflected into the observer's eyes, I(λ) is the relative spectral power distribution of the light source radiation, and k is the normalization coefficient. The three stimulus values ​​for the CIE 1931 standard observer color spectrum are shown, with Δλ being the wavelength interval. "L*" represents lightness, indicating color from black to dark; "a*" represents green-red color saturation, with "+a*" representing red and "-a*" representing green; and "b*" represents blue-yellow color saturation, with "+b*" representing yellow and "-b*" representing blue.

[0063] like Figure 3As shown, the wavelength range of the CIE standard colorimetric observer spectral tristimulus function is 380nm-780nm. Therefore, according to the calculation formula, we need to select the 380-780nm visible light band of the LIBS spectrum and use MATLAB to interpolate to obtain spectral line intensity values ​​at certain wavelength intervals. Combined with the spectral tristimulus values, we solve for the XYZ tristimulus values ​​and further obtain the L*a*b* values ​​in the CIE L*a*b* color space. This realizes the mapping of the LIBS spectrum to the CIE L*a*b* color space, converting high-dimensional spectral data into three-dimensional color features.

[0064] Select the detection delay with the best linear fit of the calibration curve, and map the spectrum under this condition to the CIE L*a*b* color space according to the above conversion method to obtain the coordinate set of the spectrum mapping point, as shown in Figure 4 shown.

[0065] 4. Based on the quantitatively optimal spectral mapping point coordinate set, a standard area is defined according to the center of the point set.

[0066] Since the points on the calibration curve are calculated using the average values ​​of the characteristic spectral line intensities at each concentration, the center point of the spectral mapping points of samples with different element concentrations is calculated. This means that the average values ​​of the coordinate points along the three dimensions are calculated, and the Euclidean distance between each point and the center point is calculated. The distances are sorted in ascending order, and the spectral mapping points corresponding to the first 1 / e distances are retained. The area covered by this set of points is the standard area.

[0067] 5. According to the standard area, the spectral mapping points under other time delays can be screened.

[0068] The LIBS spectra under other delay conditions (non-quantitative optimal delay conditions) were converted to the CIE L*a*b* color space according to step 3. The converted mapping points were screened according to the defined standard area, such as Figure 5 If a spectral mapping point falls within the standard region, the spectrum corresponding to that point is retained; conversely, if it falls outside the standard region, it is discarded. This screening mechanism allows for more accurate retention of spectral data that best reflects the rare earth element content.

[0069] 6. Establish element calibration curves for the screened spectra to verify the quantitative improvement effects before and after.

[0070] The rare earth element calibration curve was established for the screened spectral data, and the linear fit of the calibration curve before and after screening was compared. Figure 6 As shown, after screening, the R 2The CIE L*a*b* color space screening method can effectively optimize the LIBS quantitative analysis of rare earth elements, thereby improving the accuracy and reliability of the analysis.

[0071] The method of the present invention is further described below with reference to specific embodiments.

[0072] Figure 1 This is a flow chart for implementing the method of the present invention. Build a LIBS experimental system, obtain the LIBS spectra of samples under different conditions and perform simple quantitative analysis to find the spectrum with the best quantitative performance. According to the calculation formula, select the 380-780nm visible light band of the spectrum, use MATLAB to perform interpolation, obtain the spectral line intensity values ​​with an interval of 1nmm, combine the spectral tristimulus values, solve the XYZ tristimulus values, and further obtain the L*a*b* values ​​in the CIE L*a*b* color space, and map the quantitatively optimal spectrum to the CIE L*a*b* color space. Calculate the center point position of the spectral mapping points of different types of samples at this time, retain a part of the spectral mapping points closest to the center point of each type of sample, and use the area formed by this part of points as the standard to constrain the spectral characteristic points of different types of samples under other conditions. If the spectral mapping point falls within the standard area, then this point will be retained; conversely, if it falls outside the standard area, it will be eliminated. Establishing a calibration curve for the screened spectrum can improve its R 2 value (coefficient of determination, used to quantify the degree of fit of the model to the data) to improve the accuracy of quantitative analysis.

