CN108169213B - Automatic identification method for peak elements of laser-induced breakdown spectroscopy - Google Patents

Automatic identification method for peak elements of laser-induced breakdown spectroscopy Download PDF

Info

Publication number
CN108169213B
CN108169213B CN201810174758.5A CN201810174758A CN108169213B CN 108169213 B CN108169213 B CN 108169213B CN 201810174758 A CN201810174758 A CN 201810174758A CN 108169213 B CN108169213 B CN 108169213B
Authority
CN
China
Prior art keywords
spectral
peak
identified
spectrum
spectral peak
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201810174758.5A
Other languages
Chinese (zh)
Other versions
CN108169213A (en
Inventor
朱启兵
檀兵
黄敏
郭亚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jiangnan University
Original Assignee
Jiangnan University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jiangnan University filed Critical Jiangnan University
Priority to CN201810174758.5A priority Critical patent/CN108169213B/en
Publication of CN108169213A publication Critical patent/CN108169213A/en
Application granted granted Critical
Publication of CN108169213B publication Critical patent/CN108169213B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/71—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light thermally excited
    • G01N21/718—Laser microanalysis, i.e. with formation of sample plasma

Landscapes

  • Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • Optics & Photonics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Plasma & Fusion (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Spectrometry And Color Measurement (AREA)
  • Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)

Abstract

The invention provides a method for automatically identifying peak elements of a laser-induced breakdown spectroscopy, which comprises the following steps: collecting a laser-induced breakdown spectrum to be identified, and correcting a continuous background of the spectrum; performing spectral peak fitting and decomposition on the spectrum after background correction by using a Voigt function, and constructing a characteristic parameter vector of a spectral peak signal to be identified; acquiring spectral information of all elements under the same wave band as the spectrum to be identified according to a standard database, and constructing a characteristic parameter vector of the elements; respectively carrying out similarity analysis on the characteristic parameter vector of each spectral peak signal to be identified and the characteristic vector of the element spectral peak in the standard database; and according to the similarity, judging the element corresponding to the spectrum peak with the maximum similarity in the standard database as the element to which the spectrum peak to be identified belongs. The method can be quickly and accurately suitable for identifying the spectral peaks of simple and complex spectrums.

