CN107817239A - A kind of LIBS spectral correction methods based on plasma position information - Google Patents

Abstract

The present invention relates to a kind of LIBS spectrum-stable methods based on plasma position information, concretely comprise the following steps：1) LIBS spectral signals and the corresponding Plasma picture signal under different sample positions are obtained；2) plasma position information is extracted from Plasma picture information；3) relation function of LIBS spectrums region and value with corresponding plasma position is established by approximating method；Relation function of the specific the intensity of spectral line of LIBS spectrum with corresponding plasma position is established by approximating method；4) relation function obtained in using 3) establishes comprehensive correction function；5) the LIBS spectrum of Location-Unknown are corrected using correction function.Characteristic spectral line after being corrected with this method, there is higher stability, more reliable spectral line information is provided for quantitative (qualitative) analysis.

Description

A LIBS Spectral Correction Method Based on Plasma Position Information

technical field

The invention belongs to the field of spectrum analysis and material composition analysis, and specifically relates to a LIBS spectrum correction method based on plasma position information.

Background technique

Laser-induced breakdown spectroscopy (LIBS) analysis technology is an atomic emission spectroscopy analysis technology that uses laser-excited plasma as a light source. The qualitative and quantitative analysis of material composition based on LIBS technology has become a hot application in many industrial fields such as metallurgy, energy, food, chemical industry, etc. technology. However, due to the low stability of the obtained spectral signals, LIBS is only used for qualitative analysis and semi-quantitative analysis. Therefore, only by improving the stability of LIBS signal can it meet the needs of quantitative analysis, and thus be applied to real-time on-line in situ detection.

At present, the commonly used spectral preprocessing methods to improve stability are mainly the selection of reference lines and full-spectrum normalization methods. The former requires that the sample to be tested contains an element with a stable content, and the element has a clear characteristic spectral line in the detection wavelength range; the latter assumes that the sum of all spectral intensities in the detection wavelength range is Constant, this assumption is unreasonable when the detection spectrum range is narrow and the matrix element content fluctuates greatly. Correcting the spectral signal through the extracted plasma characteristic parameters is a spectral correction method based on the physical characteristics of the plasma, but it needs to detect a wide range of spectral bands, and needs to contain multiple clear atoms in this range , ion spectra, and modeled by complex fitting models.

As the most direct reflection of plasma luminescence, plasma images have always been an important way to analyze the temporal and spatial distribution of plasma and study the formation of plasma. Extracting the plasma position information from the plasma image, and then correcting the spectral signal jitter caused by the inconsistency of the surface height of the sample to be measured, is a feasible means to stabilize the spectral signal.

Contents of the invention

Aiming at the above-mentioned deficiencies in the prior art, the technical problem to be solved by the present invention is to establish a mathematical model and correct the LIBS spectral signal through the plasma position information contained in the plasma image, so as to obtain a stable Reliable characteristic line intensity.

The technical solution adopted by the present invention to achieve the above object is: a LIBS spectrum correction method based on plasma position information, comprising the following steps:

Step 1: Obtain LIBS spectral signals and corresponding plasma images at different heights of the sample;

Step 2: Determine the plasma light-emitting area according to the value of each pixel in the plasma image, and determine the plasma position according to the shape and pixel value of the light-emitting area;

Step 3: Through the fitting method, establish the corresponding functional relationship between the full spectrum segment within the LIBS spectral collection range and the plasma position information; at the same time, through the fitting method, establish the correspondence between the specific characteristic spectral line intensity of the LIBS spectrum and the plasma position information Functional relationship;

Step 4: According to the function obtained in step 3, establish a correction function for correcting the LIBS spectral signal by plasma position information;

Step 5: Apply the correction function obtained in step 4 to correct the LIBS spectrum with unknown sample height.

Step 1: Obtain LIBS spectra and corresponding plasma images at different sample heights by moving the sample, and the moving distance is judged by the plasma image.

The step 2 is as follows: firstly, set the pixel threshold to determine the plasma region, and filter out the background noise; then, determine the plasma position parameters by calculating the aspect ratio, the descending gradient of pixel values in each direction, and the center of gravity of the image in the region.

The fitting method is used to establish the corresponding functional relationship between the full spectrum segment within the LIBS spectral collection range and the plasma position information, specifically:

Calculate the spectral line intensity sum within the range of the LIBS spectrum detection band obtained under different positions, and fit the corresponding relationship between the spectral band intensity sum value and the plasma position parameter by I _SUM =f _SUM (h), where f _SUM (h ) is the fitting function.