[0073] like Figure 2 As shown, the deep-sea rare earth laser-induced breakdown spectroscopy system disclosed in the present invention includes a 1064 nm pulsed laser 1, a sample platform 5, spectral signal receiving systems 6-7, a spectrometer 8, a timing control system 9, and a computer control system 10. The pulsed laser 1 utilizes a Q-switched Nd:YAG laser with an excitation wavelength of 1064 nm and a pulse repetition frequency of 1 Hz. A reflector 2 is installed at the laser's light output port to direct the laser light into the subsequent optical path. A portion of the laser light passes through the reflector 2, where it is converted into an electrical signal by a PIN diode 9, which then synchronously triggers the spectrometer. The laser light then passes through a dichroic mirror 3 and is focused onto the sample surface by a microscope objective 4, exciting the sample to generate plasma. The light emitted by the plasma is reflected by the dichroic mirror 3 and focused by a lens 6 into a fiber optic probe 7. The optical fiber then directs the detected signal into the spectrometer 8, where it is observed and the spectrum is collected by a computer control system 10.

[0074] The specific sample spectrum acquisition process is as follows: First, a tablet press is used to compress the rare earth sample at a pressure of 10 MPa to obtain tablets of five different rare earth samples. The tablet is placed below the microscope objective and fixed on a three-dimensional translation stage. The Z axis is moved to bring the rare earth sample tablet into focus. The spectrometer delay is set to 500 ns. A Q-switched Nd:YAG laser is turned on with a pulse repetition rate of 1 Hz and a laser energy of 50 mJ. This generates a plasma on the surface of the rare earth sample tablet. The light emitted by the plasma is reflected by a dichroic mirror, and the resulting spectral signal is collected by a lens. LIBS spectra are collected continuously for 10 laser pulses at each sample point. Ten sample points are collected by moving the XY axis of the translation stage, resulting in a total of 100 LIBS spectra for each sample. The spectrometer delay is changed to 1000 ns and 1500 ns, and the above acquisition steps are repeated to obtain LIBS spectra for the five rare earth samples at different delays.

[0075] like Figure 3 As shown, taking the LIBS spectrum of the GBW07590 rare earth sample obtained under a delay of 1500ns as an example, the upper figure is the visible light band of the selected LIBS spectrum. Although the wavelength range of the spectrometer used in the present invention is 235-800nm, only the 380-780nm band can characterize the color, so this band is selected. The figure below shows the response of the LIBS spectrum to the CIE standard colorimetric observer spectral tristimulus function. Since the resolution of the system spectrometer used is 0.7nm, the spectral tristimulus value of Δλλ=1nm is selected. MATLAB is used to obtain the spectral line intensity with an interval of 1nm by interpolation, and the summation calculation is performed to match the spectral tristimulus value, and then the L*a*b* value can be obtained according to the formula.

[0076] By establishing calibration curves for the spectra at 500ns, 1000ns, and 1500ns delays, it was found that the R2 of the calibration curve at 1500ns delay was close to 1, indicating that the quantitative effect was optimal at this time. Figure 3 The process obtains the L*a*b* value of each sample spectrum under the conditions, and maps the LIBS spectrum to the CIE L*a*b* color space to achieve the characterization of the luminescent colors of different sample plasmas, such as Figure 4 shown.

[0077] Calculate the center point position of the spectrum mapping points of different types of samples under the delay, retain the part of the spectrum mapping points closest to the center point of each type of sample, and use the area formed by this part of points as the standard, such as Figure 5The red area in the middle shows the 500ns and 1000ns spectra mapped to the CIE L*a*b* color space. The spectral mapping points were then filtered based on the standard region with the best quantitative results. Spectra corresponding to points within the region were retained, while those not within the region were deleted, completing the screening of LIBS spectra under fluctuating conditions.