Description

Automatic identification method for peak elements of laser-induced breakdown spectroscopy
Technical Field
The invention relates to the technical field of spectrum analysis of sample composition and content and the like, in particular to a method for automatically identifying peak elements of a laser-induced breakdown spectroscopy.
Background
The Laser Induced Breakdown Spectroscopy (LIBS) is a spectroscopic analysis technique and has wide application in the field of sample composition and content analysis. The LIBS technique is an analysis technique for obtaining a substance component (qualitative analysis) and a concentration (quantitative analysis) by irradiating a surface of a measured object with laser light to generate plasma.
Compared with the traditional spectral analysis method, the LIBS has the characteristics of real-time, rapid, nondestructive or micro-damage detection and the like.
The LIBS technology is used for analyzing an emission spectrum of atoms or ions, a laser-induced breakdown spectrum generally contains a large number of element characteristic spectral peaks, and accurate identification of the spectral peaks is a premise and a basis for element detection and analysis. For complex spectra, accurate identification of the effective spectral peaks in the spectra can provide an effective means for qualitative and quantitative analysis.
At present, the identification method of LIBS spectrum peak can be roughly classified into three types. The first type is called as the wavelength nearest neighbor rule, and the basic idea is to compare the wavelength information of the spectral peak in the experimental spectrum with the emission wavelength of the elements in the standard database, and the elements of the standard database which are closer to the wavelength of the spectral peak of the experimental spectrum can be judged as the corresponding elements of the wavelength. The second type is a simulated spectrum comparison method, and the main idea of the method is to obtain a fitting spectrum peak of related elements through an NIST standard database, and compare the fitting spectrum peak with the spectrum data obtained by experiments to observe and identify the spectrum peak elements. The third kind is correlation analysis method, which uses standard database to establish the simulated spectrum peak of each element, and makes correlation analysis with the actual spectrum data in a certain wavelength range, and determines the element attribution of the experimental spectrum peak according to the correlation size.
Due to the influence of factors such as material characteristics and experimental equipment, the nearest neighbor method has poor identification accuracy because the wavelengths of spectral peaks in experimental spectra have spectral peak shifts to a certain extent compared with a standard database. The method for comparing and analyzing the spectrum obtained by database simulation and the experimental spectrum still depends on the judgment of human eyes, which is a simulated spectrum comparison method, not only wastes time and labor, but also is difficult to ensure the reliability of identification for the more complex spectrum. Most of the correlation analysis methods simply use the spectral intensity near a spectral peak as a feature vector, do not consider the waveform structure of the spectral data, and do not consider the inherent physical relationship between the waveform structure and the parameter to be measured of the sample, and the obtained correlation coefficient is usually small (even negative correlation), so that the elements cannot be identified.
Therefore, under the conditions that the material components are complex and the spectrum peaks have more overlapped peak interferences, the accurate identification of the spectrum peak elements is always an urgent problem to be solved. On the basis of the existing research work, a novel method for automatically identifying the peak elements of the laser-induced breakdown spectroscopy is provided. The method comprises the steps of firstly, fitting a spectrum by utilizing a Voigt function to reduce spectral peak overlapping and background noise interference; on the basis, a spectral peak characteristic parameter vector including spectral peak center wavelength, spectral intensity, full width at half maximum and spectral peak mass center under the condition of corresponding center wavelength is constructed, and automatic identification of spectral peak elements is realized according to similarity analysis.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides an automatic identification method for spectral peak elements of a laser-induced breakdown spectrum, which comprises the steps of firstly fitting the spectrum by using a Voigt function to reduce spectral peak overlapping and background noise interference; on the basis, a spectral peak characteristic parameter vector including spectral peak center wavelength, spectral intensity, full width at half maximum and spectral peak mass center under the condition of corresponding center wavelength is constructed, and automatic identification of spectral peak elements is realized according to similarity analysis. The technical scheme adopted by the invention is as follows:
a method for automatically identifying peak elements of a laser-induced breakdown spectroscopy comprises the following steps:
step (a), collecting a laser-induced breakdown spectrum to be identified, and correcting a continuous background of the spectrum;
step (b), performing spectral peak fitting and decomposition on the spectrum after background correction by using a Voigt function, and constructing a characteristic parameter vector of a spectral peak signal to be identified;
acquiring spectral information of all elements under the same wave band with the spectrum to be identified according to a standard database, and constructing a characteristic parameter vector of the elements;