The expression of the fitting function is as follows:

where sgn(·) represents the sign function, h is the parameter representing the plasma position information, h ₀ , σ ₁ , σ ₂ are the offset parameters and position parameters determined by fitting.

Said through the fitting method, the corresponding functional relationship between the intensity of the specific characteristic line of the LIBS spectrum and the plasma position information is established, specifically:

The corresponding relationship between the intensity of each characteristic spectral line and the plasma position parameter is determined by I _λ =f _λ (h), where f _λ (h) is a fitting function.

The expression of the fitting function is as follows:

where sgn(·) represents the sign function, h is the parameter representing the plasma position information, h ₀ , σ ₁ , σ ₂ are the offset parameters and position parameters determined by fitting.

The expression of the correction function is as follows:

Among them, I′ _λ represents the corrected spectral line intensity, I _λ is the original spectral line intensity, I _SUM is the original spectral segment intensity sum, f _λ (h) is the correspondence between the specific characteristic spectral line intensity of the LIBS spectrum and the plasma position information The fitting function of the functional relationship, f _SUM (h) is the fitting function of the full spectrum segment within the LIBS spectral collection range and the corresponding functional relationship with the plasma position information, where h is the plasma position parameter extracted from the plasma image , h ₀ is the h when f _λ (h) reaches the maximum value.

The present invention has the following advantages and beneficial effects:

1. The method proposed by the present invention does not need to detect a wide spectrum to obtain spectral data to calculate the plasma characteristic parameters, and does not need to change the sample position to suppress spectral jitter, extract the plasma position information from the synchronously collected plasma image, and The spectral signal is directly corrected using a mathematical model.

2. The method proposed by the present invention takes into account the fluctuation of the characteristic spectral line itself and the selected spectral region as a whole with the sample position, so that the built mathematical model takes into account the full spectrum while correcting the spectral signal through the plasma position information. Advantages of the normalization method.

3. For the model required by the method proposed by the present invention, the location information as an independent variable does not need to be strictly limited by experiments, and can be extracted from plasma images, and the complexity of experimental operations for modeling is low.

Description of drawings

Fig. 1 is the realization flow chart of the method of the present invention;

Figure 2 is the LIBS spectrum of a specific region under different sample positions;

Fig. 3 is the plasma image under different sample positions;

Fig. 4 is the fitting curve of the intensity values of the characteristic lines of the five elements Si, Fe, Cu, Mn and Mg in the aluminum alloy sample as a function of the plasma position;

Fig. 5 is the experimental result of direct calibration between the corrected characteristic spectral lines and the normalized characteristic spectral lines of the whole spectrum using the method of the present invention.

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings and examples.

The present invention aims at the problem that the laser focusing depth and the detection position of the spectrometer are different due to the different positions of the samples to be measured, thereby causing spectral jitter, obtains the plasma position information through the plasma image, and establishes a mathematical model to standardize the spectral data obtained under different sample positions processing to achieve correction of spectral data.

As shown in Figure 1, after the method starts, the spectral data and plasma image are obtained by the LIBS experimental platform, the position information is extracted from the plasma image information, and the fitting model is established by combining the intensity of the corresponding characteristic spectral lines and spectral segments, and obtained The fitting function of is to correct the spectrum to be analyzed. The specific implementation steps are as follows:

Step 1: Obtain the LIBS spectral signals at different heights of the sample (a typical correlation region LIBS spectrum is shown in Figure 2) and the corresponding plasma image signal (a typical plasma image is shown in Figure 3), where different sample heights are adjusted by adjusting the sample platform implementation.

Step 2: Determine the plasma light-emitting area according to the value of each pixel in the plasma image, and determine the plasma position according to the shape of the light-emitting area and the pixel value. First, the pixel threshold is set to determine the plasma region, and the background noise is filtered out; then, the plasma position parameters are determined by calculating the aspect ratio, the gradient of the pixel value in each direction, and the center of gravity of the image in the region.

Step 3: Establish the corresponding functional relationship between the full spectrum segment within the LIBS spectral collection range and the plasma position information by the fitting method. The sum of spectral line intensities within the range of the detection spectrum of LIBS spectra obtained at different positions is calculated, and the corresponding relationship between the intensity sum of the spectral bands and the plasma position parameters is fitted by I _SUM =f(h). in,

sgn(·) represents a sign function, h is a parameter representing plasma position information, h ₀ , σ ₁ , σ ₂ are offset parameters and position parameters determined by fitting.