[0078] Establish element calibration curve for the screened spectrum and calculate the determination coefficient R 2 value, Figure 6 This is a comparison of the quantitative results before and after screening in the embodiment of the present invention. For the LIBS quantitative analysis of rare earth elements such as Dy, the determination coefficient R of the calibration curve before and after screening is 2 The improvement was significant, with the 389.9nm characteristic line having the most significant effect. 2 The value increased from 0.339 to 0.905, proving that the screening method based on the CIE L*a*b* color space can improve the accuracy of LIBS quantitative analysis and is a fast and effective spectral screening method.

[0079] like Figure 7 As shown, the present invention also provides a spectral screening device for deep-sea rare earth element LIBS quantitative detection. The device includes at least one processor and at least one memory, as well as a communication interface and an internal bus; the memory stores a computer executable program; the memory stores a computer executable program; when the processor executes the computer executable program stored in the memory, it can cause the processor to perform the spectral screening method for deep-sea rare earth element LIBS quantitative detection as described above. The internal bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be divided into address buses, data buses, control buses, etc. For ease of illustration, the buses in the drawings of this application are not limited to only one bus or one type of bus. The memory may include high-speed RAM memory, and may also include non-volatile storage (NVM), such as at least one disk storage, and can also be a USB flash drive, a mobile hard drive, a read-only memory, a magnetic disk, or an optical disk.

[0080] The device may be provided as a terminal, a server, or other form of device. In an exemplary embodiment, the electronic device may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0081] The present invention also provides a computer-readable storage medium, which stores a computer execution program. When the computer execution program is executed by a processor, the processor can execute the spectral screening method for LIBS quantitative detection of deep-sea rare earth elements as described above.

[0082] Specifically, a system, device, or apparatus equipped with a machine-readable storage medium may be provided, wherein the machine-readable storage medium stores software program code that implements the functions of any of the above-described embodiments, and the system, device, or apparatus is configured to read and execute the instructions stored in the machine-readable storage medium. In this case, the program code read from the machine-readable storage medium itself can implement the functions of any of the above-described embodiments, and thus the machine-readable code and the machine-readable storage medium storing the machine-readable code constitute part of the present invention.

[0083] The above description is merely a preferred embodiment of the present application and is not intended to limit the present application. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present application shall be included within the scope of protection of the present application.

[0084] Although the above describes the specific implementation methods of the present invention, it does not limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A spectral screening method for LIBS quantitative detection of deep-sea rare earth elements, characterized in that: The following processes are included: Step 1: Based on the established LIBS detection system, LIBS spectral data of multiple groups of rare earth samples with target element concentration gradients under different detection delays of the spectrometer are obtained, and each delay condition includes a total of K LIBS spectral data of multiple groups of rare earth samples; The constructed LIBS detection system includes a pulsed laser, a sample platform, a spectral signal receiving system, a spectral detection system, a timing control system and a computer control system; A reflector is installed at the pulse laser light outlet to guide the pulse laser into the subsequent light path; The sample platform is a three-dimensional movable platform on which the pressed sample is placed. After the laser passes through the dichroic mirror, it is focused on the sample surface through the microscope objective lens. The spectral signal receiving system includes a lens and a fiber optic probe. The spectral signal receiving system adopts backward collection. The light emitted by the plasma is reflected by the dichroic mirror. The lens converges the LIBS spectral signal into the fiber optic probe and connects to the spectral detection system through an optical fiber. The spectral detection system is controlled by a timing control system and a computer control system. The spectrum detection system is realized by a spectrometer; The timing control system is implemented by a PIN tube; a small portion of the laser light passes through the reflector and reaches the PIN tube, which triggers the spectrometer to collect data after responding. The computer control system is connected to the spectrum detection system to set the spectrometer parameters and display the LIBS spectrum of the collected samples; Step 2: Based on the acquired LIBS spectral data, establish an element calibration curve to find a total of K LIBS spectral data of multiple groups of rare earth samples under the detection delay condition with the best quantitative effect; Step 3: Based on the K pieces of LIBS spectral data under the optimal detection delay condition for quantitative effect, calculate the CIE L*a*b* value corresponding to each piece of spectral data to obtain the coordinate set of the spectral mapping point; the specific process is as follows: Convert the spectral data to the CIE L*a*b* color space using the following formula: Where X, Y, and Z are the tristimulus values ​​of the CIE1931 standard colorimetric system. n 、Y n , Z n is the tristimulus value of the CIE standard illuminant irradiated on a completely diffuse reflector and then reflected into the observer's eyes, I(λ) is the relative spectral power distribution of the light source radiation, k is the normalization coefficient, is the CIE1931 standard colorimetric observer spectrum tristimulus value, Δλ is the wavelength interval; "L*" represents lightness, indicating color from black to white; "a*" represents green-red color saturation, "+a*" represents red, "-a*" represents green; "b*" represents blue-yellow color saturation, "+b*" represents yellow, "-b*" represents blue; According to the calculation formula, the 380-780nm visible light band of the LIBS spectrum was selected and interpolated using MATLAB to obtain the spectral line intensity values ​​at certain wavelength intervals. Combined with the spectral tristimulus values, the XYZ tristimulus values ​​were solved to obtain the L*a*b* values ​​in the CIE L*a*b* color space. This enabled the mapping of LIBS spectral data to the CIE L*a*b* color space, converting the high-dimensional spectral data into three-dimensional color features and obtaining the coordinate set of the spectral data mapping points. Step 4: Based on the spectral mapping point coordinate set obtained in step 3, a standard area is defined according to the center of the point set; In step 5, according to the standard area, the LIBS spectral data under other delay conditions are converted to the CIE L*a*b* color space according to step 3 to obtain the coordinates of the spectral data mapping points under other delay conditions. The spectral data falling outside the standard area are filtered out.