step (d), respectively carrying out similarity analysis on the characteristic parameter vector of each spectral peak signal to be identified and the characteristic vector of the element spectral peak in the standard database;
the similarity calculation is defined as follows:
Figure GDA0002454060490000021
wherein c isi,tRepresenting the similarity between the ith spectral peak signal to be identified and the spectral peak signal of the tth possible element in the standard database; fiCharacteristic parameter vector representing the ith spectral peak signal to be identified, FtA feature parameter vector representing a spectral peak signal of a tth possible element in the standard database;
and (e) according to the similarity, judging the element corresponding to the spectrum peak with the maximum similarity in the standard database as the element to which the spectrum peak to be identified belongs.
In step (b), the Voigt function is defined as follows:
Figure GDA0002454060490000022
wherein I (λ) represents the spectral intensity at wavelength λ; lambda [ alpha ]cIs the center wavelength, I, of the spectral peak signalcW represents the spectral intensity and the full width at half maximum of the spectral peak corresponding to the central wavelength respectively; theta is a Gauss-Lorentz coefficient and takes a constant between (0 and 1);
performing Voigt function fitting on a spectrum to be identified to obtain a spectrum peak centroid A as a substitute characteristic parameter of a waveform characteristic parameter theta; thus, for a spectral peak to be identified, its characteristic parameter vector F ═ I is obtainedc,λc,w,A]；
The centroid calculation is defined as follows:
Figure GDA0002454060490000023
wherein A represents the centroid of the spectral peak at wavelength λ, I (λ) represents the spectral intensity at wavelength λ, [ λ [ [ λ ]L,λR]The interval where the peak of the spectrum is located.
When there is a spectral peak overlap phenomenon, in order to obtain more accurate spectral peak characteristic parameters, the Voigt function is now replaced by the following formula:
Figure GDA0002454060490000031
wherein l is the number of overlapping spectral peaks, lambdat、It、wtAnd thetatRespectively is the center wavelength of the t-th spectral peak, the spectral intensity corresponding to the center wavelength, the full width at half maximum of the spectral peak and the Gaussian-Lorentz coefficient.
The invention has the advantages that:
1) the method for automatically identifying the peak elements of the laser-induced breakdown spectroscopy provided by the invention is executed by a computer program, does not need human participation, and can automatically identify the effective spectral peak in the whole spectral range.
2) The method can be quickly and accurately suitable for identifying the spectral peaks of simple and complex spectrums, and is an effective means for spectrum analysis.
3) The method and the device consider a plurality of characteristic information of the spectral peak, and can effectively improve the accuracy of spectral peak identification.
4) The invention not only realizes the spectral peak identification of the laser-induced breakdown spectrum, but also corrects the background and overlapping peaks in the spectrum, thereby extracting useful information in the spectrum.
Drawings
FIG. 1 is a flow chart of the implementation steps of the present invention.
FIG. 2 is a schematic diagram of determining an overlapping peak according to the present invention.
FIG. 3 is a graph illustrating the results of decomposition reconstruction of a standard simulated spectrum according to an embodiment of the present invention.
Fig. 4 is a diagram illustrating the peak identification result of a simulated spectrum according to an embodiment of the present invention.
FIG. 5 is a graph showing the result of decomposition and reconstruction of a spectrum of tea according to an embodiment of the present invention.
Fig. 6 is a diagram illustrating the peak identification result of the spectrum of tea according to an embodiment of the present invention.
Detailed Description
The invention is further illustrated by the following specific figures and examples.
The invention provides a method for automatically identifying peak elements of a laser-induced breakdown spectrum, which can be used for carrying out continuous background correction and overlapped spectrum peak decomposition on an original spectrum to be identified before spectrum peak identification, wherein the background spectrum and the overlapped spectrum peak can seriously interfere extraction of spectrum information to be identified, and thus the accuracy of spectrum peak identification can be influenced. The method comprises the following specific steps:
step (a), collecting a laser-induced breakdown spectrum to be identified, and correcting a continuous background of the spectrum;
step (b), performing spectral peak fitting and decomposition on the spectrum after background correction by using a Voigt function, and constructing a characteristic parameter vector of a spectral peak signal to be identified; the Voigt function is defined as follows:
Figure GDA0002454060490000032
wherein I (λ) represents the spectral intensity at wavelength λ; lambda [ alpha ]cIs the center wavelength, I, of the spectral peak signalcW represents the spectral intensity and the full width at half maximum of the spectral peak corresponding to the central wavelength respectively; theta is a Gauss-Lorentz coefficient and takes a constant between (0 and 1); the above formula (1) indicates that: fitting the measured laser induced spectral peak with Voigt function can obtain lambdac、IcW and theta are 4 waveform characteristic parameters; using lambdac、IcW and theta can construct a characteristic parameter vector of a spectral peak signal to be identified;