Step 4: For different characteristic spectral lines, according to the fitting function of step 3, determine the corresponding relationship between the intensity of each characteristic spectral line and the plasma position parameter by _Iλ =f(h);

Step 5: Combining the fitting functions determined in steps 3 and 4, correct the comprehensive model of the LIBS spectral signal through the plasma position information. The specific correction function form is as follows:

Among them, I′ _λ represents the spectral line intensity after correction, I _λ is the original spectral line intensity, I _SUM is the original spectral segment intensity sum, f _λ (h) is the fitting function determined in step 4, and f _SUM (h) is The fitting function determined in step 3, where h is the plasma position parameter extracted from the plasma image, and h ₀ is h when f _λ (h) reaches the maximum value.

Step 6: Apply the model obtained in step 5 to analyze and correct the LIBS spectrum of unknown sample height.

The corrected spectral lines obtained by the method can be directly used for calibration analysis, and can also be used as the input of a complex calibration model, and the quantitative (qualitative) analysis ability of LIBS can be further improved by the multivariate analysis method.

Establish a correction model according to the above method to calibrate the spectra collected under 1mm sample position shaking of the five characteristic spectral lines corresponding to the five elements of Si, Fe, Cu, Mn and Mg in five aluminum alloy samples, and use the corrected The spectral line intensities were directly calibrated. The obtained fitting curve is shown in Figure 4, and the specific fitting parameters are as follows:

The results of calibration of the element concentrations using the corrected spectral line intensities and the results of calibration using the normalized intensity of the full spectrum are shown in Figure 5. Wherein the blue line corresponds to the full spectrum normalization method, and the red line represents the method proposed by the present invention. The coefficient of determination (R ² ), relative standard error (RSD), root mean square error (RMSE) and other performance indicators all show that this method can achieve better spectral line correction effect than the full-spectrum normalization method. More specific index comparison See the table below:

Claims (8)

1. A LIBS spectrum correction method based on plasma position information is characterized by comprising the following steps:

step 1: acquiring LIBS spectral signals and corresponding plasma images of a sample at different heights;

step 2: determining a light-emitting area of the plasma according to the value of each pixel point in the plasma image, and determining the position of the plasma according to the appearance and the pixel value of the light-emitting area;

and step 3: establishing a corresponding function relation between the full spectrum in the LIBS spectrum acquisition range and the plasma position information by a fitting method; meanwhile, establishing a corresponding function relation between the LIBS spectrum specific characteristic spectral line intensity and the plasma position information by a fitting method;

and 4, step 4: establishing a correction function for correcting the LIBS spectral signal according to the function obtained in the step 3;

and 5: and (4) correcting the LIBS spectrum with unknown sample height by applying the correction function obtained in the step (4).

2. The method as claimed in claim 1, wherein the step 1 is to obtain LIBS spectra and corresponding plasma images at different sample heights by moving the sample, and the moving distance is determined by the plasma images.

3. The method as claimed in claim 1, wherein the step 2 is specifically as follows: firstly, setting a pixel threshold value to determine a plasma region, and filtering background noise; and then determining plasma position parameters by calculating the length-width ratio, the descending gradient of the pixel values in all directions and the gravity center of the image in the region.

4. The method as claimed in claim 1, wherein the method for LIBS spectrum calibration based on plasma location information comprises the following steps of establishing a full spectrum range in the LIBS spectrum collection range and a corresponding functional relationship with the plasma location information by a fitting method:

calculating the sum of the spectral line intensities in the LIBS spectrum detection spectral band range obtained at different positions, passing through I_SUM＝f_SUM(h) Fitting the corresponding relation between the intensity and value of the spectrum and the position parameter of the plasma, wherein f_SUM(h) Is a fitting function.

5. The method of claim 4, wherein the fitting function is expressed as follows:

<mrow> <msub> <mi>f</mi> <mrow> <mi>S</mi> <mi>U</mi> <mi>M</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>h</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> <mrow> <mo>(</mo> <mi>sgn</mi> <mo>(</mo> <mrow> <mi>h</mi> <mo>-</mo> <msub> <mi>h</mi> <mn>0</mn> </msub> </mrow> <mo>)</mo> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>&amp;times;</mo> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mfrac> <msup> <mrow> <mo>(</mo> <mi>h</mi> <mo>-</mo> <msub> <mi>h</mi> <mn>0</mn> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mrow> <mn>2</mn> <msubsup> <mi>&amp;sigma;</mi> <mn>1</mn> <mn>2</mn> </msubsup> </mrow> </mfrac> </mrow> </msup> <mo>+</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> <mrow> <mo>(</mo> <mi>sgn</mi> <mo>(</mo> <mrow> <msub> <mi>h</mi> <mn>0</mn> </msub> <mo>-</mo> <mi>h</mi> </mrow> <mo>)</mo> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>&amp;times;</mo> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mfrac> <msup> <mrow> <mo>(</mo> <mi>h</mi> <mo>-</mo> <msub> <mi>h</mi> <mn>0</mn> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mrow> <mn>2</mn> <msubsup> <mi>&amp;sigma;</mi> <mn>2</mn> <mn>2</mn> </msubsup> </mrow> </mfrac> </mrow> </msup> </mrow>

wherein sgn (. cndot.) represents a sign function, h is a parameter representing plasma position information, h₀,σ₁,σ₂Is the offset parameter and the position parameter determined by fitting.