2. The spectral screening method for LIBS quantitative detection of deep-sea rare earth elements according to claim 1, characterized in that: The method of obtaining LIBS spectral data of multiple groups of rare earth samples under different detection delays of the spectrometer is to continuously collect LIBS spectra of p laser pulses at each point of each group of samples under different detection delays of the spectrometer, collect m points in total, and obtain multiple LIBS spectra for each group of samples in total, thereby realizing the acquisition of LIBS spectra of samples under different delays.

3. The spectral screening method for LIBS quantitative detection of deep-sea rare earth elements according to claim 1, characterized in that: The calibration curve in step 2 is a regression curve between the intensity of the characteristic spectral line and the element content established using a series of standard samples. First, based on the acquired spectral data, the characteristic spectral line of the rare earth element is selected, and a linear fit is performed with the element content as the horizontal axis and the intensity of the characteristic spectral line of the element at different contents as the vertical axis. A calibration curve of a total of K LIBS spectral data of multiple groups of samples under each detection delay is established and the determination coefficient R is calculated. 2 The maximum value is 1 and the minimum value is 0. The closer the value is to 1, the better the model is. K pieces of LIBS spectral data under the detection delay condition closest to 1 are selected.

4. The spectral screening method for LIBS quantitative detection of deep-sea rare earth elements according to claim 1, characterized in that: The specific process of step 4 is as follows: Calculate the center point position of the spectral mapping points of samples with different element concentrations, that is, solve the average value of the coordinate points according to the three dimensions respectively, and calculate the Euclidean distance between each point and the center point; sort the distances from small to large, and retain the spectral mapping points corresponding to the first 1 / e distances. The area covered by the partial point set is the standard area.

5. A spectral screening device for LIBS quantitative detection of deep-sea rare earth elements, characterized by: The device includes at least one processor and at least one memory, the processor and the memory are coupled; the memory stores a computer execution program; when the processor executes the computer execution program stored in the memory, the processor executes the spectral screening method for LIBS quantitative detection of deep-sea rare earth elements as described in any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program or instruction, which, when executed by a processor, causes the processor to execute the spectral screening method for LIBS quantitative detection of deep-sea rare earth elements as described in any one of claims 1 to 4.

Images