in constructing the feature parameter vector, λc、IcW is three important parameters reflecting the waveform characteristics of the spectral peaks and can be obtained by consulting a NIST (national institute of standards and technology) standard database; values of all elements theta are not provided in the NIST standard database, and the fact that the value of the waveform characteristic parameter theta mainly influences the distribution size of all light intensities near a spectral peak is considered, so that a spectral peak centroid A obtained after Voigt function fitting is carried out on a spectrum to be identified is used as a substitute characteristic parameter of the waveform characteristic parameter theta; thus, for a spectral peak to be identified, its characteristic parameter vector F ═ I can be obtainedc,λc,w,A]；
The centroid calculation is defined as follows:
Figure GDA0002454060490000041
wherein A represents the centroid of the spectral peak at wavelength λ, I (λ) represents the spectral intensity at wavelength λ, [ λ [ [ λ ]L,λR]Is the interval where the spectrum peak is located;
when the Voigt function is used for carrying out spectrum peak fitting and decomposition on a spectrum to be identified, under the condition that all spectrum peaks are not overlapped, the characteristic parameters of the spectrum peaks can be well obtained by using the Voigt function; however, since the spectral peak overlap is a general phenomenon in practical measurement, in order to obtain more accurate spectral peak characteristic parameters, the Voigt function can be replaced by the following formula:
Figure GDA0002454060490000042
wherein l is the number of overlapping spectral peaks, lambdat、It、wtAnd thetatRespectively the center wavelength of the t-th spectral peak, the spectral intensity corresponding to the center wavelength, the full width at half maximum of the spectral peak and the Gaussian-Lorentz coefficient;
the specific implementation of the construction of the characteristic parameter vector of the spectral peak signal to be identified is as follows:
using the maximum value method of adjacent points to identify the spectral interval [ lambda ]min,λmax]Searching is carried out to obtain spectrum extreme value information
Figure GDA0002454060490000043
N is the number of the extreme points,
Figure GDA0002454060490000044
respectively the wavelength position and the spectrum peak intensity at the jth extreme point;
obtained extreme point
Figure GDA0002454060490000045
The spectrum peak information corresponding to real elements and weak spectrum peaks with low noise and intensity (in practice, the spectrum peaks are usually not identified);
removing noise and weak spectral peaks with low intensity, namely invalid spectral peaks; background intensity value pair extreme point capable of introducing original signal
Figure GDA0002454060490000046
Judging and screening; is provided with
Figure GDA0002454060490000047
As the position of the spectrum
Figure GDA0002454060490000048
Corresponding background light intensity value if
Figure GDA0002454060490000049
Then extreme point
Figure GDA00024540604900000410
Identifying and judging the effective spectral peak to be identified, otherwise discarding; t ishTo select the threshold, it may be selected according to the specific identification needs;
after removing noise and invalid spectral peaks, all the remaining extreme points (spectral peaks to be identified) are recorded as
Figure GDA00024540604900000411
M is the number of the spectral peaks to be identified,
Figure GDA00024540604900000412
is the wavelength and the spectrum peak intensity of the ith spectrum peak to be identified;
judging the spectral peak to be identified
Figure GDA00024540604900000413
Whether it is an overlapping spectral peak; is provided with
Figure GDA00024540604900000414
Respectively the spectral peak interval
Figure GDA0002454060490000051
And
Figure GDA0002454060490000052
the position of the minimum value of the light intensity and the intensity thereof (see fig. 2); if its slope
Figure GDA0002454060490000053
Less than a given threshold value TkThen the spectral peak to be identified can be judged
Figure GDA0002454060490000054
Is a non-overlapping spectral peak and records the interval of the spectral peak to be fitted as
Figure GDA0002454060490000055
If K is greater than or equal to TkThen the spectral peak to be identified can be judged
Figure GDA0002454060490000056
And
Figure GDA0002454060490000057
for overlapping spectral peaks, in
Figure GDA0002454060490000058
Finding light intensity minimum value position and intensity in spectral peak interval
Figure GDA0002454060490000059
Calculating slope
Figure GDA00024540604900000510
If K is greater than or equal to TkThis process is repeated until the slope is less than a given threshold TkRecording the peak interval of the spectrum to be fitted as
Figure GDA00024540604900000511
l is the number of peaks in the region to be fitted;
in the interval of the peak of the spectrum to be fitted
Figure GDA00024540604900000512
Fitting the spectral curve by using a formula (2) to obtain It，wt，λt，θt(t ═ 1,2, …, l) characteristic parameters, where t ═ 1,2, …, l represent the ith, i +1, …, i + l spectral peaks to be identified, respectively;
repeating the above steps for the ith spectral peak to be identified
Figure GDA00024540604900000513
Characteristic parameter vector F is constructed by utilizing characteristic parameters obtained by Voigt function decompositioni＝[Ii,λi,wi,Ai]；
Acquiring spectral information of all elements under the same wave band with the spectrum to be identified according to an NIST standard database, and constructing a characteristic parameter vector of the elements;