6. The LIBS spectrum correction method based on the plasma position information as claimed in claim 1, wherein the corresponding functional relationship between LIBS spectrum specific characteristic spectral line intensity and the plasma position information is established by a fitting method, specifically:

through I_λ＝f_λ(h) Determining the corresponding relation between the characteristic spectral line intensity and the plasma position parameter, wherein f_λ(h) Is a fitting function.

7. The method of claim 6, wherein the fitting function is expressed as follows:

<mrow> <msub> <mi>f</mi> <mi>&amp;lambda;</mi> </msub> <mrow> <mo>(</mo> <mi>h</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> <mrow> <mo>(</mo> <mi>s</mi> <mi>g</mi> <mi>n</mi> <mo>(</mo> <mrow> <mi>h</mi> <mo>-</mo> <msub> <mi>h</mi> <mn>0</mn> </msub> </mrow> <mo>)</mo> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>&amp;times;</mo> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mfrac> <msup> <mrow> <mo>(</mo> <mi>h</mi> <mo>-</mo> <msub> <mi>h</mi> <mn>0</mn> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mrow> <mn>2</mn> <msubsup> <mi>&amp;sigma;</mi> <mn>1</mn> <mn>2</mn> </msubsup> </mrow> </mfrac> </mrow> </msup> <mo>+</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> <mrow> <mo>(</mo> <mi>s</mi> <mi>g</mi> <mi>n</mi> <mo>(</mo> <mrow> <msub> <mi>h</mi> <mn>0</mn> </msub> <mo>-</mo> <mi>h</mi> </mrow> <mo>)</mo> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>&amp;times;</mo> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mfrac> <msup> <mrow> <mo>(</mo> <mi>h</mi> <mo>-</mo> <msub> <mi>h</mi> <mn>0</mn> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mrow> <mn>2</mn> <msubsup> <mi>&amp;sigma;</mi> <mn>2</mn> <mn>2</mn> </msubsup> </mrow> </mfrac> </mrow> </msup> </mrow>

wherein sgn (. cndot.) represents a sign function, h is a parameter representing plasma position information, h₀,σ₁,σ₂Is the offset parameter and the position parameter determined by fitting.

8. The method of claim 1, wherein the correction function is expressed as follows:

<mrow> <msubsup> <mi>I</mi> <mi>&amp;lambda;</mi> <mo>&amp;prime;</mo> </msubsup> <mo>=</mo> <mfrac> <mrow> <msub> <mi>I</mi> <mi>&amp;lambda;</mi> </msub> <mo>&amp;CenterDot;</mo> <msub> <mi>f</mi> <mrow> <mi>S</mi> <mi>U</mi> <mi>M</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>h</mi> <mo>)</mo> </mrow> </mrow> <mrow> <msub> <mi>I</mi> <mrow> <mi>S</mi> <mi>U</mi> <mi>M</mi> </mrow> </msub> <mo>&amp;CenterDot;</mo> <msub> <mi>f</mi> <mrow> <mi>S</mi> <mi>U</mi> <mi>M</mi> </mrow> </msub> <mrow> <mo>(</mo> <msub> <mi>h</mi> <mn>0</mn> </msub> <mo>)</mo> </mrow> <mo>&amp;CenterDot;</mo> <msub> <mi>f</mi> <mi>&amp;lambda;</mi> </msub> <mrow> <mo>(</mo> <mi>h</mi> <mo>)</mo> </mrow> </mrow> </mfrac> </mrow>

wherein, I'_λDenotes corrected line intensity, I_λIs the original line intensity, I_SUMIs the sum of the intensities of the original spectral bands, f_λ(h) Is a fitting function of corresponding function relationship between LIBS spectrum specific characteristic spectral line intensity and plasma position information, f_SUM(h) Is a fitting function of the corresponding function relationship between the full spectrum band in the LIBS spectrum acquisition range and the plasma position information, wherein h is a plasma position parameter extracted from a plasma image, h is a parameter of the plasma position extracted from the plasma image₀Is f_λ(h) H when the maximum value is obtained.