when constructing the feature parameter vector of an element using the NIST standard database, since the plasma electron density Ne and the electron temperature T are unknown, this gives I in the feature parameter vectorcThe calculation of w presents difficulties; grid search and similarity analysis method are combined to calculate I in feature parameter vectorcW; the method comprises the following specific steps:
the value ranges of the plasma electron density Ne and the electron temperature T are set as follows: log (ne) ═ 15,20], T ═ 0.5,2] (eV); dividing log (Ne) by step 1, T by step 0.25 within a given range, resulting in a total of 42 combinations of plasma electron density Ne and electron temperature T; thus obtaining:
uj＝(log(Ne),T)∈{(15,0.5),(15,0.75),…,(15,2),(16,0.5),…,(20,2)},j＝1,…,42
for any given one uj(log (Ne), T) in combination, depending on the wavelength information λ of the spectral peak to be identifiediObtaining [ lambda ] in NIST standard databasei-Δλ,λi+Δλ]Calculating and obtaining the spectral peak characteristic parameter vector of all possible S elements in the range
Figure GDA00024540604900000514
t-1, 2, … S, where Δ λ is the possible wavelength shift of the experimental spectrum from the standard spectrum, selected here as 0.2;
step (d), respectively carrying out similarity analysis on the characteristic parameter vector of each spectral peak signal to be identified and the characteristic vector of the element spectral peak in the standard database;
the similarity calculation is defined as follows:
Figure GDA00024540604900000515
wherein c isi,tRepresenting the similarity between the ith spectral peak signal to be identified and the spectral peak signal of the tth possible element in the standard database; fiCharacteristic parameter vector representing the ith spectral peak signal to be identified, FtA feature parameter vector representing a spectral peak signal of a tth possible element in the standard database;
the concrete implementation is as follows:
calculating the ith spectral peak signal feature vector F to be identified by adopting the similarity of Cosini＝[Ii,λi,wi,Ai]And element characteristic parameter vector in NIST standard database
Figure GDA0002454060490000061
The similarity of (2); the expression is as follows:
Figure GDA0002454060490000062
wherein
Figure GDA0002454060490000063
Expressed in the plasma density and electron temperature combination ujUnder the condition, the ith spectral peak signal characteristic parameter vector F to be identifiediAnd the tth possible element in the NIST standard database
Figure GDA0002454060490000064
Cosin similarity of (a);
and (e) according to the similarity, judging the element corresponding to the spectrum peak with the maximum similarity in the standard database as the element to which the spectrum peak to be identified belongs.
The concrete implementation is as follows:
according to the similarity, the feature vector F of the spectral peak signal to be identified is compared with the feature vector F of the spectral peak signal to be identifiediWith maximum similarity
Figure GDA0002454060490000065
The NIST standard database elements are used as corresponding elements of the spectral peaks to be identified and are recorded as
Figure GDA0002454060490000066
The similarity is recorded as
Figure GDA0002454060490000067
Repeating the steps (c) and (d) to obtain ujUnder the conditions of log (Ne) and T, the element identification results and the similarity of all the spectral peaks to be identified
Figure GDA0002454060490000068
And obtaining the average similarity thereof
Figure GDA0002454060490000069
Repeating the above steps (c), (d) and (e) to obtain all ujThe average similarity under the condition of (log (Ne), T) is selected, and u corresponding to the maximum average similarity is selectedjFor optimal plasma electron density and electron temperature; the element recognition result under this condition is the final result.
In one embodiment, the spectral peak identification of the analog spectrum and the tea spectrum is realized, and the spectral peak identification of the spectrum can be better carried out by the automatic identification method of the spectral peak elements of the laser-induced breakdown spectrum; effective spectral peak information in the spectrum to be identified can be extracted after background correction and overlapped peak decomposition, interference of noise and weak spectral peaks with low intensity is removed, and identification of effective spectral peaks is achieved.
See FIGS. 3-6, which are graphs of the effect of spectral peak identification on partial spectra in this example: FIG. 3 is a graph showing the results of background correction and overlapping peak decomposition of a simulated spectrum obtained from a standard database; wherein, the broken line represents the reconstructed spectrum obtained after the effective spectral peak information is linearly superposed; FIG. 4 is a graph showing the identification of spectral peaks in a simulated spectrum; wherein the thin dashed line is marked as a missing judgment occurring in the spectral peak identification process of the element, and the thick dashed line is marked as a false judgment. FIG. 5 is a graph showing the results of background correction and overlapping peak decomposition of the spectra of tea leaves; wherein, the broken line represents the reconstructed spectrum obtained after the effective spectral peak information is linearly superposed; fig. 6 shows the identification result of the effective spectrum peak in the tea spectrum. From the above figures, the present invention can better identify the spectrum peak in the spectrum, and can automatically identify the element spectrum peak for the simple spectrum and the complex spectrum with the overlapping spectrum peak and the noise influence.
Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention has been described in detail with reference to examples, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.

Claims (6)

1. A method for automatically identifying peak elements of a laser-induced breakdown spectroscopy is characterized by comprising the following steps:
step (a), collecting a laser-induced breakdown spectrum to be identified, and correcting a continuous background of the spectrum;
step (b), performing spectral peak fitting and decomposition on the spectrum after background correction by using a Voigt function, and constructing a characteristic parameter vector of a spectral peak signal to be identified;
acquiring spectral information of all elements under the same wave band with the spectrum to be identified according to a standard database, and constructing a characteristic parameter vector of the elements;
step (d), respectively carrying out similarity analysis on the characteristic parameter vector of each spectral peak signal to be identified and the characteristic vector of the element spectral peak in the standard database;
the similarity calculation is defined as follows:
Figure FDA0002669276380000011
wherein c isi,tRepresenting the similarity between the ith spectral peak signal to be identified and the spectral peak signal of the tth possible element in the standard database; fiCharacteristic parameter vector representing the ith spectral peak signal to be identified, FtFeature parameters representing spectral peak signals of the tth possible element in the normative databaseA number vector;
step (e), according to the similarity, judging the element corresponding to the spectrum peak with the maximum similarity in the standard database as the element belonging to the spectrum peak to be identified;
in step (b), the Voigt function is defined as follows:
Figure FDA0002669276380000012
wherein I (λ) represents the spectral intensity at wavelength λ; lambda [ alpha ]cIs the center wavelength, I, of the spectral peak signalcW represents the spectral intensity and the full width at half maximum of the spectral peak corresponding to the central wavelength respectively; theta is a Gauss-Lorentz coefficient and takes a constant between (0 and 1);
performing Voigt function fitting on a spectrum to be identified to obtain a spectrum peak centroid A as a substitute characteristic parameter of a waveform characteristic parameter theta; thus, for a spectral peak to be identified, its characteristic parameter vector F ═ I is obtainedc,λc,w,A]；
The centroid calculation is defined as follows:
Figure FDA0002669276380000013
wherein A represents the centroid of the spectral peak at wavelength λ, I (λ) represents the spectral intensity at wavelength λ, [ λ [ [ λ ]L,λR]The interval where the peak of the spectrum is located.
2. The method for automatically identifying peak elements in a laser-induced breakdown spectroscopy of claim 1,
in order to obtain more accurate characteristic parameters of the spectral peaks when the spectral peak overlapping phenomenon exists, the Voigt function is replaced by the following formula:
Figure FDA0002669276380000021
wherein l is an overlapNumber of spectral peaks, λt、It、wtAnd thetatRespectively is the center wavelength of the t-th spectral peak, the spectral intensity corresponding to the center wavelength, the full width at half maximum of the spectral peak and the Gaussian-Lorentz coefficient.
3. The method for automatically identifying peak elements in a laser-induced breakdown spectroscopy of claim 2,
in the step (b), constructing a characteristic parameter vector of a spectral peak signal to be identified, which is specifically realized as follows:
using the maximum value method of adjacent points to identify the spectral interval [ lambda ]min,λmax]Searching is carried out to obtain spectrum extreme value information
Figure FDA0002669276380000022
N is the number of the extreme points,
Figure FDA0002669276380000023
respectively the wavelength position and the spectrum peak intensity at the jth extreme point;
removing noise and weak spectral peaks with low intensity, namely invalid spectral peaks;
after removing noise and invalid spectral peaks, all the retained extreme points, i.e. spectral peaks to be identified, are recorded as
Figure FDA0002669276380000024
M is the number of the spectral peaks to be identified,
Figure FDA0002669276380000025
is the wavelength and the spectrum peak intensity of the ith spectrum peak to be identified;
judging the spectral peak to be identified
Figure FDA0002669276380000026
Whether it is an overlapping spectral peak; is provided with
Figure FDA0002669276380000027
Respectively the spectral peak interval
Figure FDA0002669276380000028
And
Figure FDA0002669276380000029
the position of the minimum value of the light intensity and the intensity thereof; if its slope
Figure FDA00026692763800000210
Less than a given threshold value TkThen the spectral peak to be identified can be judged
Figure FDA00026692763800000211
Is a non-overlapping spectral peak and records the interval of the spectral peak to be fitted as
Figure FDA00026692763800000212
If K is greater than or equal to TkThen the spectral peak to be identified can be judged
Figure FDA00026692763800000213
And
Figure FDA00026692763800000214
for overlapping spectral peaks, in
Figure FDA00026692763800000215
Finding light intensity minimum value position and intensity in spectral peak interval
Figure FDA00026692763800000216
Calculating slope
Figure FDA00026692763800000217
If K is greater than or equal to TkThis process is repeated until the slope is less than a given threshold TkRecording the peak interval of the spectrum to be fitted as
Figure FDA00026692763800000218
l is the number of peaks in the region to be fitted;
in the interval of the peak of the spectrum to be fitted
Figure FDA00026692763800000219
Fitting the spectral curve by using a formula (2) to obtain It，wt，λt，θt(t ═ 1,2, …, l) characteristic parameters, where t ═ 1,2, …, l represent the ith, i +1, …, i + l spectral peaks to be identified, respectively;
repeating the above steps for the ith spectral peak to be identified
Figure FDA00026692763800000220
Characteristic parameter vector F is constructed by utilizing characteristic parameters obtained by Voigt function decompositioni＝[Ii,λi,wi,Ai]。
4. The method for automatically identifying peak elements in a laser-induced breakdown spectroscopy of claim 3,
in the step (c), according to the NIST standard database, a feature parameter vector of the element is constructed, which specifically includes:
setting the value ranges of the plasma electron density Ne and the electron temperature T: log (Ne) and T; dividing log (Ne) and T in a given range according to respective step sizes to obtain a plurality of combinations of plasma electron density Ne and electron temperature T; thus obtaining: u. ofj＝(log(Ne),T)；
For any given one uj(log (Ne), T) in combination, depending on the wavelength information λ of the spectral peak to be identifiediObtaining [ lambda ] in NIST standard databasei-Δλ,λi+Δλ]Calculating and obtaining the spectral peak characteristic parameter vector of all possible S elements in the range
Figure FDA0002669276380000031
Where Δ λ is the possible wavelength shift of the experimental spectrum relative to the standard spectrum.
5. The method for automatically identifying peak elements of laser-induced breakdown spectroscopy of claim 4, wherein,
the step (d) is specifically realized as follows:
calculating the ith spectral peak signal feature vector F to be identified by adopting the similarity of Cosini＝[Ii,λi,wi,Ai]And element characteristic parameter vector in NIST standard database
Figure FDA0002669276380000032
The similarity of (2); the expression is as follows:
Figure FDA0002669276380000033
wherein
Figure FDA0002669276380000034
Expressed in the plasma density and electron temperature combination ujUnder the condition, the ith spectral peak signal characteristic parameter vector F to be identifiediAnd the tth possible element in the NIST standard database
Figure FDA0002669276380000035
Cosin similarity of (a).
6. The method for automatically identifying peak elements of laser-induced breakdown spectroscopy of claim 5, wherein,
the step (e) is specifically realized as follows:
according to the similarity, the feature vector F of the spectral peak signal to be identified is compared with the feature vector F of the spectral peak signal to be identifiediWith maximum similarity
Figure FDA0002669276380000036
The NIST standard database elements are used as corresponding elements of the spectral peaks to be identified and are recorded as
Figure FDA0002669276380000037
The similarity is recorded as
Figure FDA0002669276380000038
Repeating the steps (c) and (d) to obtain ujUnder the conditions of log (Ne) and T, the element identification results and the similarity of all the spectral peaks to be identified
Figure FDA0002669276380000039
And obtaining the average similarity thereof
Figure FDA00026692763800000310
Repeating the above steps (c), (d) and (e) to obtain all ujThe average similarity under the condition of (log (Ne), T) is selected, and u corresponding to the maximum average similarity is selectedjFor optimal plasma electron density and electron temperature; the element recognition result under this condition is the final result.
CN201810174758.5A 2018-03-02 2018-03-02 Automatic identification method for peak elements of laser-induced breakdown spectroscopy Active CN108169213B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810174758.5A CN108169213B (en) 2018-03-02 2018-03-02 Automatic identification method for peak elements of laser-induced breakdown spectroscopy

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810174758.5A CN108169213B (en) 2018-03-02 2018-03-02 Automatic identification method for peak elements of laser-induced breakdown spectroscopy

Publications (2)

Publication Number Publication Date
CN108169213A CN108169213A (en) 2018-06-15
CN108169213B true CN108169213B (en) 2020-12-15

Family

ID=62511641

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810174758.5A Active CN108169213B (en) 2018-03-02 2018-03-02 Automatic identification method for peak elements of laser-induced breakdown spectroscopy

Country Status (1)

Country Link
CN (1) CN108169213B (en)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109596558B (en) * 2018-12-17 2020-05-19 华中科技大学 Spectrogram basic dimension correction and differential analysis method based on moving least square method
EP3674675A1 (en) * 2018-12-27 2020-07-01 INESC TEC - Instituto de Engenharia de Sistemas e Computadores, Tecnologia e Ciência A calibration method of a spectroscopy device comprising a plurality of sensors and of transfer of spectral information obtained from at least two calibrated spectroscopy devices
CN113075136A (en) * 2020-01-03 2021-07-06 应急管理部化学品登记中心 Device and method for identifying sulfur-containing substances in fire comburent
CN111666532B (en) * 2020-05-14 2023-04-07 山东省科学院海洋仪器仪表研究所 Peak-splitting calculation method for seawater radionuclide peak overlapping
US11614408B2 (en) 2020-05-19 2023-03-28 Jiangnan University Method for improving identification accuracy of mixture components by using known mixture Raman spectrum
CN111504979B (en) * 2020-05-19 2021-03-16 江南大学 Method for improving mixture component identification precision by using Raman spectrum of known mixture
CN112557332B (en) * 2020-12-03 2023-10-03 华东理工大学 Spectrum segmentation and spectrum comparison method based on spectrum peak-splitting fitting
CN113008874B (en) * 2021-03-11 2022-07-26 合肥工业大学 Method for improving qualitative detection capability of laser-induced breakdown spectroscopy technology based on baseline correction and spectral peak recognition
CN113848594A (en) * 2021-09-23 2021-12-28 云南电网有限责任公司电力科学研究院 Method for identifying spectral information of lightning channel
CN116609319B (en) * 2023-06-09 2024-04-09 中国海洋大学 Element type pre-screening method and device for LIBS spectrum element identification
CN117332258B (en) * 2023-12-01 2024-01-30 奥谱天成(成都)信息科技有限公司 Near infrared absorption peak identification method, system and medium based on multi-scale Lorentz

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103217404A (en) * 2013-03-30 2013-07-24 中国科学院安徽光学精密机械研究所 Method for identifying affiliations of spectrum lines of elements by laser-induced breakdown spectroscopy

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103063621B (en) * 2011-10-21 2015-02-11 中国科学院沈阳自动化研究所 Element identification method based on light spectrum correlation
AU2015275734B2 (en) * 2014-06-20 2020-10-08 National Research Council Of Canada Method for laser-induced breakdown spectroscopy and calibration
CN104697965A (en) * 2015-03-10 2015-06-10 西北大学 Method for recognizing slag variety by combining with laser-induced breakdown spectroscopy based on least squares support vector machine
CN106596513A (en) * 2017-01-19 2017-04-26 江南大学 Tea leaf variety identification method based on laser induced breakdown spectroscopy
CN107677640B (en) * 2017-09-29 2018-07-27 吉林大学 A kind of plastic sample elemental recognition method based on laser induced breakdown spectroscopy

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103217404A (en) * 2013-03-30 2013-07-24 中国科学院安徽光学精密机械研究所 Method for identifying affiliations of spectrum lines of elements by laser-induced breakdown spectroscopy

Also Published As

Publication number Publication date
CN108169213A (en) 2018-06-15

Similar Documents

Publication Publication Date Title
CN108169213B (en) Automatic identification method for peak elements of laser-induced breakdown spectroscopy
CN105334279B (en) A kind of processing method of high resolution mass spectrum data
JP6091493B2 (en) Spectroscopic apparatus and spectroscopy for determining the components present in a sample
CN109993155A (en) For the characteristic peak extracting method of low signal-to-noise ratio uv raman spectroscopy
CN110672582B (en) Raman characteristic spectrum peak extraction method based on improved principal component analysis
CN108956583A (en) Characteristic spectral line automatic selecting method for laser induced breakdown spectroscopy analysis
CN113340874B (en) Quantitative analysis method based on combination ridge regression and recursive feature elimination
CN103543132B (en) A kind of coal characteristic measuring method based on wavelet transformation
CN113008805A (en) Radix angelicae decoction piece quality prediction method based on hyperspectral imaging depth analysis
CN113237836A (en) Flue-cured tobacco leaf moisture content estimation method based on hyperspectral image
CN110032988A (en) Uv raman spectroscopy system real-time noise-reducing Enhancement Method
CN105223140A (en) The method for quickly identifying of homology material
CN114611582A (en) Method and system for analyzing substance concentration based on near infrared spectrum technology
CN112485238B (en) Method for identifying turmeric essential oil producing area based on Raman spectrum technology
Charlton et al. Improving the technological readiness of time of Flight-Secondary Ion Mass Spectrometry for enhancing fingermark recovery-towards operational deployment
CN116818687B (en) Soil organic carbon spectrum prediction method and device based on spectrum guide integrated learning
CN117629971A (en) Method for improving precision of measuring carbon element in shale by laser-induced breakdown spectroscopy
CN110836878A (en) Convolution interpolation coupling Gaussian mixture model rapid three-dimensional fluorescence peak searching method
CN113406037B (en) Infrared spectrum online rapid identification analysis method based on sequence forward selection
WO2023123329A1 (en) Method and system for extracting net signal in near-infrared spectrum
CN114002203A (en) Method and device for analyzing content of wood components based on Raman spectrum
CN114283338A (en) Automatic chromatogram analysis method based on image recognition
JP2022013285A (en) Machine learning method
CN111624193B (en) LIBS (laser-induced breakdown Spectroscopy) identification method and system for wild gentiana rigescens
CN109145887A (en) A kind of Threshold Analysis method for obscuring differentiation based on spectrum latent variable

Legal Events

Date Code Title Description
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant