CN111766260B - Method for rapidly generating X fluorescence spectrum based on genetic algorithm optimization basic parameters - Google Patents

Method for rapidly generating X fluorescence spectrum based on genetic algorithm optimization basic parameters Download PDF

Info

Publication number
CN111766260B
CN111766260B CN202010541483.1A CN202010541483A CN111766260B CN 111766260 B CN111766260 B CN 111766260B CN 202010541483 A CN202010541483 A CN 202010541483A CN 111766260 B CN111766260 B CN 111766260B
Authority
CN
China
Prior art keywords
spectrum
ray
wavelength
incident
analysis
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
CN202010541483.1A
Other languages
Chinese (zh)
Other versions
CN111766260A (en
Inventor
李福生
马捷思
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.)
University of Electronic Science and Technology of China
Original Assignee
University of Electronic Science and Technology of China
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 University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN202010541483.1A priority Critical patent/CN111766260B/en
Publication of CN111766260A publication Critical patent/CN111766260A/en
Application granted granted Critical
Publication of CN111766260B publication Critical patent/CN111766260B/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
    • G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
    • G01N23/22—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by measuring secondary emission from the material
    • G01N23/223—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by measuring secondary emission from the material by irradiating the sample with X-rays or gamma-rays and by measuring X-ray fluorescence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2223/00—Investigating materials by wave or particle radiation
    • G01N2223/07—Investigating materials by wave or particle radiation secondary emission
    • G01N2223/076—X-ray fluorescence
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2223/00—Investigating materials by wave or particle radiation
    • G01N2223/30—Accessories, mechanical or electrical features
    • G01N2223/305—Accessories, mechanical or electrical features computer simulations

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Genetics & Genomics (AREA)
  • Data Mining & Analysis (AREA)
  • Analytical Chemistry (AREA)
  • Chemical & Material Sciences (AREA)
  • Physiology (AREA)
  • Pathology (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Computational Linguistics (AREA)
  • Biochemistry (AREA)
  • Evolutionary Computation (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Immunology (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)

Abstract

The invention provides a method for rapidly generating an X-ray fluorescence spectrum based on genetic algorithm optimization basic parameters, and relates to the field of X-ray fluorescence spectroscopyA domain. Firstly, establishing an optimal instrument factor G of No. 12-92 elementsi'database', then generating discrete spectrum of each analysis element of any sample based on Sherman equation forward direction, broadening into continuous spectrum through Gaussian function, and based on optimal instrument factor G corresponding to each analysis elementiRapidly generating successive X-ray fluorescence spectra of the sample, wherein each analytical element has an optimum instrument factor Gi' optimization by genetic Algorithm based on Instrument factor GiThe mean square prediction error of the continuous spectrum of the standard sample and the actually measured spectrum is less than 0.2, and the instrument factor of each analysis element is the optimal instrument factor Gi'. The invention realizes the generation of multi-element sample spectra with random composition at the speed of second order, and constructs X-ray fluorescence spectrum information with a large amount of data at extremely low time cost.

Description

Method for rapidly generating X fluorescence spectrum based on genetic algorithm optimization basic parameters
Technical Field
The invention relates to the field of X-ray fluorescence spectroscopy, in particular to a method for rapidly generating an X-ray fluorescence spectrum based on genetic algorithm optimization basic parameters.
Background
With the gradual development of the spectrum science research and the continuous expansion of the application field of the spectrum analysis technology, the actual demand of the sample component analysis technology based on the spectrum information is larger and larger, and the rapidity and the accuracy in the qualitative and quantitative detection technology become two key factors. Therefore, it is important how to generate an X-Ray Fluorescence spectrum (XRF) rapidly, accurately and nondestructively and apply the XRF to qualitative and quantitative analysis.
The traditional method for generating the X-ray fluorescence spectrum is mainly realized by a Monte Carlo simulation program and an experimental measurement standard sample, and has the problems of long time consumption, low detection efficiency and the like. The basic parameter algorithm (FP) based on Sherman equation is applied to the field of X-ray fluorescence spectrum analysis, so that the problems in the traditional method for generating the X-ray fluorescence spectrum are solved, and the method has the following advantages: (1) the X-ray fluorescence spectrum generation speed is high, the problem of long time consumption of Monte Carlo simulation calculation is not involved, and a spectrum information database of multi-component elements is favorably established; (2) the problem of spectral interference between adjacent element spectral lines is solved, and the accuracy of element qualitative and quantitative analysis is improved; (3) compared with an experimental measurement method, a standard sample does not need to be purchased, the problems that part of samples are expensive and not easy to purchase are not involved, and the detection and analysis cost is reduced; (4) the background absorption of the obtained XRF spectrum is basically negligible, the subsequent analysis difficulty is reduced, and the detection efficiency is improved.
The Sherman equation is currently used only for the calculation of the relative intensities and for the correction of the matrix effect in X-ray fluorescence spectroscopy, whereas the instrument factor G is used for the calculation of the relative intensitiesiCan be eliminated, so there is no instrument factor G yetiThe calculation and research work of (1). In addition, relative intensity calculations do not allow for the direct and rapid generation of X-ray fluorescence spectra via mathematical models and physical processes, and do not satisfy the requirements of building large databases in XRF elemental spectroscopy.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a method for rapidly generating an X-ray fluorescence spectrum based on genetic algorithm optimization basic parameters, establishes a theoretical model for rapidly generating a continuous X-ray fluorescence spectrum, realizes generation of a multi-element sample spectrum with any composition at a speed of second level, and constructs X-ray fluorescence spectrum information with a large amount of data at extremely low time cost.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a method for rapidly generating an X fluorescence spectrum based on optimization of basic parameters by a genetic algorithm is characterized by comprising the following steps:
step 1: preliminarily generating an X-ray fluorescence spectrum of a standard sample according to a Sherman equation, comparing the X-ray fluorescence spectrum with actually measured spectrum data of the standard sample, and optimizing the instrument factor G of each analysis element in the standard sample based on a genetic algorithmiFurther obtaining the optimal instrument factor G of No. 12-92 elements in the periodic tablei', and establishing an optimal instrument factor Gi' database, the concrete steps are as follows:
step 1.1: introducing content information of the standard sample, and converting the content information of each analysis element in the standard sample into emission spectrum intensity I according to Sherman equationi(λi) And generating a discrete spectrum (Flux); wherein i is an analysis element, namely an element with the sequence number of 12-92 in the standard sample, and lambdaiWavelength of the characteristic X-ray for the analysis element i;
step 1.2: by Gaussian functionThe discrete spectrum I of each analysis element in the standard sample obtained in the step 1.1i(λi) Spread out as a continuous spectrum (Spectrometry)
Figure BDA0002539000820000021
Step 1.3: assuming that the initial value of the instrument factor of each analysis element in the standard sample is 1, the continuous spectrum of each analysis element in the standard sample is obtained
Figure BDA0002539000820000022
Multiplying the continuous spectrum I (lambda) of the standard sample by the corresponding instrument factor and performing accumulative calculation to obtain the continuous spectrum I (lambda) of the standard samplei) The formula is as follows:
Figure BDA0002539000820000023
step 1.4: continuous Spectrum I (. lamda.) of the Standard sample obtained in step 1.3i) The Mean Square prediction Error (EMSPE) of the spectrum I actually measured by the standard sample is used as an evaluation function, and the instrument factor G of the analysis element I in the evaluation function is optimized through a genetic algorithmiSetting the optimized optimal instrument factor of the Fe element as a reference 1, and obtaining the optimized optimal instrument factor G of other analysis elements in proportioni', finally establishing the optimal instrument factor G of No. 12-92 elements in the periodic tablei' a database; wherein, i is 1 … N, and N is the sum of the numbers of the analysis elements in the standard sample;
step 2: inputting the content information of the sample to be detected, and generating the emission spectrum intensity I of each analysis element in the sample to be detected according to the method in the step 1.1i′(λi) And generating a discrete spectrum;
and step 3: respectively broadening the discrete spectrum of each analysis element in the sample to be detected obtained in the step 2 into a continuous spectrum through a Gaussian function
Figure BDA0002539000820000024
And 4, step 4: optimal instrument factor G based on each analysis element obtained in step 1i' and continuous Spectrum obtained in step 3
Figure BDA0002539000820000025
Calculating to obtain the continuous X-ray fluorescence spectrum I' (lambda) of the sample to be measuredi) The formula is as follows:
Figure BDA0002539000820000026
further, in step 1.1, the content information of the analysis element i is converted into the wavelength lambda according to Sherman equationiCharacteristic X-ray emission spectral intensity Ii(λi) The conversion expression of (1) is:
Figure BDA0002539000820000027
ki＝Ji·ωi·pi (4)
Figure BDA0002539000820000028
Figure BDA0002539000820000031
Figure BDA0002539000820000032
Figure BDA0002539000820000033
μs(λj)＝Ciμi(λj)+Cjμj(λj) (9)
wherein i is an analysis element; j is a momentArray elements, namely other elements except the analysis element i in the standard sample; lambda [ alpha ]iWavelength of the characteristic X-ray for the analysis element i; i isi(λi) For analysis of element i the emission wavelength is lambdaiThe spectral intensity of the characteristic X-ray of (a); giAn instrument factor to analyze element i; ci、CjRespectively mass fractions of an analysis element i and a matrix element j in a standard sample; k is a radical ofiTo analyze the excitation factor of element i as the fluorescence yield ωiSpectral line fraction piAnd absorption transition factor JiThe product of the three; lambda [ alpha ]0Is the minimum wavelength of the incident X-rays; lambda [ alpha ]edgeiIs the absorption edge wavelength of the incident X-ray for the analysis element i; lambda is the incident wavelength of the incident X-ray and ranges from 0.177A to 0.413A; i iso(λ) is the spectral intensity distribution of the incident wavelength λ; mu.si(λ) is the mass absorption coefficient of the analytical element i at the incident wavelength λ; s is a representation of the interaction of elements; mu's's(λ) is the effective mass absorption coefficient of the standard sample at the incident wavelength λ; mu ″)s(λi) For a standard sample at a characteristic X-ray wavelength λiEffective mass absorption coefficient of; deltaij(λ) is the superposition of the matrix effects of the analysis element i and the matrix element j; phi' is the incident angle of the incident X-ray; φ "is the exit angle of the characteristic X-ray; n is the sum of the number of analysis elements in the standard sample; cnIs the mass fraction of the analysis element n; mu.sn(λ) is the mass absorption coefficient of the analytical element n at the incident wavelength λ; mu.sn(λi) For analysing the element n at a characteristic X-ray wavelength lambdaiThe lower mass absorption coefficient; when the incident wavelength λ is smaller than the characteristic X-ray wavelength of the matrix element j, Dj(λ) ═ 1, otherwise 0; when the characteristic X-ray wavelength λ of the matrix element jjLess than the characteristic X-ray wavelength of the analysis element i, Di(λj) 1, otherwise 0; k is a radical ofjAn excitation factor of matrix element j; mu.sj(λ) is the mass absorption coefficient of the matrix element j at the incident wavelength λ; mu.si(λj) For analysing element i at a characteristic X-ray wavelength lambdajThe lower mass absorption coefficient; mu.ss(λj) Is a standardSample at characteristic X-ray wavelength λjThe lower mass absorption coefficient; pijIs a representation of the secondary fluorescence of i and matrix element j.
Further, the gaussian function in step 1.2 and step 3 is defined as:
Figure BDA0002539000820000034
Ei＝12.3981/λ (11)
wherein λ is the incident wavelength of the incident X-rays, EiFor the energy of the incident X-ray, E' is the energy deposition part in the X-ray measuring detector, i.e. the energy of the incident X-ray minus the energy of the emergent secondary fluorescence ray, σ is an important parameter for controlling the width of the Gaussian peak, and σ (E)i) At an energy of incident X-rays of EiThe sigma parameter of time.
Further, step 1.4 describes the continuous spectrum I (. lamda.) of the standard sample obtained in step 1.3i) The mean square prediction error expression of the spectrum I actually measured by the standard sample is as follows:
Figure BDA0002539000820000041
wherein n is the number of spectrum channels of the actually measured spectrum of the standard sample, and m is the mth spectrum.
The invention has the beneficial effects that:
the invention generates discrete spectrum of each analysis element of any sample based on Sherman equation forward direction, widens the spectrum into continuous spectrum through Gaussian function, and generates optimal instrument factor G based on each analysis elementiRapidly generating successive X-ray fluorescence spectra of the sample, wherein each analytical element has an optimum instrument factor Gi' optimization by genetic Algorithm based on Instrument factor GiThe mean square prediction error of the continuous spectrum of the standard sample and the actually measured spectrum is less than 0.2, and the instrument factor of each analysis element is the optimal instrument factor Gi'; the invention establishes a theoretical model for rapidly generating continuous X-ray fluorescence spectrumCompared with Monte Carlo simulation, the method has the advantages that the calculation speed is increased by more than 2 orders of magnitude, the generation of multi-element sample spectra with any composition at the speed of second level is realized, the X-ray fluorescence spectrum information with a large amount of data is built at extremely low time cost, and a basic theoretical model is provided for the realization of building a spectrum information big database through calculation substitution experiments.
Drawings
FIG. 1 is a flowchart of a method for rapidly generating an X-ray fluorescence spectrum based on genetic algorithm optimization basic parameters in example 1 of the present invention;
FIG. 2 is a dispersion spectrum and a continuous spectrum of elemental Fe in example 1 of the present invention;
FIG. 3 is the instrument factor G for fast generation of YSBS37351_6 spectrum of standard alloy in embodiment 1 of the present inventioniBefore and after optimization, comparing the spectrum with actual test data;
FIG. 4 is the instrument factor G for fast generation of the spectrum of standard alloy A3 in example 1 of the present inventioniBefore and after optimization, comparing the spectrum with actual test data;
FIG. 5 shows the instrument factor G for fast generation of the YSBS 41346-2014-904L spectrum of the standard alloy in example 1iAnd (4) comparing the spectrum of the data before and after optimization with the spectrum of the actual test data.
Detailed Description
Example 1
The embodiment provides a method for rapidly generating an X fluorescence spectrum based on genetic algorithm optimization basic parameters, the working process is shown in figure 1, and an optimal instrument factor G of elements 12-92 in a periodic table of elements is establishedi' database, taking the standard alloy sample YSBS37351_6, standard alloy sample A3 and standard alloy sample YSBS41346_2014_904L as examples, the element numbers and contents of the three standard alloy samples are shown in Table 1,
TABLE 1 elemental numbers and contents of standard alloy samples
Figure BDA0002539000820000051
The specific steps for obtaining the optimal instrument factors of the elements in the three standard alloy samples are as follows:
step 1: establishing a basic parameter database of different elements, wherein the basic parameter database comprises element serial numbers, mass absorption coefficients, fluorescence yield, spectral line fractions, absorption jump factors and absorption edge wavelengths; then introducing the components and element content of the sample; emitting incident X-rays by using a multi-wavelength light source (X-ray tube) with the wavelength of 0.177A-0.413A, and introducing the incident angle (23 ℃) of the incident X-rays and the emergent angle (37 ℃) of characteristic X-rays;
respectively converting the content information of each analysis element in the standard alloy sample YSBS37351_6, the standard alloy sample A3 and the standard alloy sample YSBS41346_2014_904L into emission spectrum intensity according to Sherman equation, and generating a discrete spectrum I of each analysis elementi(λi) (ii) a Wherein i is an analysis element, namely an element with the serial number of 12-92 in the standard alloy sample, and lambdaiWavelength of the characteristic X-ray for the analysis element i; based on Sherman equation, content information of analytical element i in standard alloy sample YSBS37351_6 is converted into excitation wavelength lambdaiEmission spectrum intensity Ii(λi) The conversion expression of (1) is:
Figure BDA0002539000820000052
ki＝Ji·ωi·pi (4)
Figure BDA0002539000820000053
Figure BDA0002539000820000061
Figure BDA0002539000820000062
Figure BDA0002539000820000063
μs(λj)＝Ciμi(λj)+Cjμj(λj) (9)
wherein, the meaning of parameter expression in Sherman formula is as follows:
i-analytical element;
j-matrix elements, i.e., other elements in the standard alloy sample YSBS37351_6 except for the analysis element i;
n is the sum of the number of analytical elements in the standard alloy sample YSBS37351_ 6;
s-representation of element interactions;
λi-analyzing the wavelength of the characteristic X-ray of element i;
λ0-the minimum wavelength of the incident X-rays;
λedgei-the absorption edge wavelength of the incident X-ray for the analysis element i;
λ — the incident wavelength of incident X-rays;
Ii(λi) -the emission wavelength of the analysis element i is λiThe spectral intensity of the characteristic X-ray of (a);
Io(λ) — the intensity distribution of the incident wavelength λ;
Gi-analyzing the instrument factor for element i;
Ci-mass fraction of analytical element i in standard alloy sample YSBS37351 — 6;
Cj-mass fraction of matrix element j in standard alloy sample YSBS37351 — 6;
ki-analyzing the excitation factor of element i;
Ji-an absorption step factor;
ωi-fluorescence yield;
pi-a spectral line score;
phi' is the angle of incidence of the incident X-rays;
φ "-the exit angle of the characteristic X-ray;
μi(λ) — analyzing the mass absorption coefficient of element i at the incident wavelength λ;
μ′s(λ) — the effective mass absorption coefficient at the incident wavelength λ of the standard alloy sample YSBS37351 — 6;
μ″s(λi) -the standard alloy sample YSBS37351 — 6 at the characteristic X-ray wavelength λiEffective mass absorption coefficient of;
δij(λ) -analyzing the superposition of the matrix effect of element i and matrix element j;
Cn-mass fraction of analytical element n in standard alloy sample YSBS37351 — 6;
μn(λ) — analyzing the mass absorption coefficient of element n at the incident wavelength λ;
μn(λi) -analyzing the element n at a characteristic X-ray wavelength λiThe lower mass absorption coefficient;
kj-the excitation factor of matrix element j;
μj(λ) — the mass absorption coefficient of matrix element j at the incident wavelength λ;
μi(λj) -analyzing the element i at a characteristic X-ray wavelength λjThe lower mass absorption coefficient;
μs(λj) -the standard sample is at a characteristic X-ray wavelength λjThe lower mass absorption coefficient;
Dj(λ) -D when the incident wavelength λ is less than the characteristic X-ray wavelength of matrix element jj(λ) ═ 1, otherwise 0;
Di(λj) -when the characteristic X-ray wavelength λ of the matrix element j isjLess than the characteristic X-ray wavelength of the analysis element i, Di(λj) 1, otherwise 0;
Pij-representation of i and matrix element j secondary fluorescence;
in addition, the first and second substrates are,kithe reaction probability of the element i under the X-ray is analyzed, namely an excitation factor which is the product of the fluorescence yield, the spectral line fraction and the absorption jump factor;
step 2: broadening the discrete spectrum of each analysis element in the standard alloy sample obtained in the step 1 into a continuous spectrum through a Gaussian function of the spectrum
Figure BDA0002539000820000071
Wherein, defining the Gaussian function as:
Figure BDA0002539000820000072
Ei＝12.3981/λ (11)
where λ is the wavelength of the incident X-rays, EiFor the incident X-ray energy, E' is the fraction of energy deposited in the X-ray measurement detector, i.e. the energy of the incident X-ray minus the energy of the emergent secondary fluorescence ray, σ is an important parameter for controlling the width of the gaussian peak, and σ ═ width at half maximum/2.355, σ (E) is obtained by experimentally measuring the width at half maximum of the broadening peak in this examplei) At an energy of incident X-rays of EiA sigma parameter of time;
fig. 2 shows a discrete spectrum and a broadened continuous spectrum of the element Fe.
And step 3: for a particular physical model, for a particular element i, its instrumental factor GiThe proportionality constant for the analytical element is instrument dependent. Assuming that the initial value of the instrument factor of each analysis element in the standard alloy sample is 1, the continuous spectrum of each analysis element in the standard alloy sample is subjected to
Figure BDA0002539000820000073
Corresponding instrument factor GiMultiplying and accumulating to obtain continuous spectrum I (lambda) of the standard alloy samplei) The formula is as follows:
Figure BDA0002539000820000081
wherein G isiAn instrument factor to analyze element i;
step 1.4: in the process of optimizing the instrument factor, the continuous spectrum I (lambda) of the standard alloy sample obtained in the step 3i) The mean square prediction error EMSPE of the spectrum I actually measured by the standard alloy sample is an evaluation function, and the formula is as follows:
Figure BDA0002539000820000082
wherein n is the number of spectrum traces of the actually measured spectrum of the standard sample, m is the mth spectrum, and n is 2048 in the model of the embodiment;
optimizing an Instrument factor G for analysis element i by a genetic AlgorithmiTo reduce the EMSPE, wherein the EMSPE is used<0.2 computing the Instrument factor G as a genetic AlgorithmiThe optimized Fe element optimal instrument factor G is evaluatedi' setting as reference 1, obtaining optimized optimal instrument factor G of other analysis elements according to proportioni'; wherein i is 1 … N, and N is the sum of the numbers of the analytic elements in each standard alloy sample;
the specific process of optimizing the instrument factors of each analysis element in the evaluation function through the genetic algorithm in the step 4 is as follows:
step 4.1: optimizing the instrumental factor G of the analytical element i in a Standard alloy sampleiIf EMSPE is more than or equal to 0.2, continuously optimizing the instrument factor GiIf EMSPE<0.2, the optimization is finished, and the instrument factor of the analysis element i is the optimal instrument factor Gi′；
Step 4.2: optimizing the Instrument factor G of analytical element i in the Standard alloy sample in step 1.4.1iThe instrument factors of other analysis elements in the standard alloy sample are optimized, and the optimal instrument factors of all the analysis elements in the standard alloy sample are finally obtained.
Instrument factor G of three alloy samples including standard alloy sample YSBS37351_6, standard alloy sample A3 and standard alloy sample YSBS41346_2014_904LiBefore optimization, instrument factor GiAfter optimization, the product is mixed withThe spectrogram of the actual test data is shown in FIG. 3, FIG. 4 and FIG. 5, respectively, and the instrument factor G is showniGenerated spectrogram after optimization compared with instrument factor GiBefore optimization, the method is closer to an actually measured spectrogram, and the accuracy of the method for generating the X-ray fluorescence spectrogram in a simulation way is embodied.
Standard alloy sample YSBS37351_6, standard alloy sample A3, and standard alloy sample YSBS41346_2014_904L were optimized instrument factor G based on each analytical elementi' comparison of database-optimized EMSPE with Pre-optimized EMSPE, results are shown in Table 2, see instrument factor GiThe EMSPE of the spectrogram generated by simulation after optimization is obviously smaller than that before optimization of the instrument factor, which shows that the X-ray fluorescence spectrogram generated quickly based on the model of the embodiment after optimization of the instrument factor is closer to the actual spectrogram.
TABLE 2 EMSPE before and after optimization of Standard alloy samples
Alloy sample EMSPE before optimization Optimized EMSPE
YSBS37351_6 11.508 7.348
A3 0.433 0.014
YSBS41346_2014_904L 36.509 5.642
Based on the invention, the optimal instrument factor G of No. 12-92 elements in the periodic table of elements is establishedi' database, the procedure for rapidly generating the X-ray fluorescence spectrogram of the Standard alloy sample YSBS37351_6 is as follows:
step 1: inputting the content information of the YSBS37351_6 standard alloy sample, and converting the content information of each analysis element in the YSBS37351_6 standard alloy sample into emission spectrum intensity I according to Sherman equationi′(λi) And generating a discrete spectrum;
step 2: broadening the discrete spectrum obtained in the step 1 into a continuous spectrum through a Gaussian function
Figure BDA0002539000820000091
And step 3: based on optimal instrument factor Gi' database and continuous spectra from step 2
Figure BDA0002539000820000092
Calculating to obtain a continuous X-ray fluorescence spectrum I' (lambda) of the YSBS37351_6 standard alloy samplei) The formula is as follows:
Figure BDA0002539000820000093
the time consumed for generating the X-ray fluorescence spectrogram of the standard alloy sample YSBS37351_6 is only 0.907 second, and the rapidity and the superiority of the method are further embodied.

Claims (4)

1. A method for rapidly generating an X fluorescence spectrum based on optimization of basic parameters by a genetic algorithm is characterized by comprising the following steps:
step 1: establishing an optimal instrument factor G of No. 12-92 elements in the periodic tablei' database, the concrete steps are as follows:
step 1.1: introducing the content information of the standard sample, and carrying out the standard sample according to Sherman equationThe content information of each analysis element is converted into emission spectrum intensity Ii(λi) And generating a discrete spectrum; wherein i is an analytical element, λiWavelength of the characteristic X-ray for the analysis element i;
step 1.2: broadening the discrete spectrum obtained in step 1.1 into a continuous spectrum by a Gaussian function
Figure FDA0003289872970000011
Step 1.3: assuming that the initial value of the instrument factor of each analysis element in the standard sample is 1, combining the continuous spectrum obtained in the step 1.2
Figure FDA0003289872970000012
Calculating to obtain a continuous spectrum I (lambda) of the standard samplei) The formula is as follows:
Figure FDA0003289872970000013
wherein G isiAn instrument factor to analyze element i;
step 1.4: continuous Spectrum I (. lamda.) of the Standard sample obtained in step 1.3i) The mean square prediction error of the spectrum I actually measured by the standard sample is an evaluation function, and the instrument factor G of the analysis element I in the evaluation function is optimized through a genetic algorithmiMaking the mean square prediction error less than 0.2, setting the optimized optimal instrument factor of the Fe element as a reference, setting the value as 1, and obtaining the optimized optimal instrument factors G of other analysis elements in proportioni', finally establishing the optimal instrument factor G of No. 12-92 elements in the periodic tablei' a database; n, wherein i is 1.. N, and N is the sum of the numbers of analytical elements in the standard sample;
step 2: inputting the content information of the sample to be detected, and generating the emission spectrum intensity I of each analysis element in the sample to be detected according to the method in the step 1.1i′(λi) And generating a discrete spectrum;
and step 3: the discrete spectrum obtained in the step 2 is obtained by a Gaussian functionSpread to a continuous spectrum
Figure FDA0003289872970000014
And 4, step 4: optimal instrument factor G based on each analysis element obtained in step 1i' and continuous Spectrum obtained in step 3
Figure FDA0003289872970000015
Calculating to obtain the continuous X-ray fluorescence spectrum I' (lambda) of the sample to be measuredi) The formula is as follows:
Figure FDA0003289872970000016
2. the method for rapidly generating X-ray fluorescence spectrum based on genetic algorithm optimization basic parameters as claimed in claim 1, wherein in step 1.1, the content information of analysis element i is converted into the wavelength λ according to Sherman equationiCharacteristic X-ray emission spectral intensity Ii(λi) The conversion expression of (1) is:
Figure FDA0003289872970000017
ki＝Ji·ωi·pi (4)
Figure FDA0003289872970000018
Figure FDA0003289872970000021
Figure FDA0003289872970000022
Figure FDA0003289872970000023
μs(λj)＝Ciμi(λj)+Cjμj(λj) (9)
wherein i is an analysis element; j is a matrix element; lambda [ alpha ]iWavelength of the characteristic X-ray for the analysis element i; i isi(λi) For analysis of element i the emission wavelength is lambdaiThe spectral intensity of the characteristic X-ray of (a); giAn instrument factor to analyze element i; ci、CjRespectively mass fractions of an analysis element i and a matrix element j in a standard sample; k is a radical ofiTo analyze the excitation factor of element i as the fluorescence yield ωiSpectral line fraction piAnd absorption transition factor JiThe product of the three; lambda [ alpha ]0Is the minimum wavelength of the incident X-rays; lambda [ alpha ]edgeiIs the absorption edge wavelength of the incident X-ray for the analysis element i; lambda is the incident wavelength of the incident X-ray and ranges from 0.177A to 0.413A; i iso(λ) is the spectral intensity distribution of the incident wavelength λ; mu.si(λ) is the mass absorption coefficient of the analytical element i at the incident wavelength λ; s is a representation of the interaction of elements; mu's's(λ) is the effective mass absorption coefficient of the standard sample at the incident wavelength λ; mu ″)s(λi) For a standard sample at a characteristic X-ray wavelength λiEffective mass absorption coefficient of; deltaij(λ) is the superposition of the matrix effects of the analytical element i and the matrix element j at the incident wavelength λ; phi' is the incident angle of the incident X-ray; φ "is the exit angle of the characteristic X-ray; n is the sum of the number of analysis elements in the standard sample; cnIs the mass fraction of the analysis element n; mu.sn(λ) is the mass absorption coefficient of the analytical element n at the incident wavelength λ; mu.sn(λi) For analysing the element n at a characteristic X-ray wavelength lambdaiThe lower mass absorption coefficient; when the incident wavelength λ is smaller than the characteristic X-ray wavelength of the matrix element j, Dj(λ) ═ 1, otherwise0; when the characteristic X-ray wavelength λ of the matrix element jjLess than the characteristic X-ray wavelength of the analysis element i, Di(λj) 1, otherwise 0; k is a radical ofjAn excitation factor of matrix element j; mu.sj(λ) is the mass absorption coefficient of the matrix element j at the incident wavelength λ; mu.si(λj) For analysing element i at a characteristic X-ray wavelength lambdajThe lower mass absorption coefficient; mu.ss(λj) For a standard sample at a characteristic X-ray wavelength λjThe lower mass absorption coefficient; pijIs a representation of the secondary fluorescence of i and matrix element j.
3. The method for rapidly generating X-ray fluorescence spectrum based on genetic algorithm optimization basic parameters as claimed in claim 1, wherein the Gaussian function defined in step 1.2 and step 3 is:
Figure FDA0003289872970000024
Ei＝12.3981/λ (11)
wherein λ is the incident wavelength of the incident X-rays, EiFor the energy of the incident X-ray, E' is the fraction of the energy deposit in the X-ray measurement detector, σ is a parameter for the Gaussian peak width, σ (E)i) At an energy of incident X-rays of EiSigma of time.
4. The method for rapidly generating X-ray fluorescence spectrum based on genetic algorithm optimization basic parameters as claimed in claim 1, wherein the expression of mean square prediction error in step 1.4 is:
Figure FDA0003289872970000031
wherein n is the number of spectrum channels of the actually measured spectrum of the standard sample, and m is the mth spectrum.
CN202010541483.1A 2020-06-15 2020-06-15 Method for rapidly generating X fluorescence spectrum based on genetic algorithm optimization basic parameters Active CN111766260B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010541483.1A CN111766260B (en) 2020-06-15 2020-06-15 Method for rapidly generating X fluorescence spectrum based on genetic algorithm optimization basic parameters

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010541483.1A CN111766260B (en) 2020-06-15 2020-06-15 Method for rapidly generating X fluorescence spectrum based on genetic algorithm optimization basic parameters

Publications (2)

Publication Number Publication Date
CN111766260A CN111766260A (en) 2020-10-13
CN111766260B true CN111766260B (en) 2021-12-03

Family

ID=72721627

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010541483.1A Active CN111766260B (en) 2020-06-15 2020-06-15 Method for rapidly generating X fluorescence spectrum based on genetic algorithm optimization basic parameters

Country Status (1)

Country Link
CN (1) CN111766260B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112382714B (en) * 2020-11-13 2022-09-13 福建鸿博光电科技有限公司 LED spectrum synthesis method and terminal
CN116106347B (en) * 2022-11-07 2023-06-20 成都物熙科技有限公司 Second-order X-ray fluorometer and measuring method thereof

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104111263B (en) * 2014-07-07 2016-08-24 大连理工大学 A kind of X-ray fluorescence spectra fundamental parameters method of dummy synthesis standard specimen
CN107290376B (en) * 2017-06-26 2019-10-18 中国建材检验认证集团股份有限公司 A kind of calculation method of the matrix correction coefficient of XRF analysis
CN107748172B (en) * 2017-11-30 2019-10-25 南京航空航天大学 A kind of xrf analysis instrument optimum analysis time determines method
CN108051466A (en) * 2017-12-17 2018-05-18 中国科学院合肥物质科学研究院 Chemical fertilizer ingredient harmless quantitative detection method based on X-ray fluorescence spectra analysis
US11403083B2 (en) * 2018-10-30 2022-08-02 Nippon Telegraph And Telephone Corporation Offloading server and offloading program

Also Published As

Publication number Publication date
CN111766260A (en) 2020-10-13

Similar Documents

Publication Publication Date Title
Yan et al. A novel hybrid feature selection strategy in quantitative analysis of laser-induced breakdown spectroscopy
CN111766260B (en) Method for rapidly generating X fluorescence spectrum based on genetic algorithm optimization basic parameters
Rousseau Corrections for matrix effects in X-ray fluorescence analysis—A tutorial
Shi et al. Quantitative analysis of sedimentary rocks using laser-induced breakdown spectroscopy: comparison of support vector regression and partial least squares regression chemometric methods
Sarkar et al. Laser ablation molecular isotopic spectrometry of water for 1D2/1H1 ratio analysis
Wolff et al. A new fundamental parameter based calibration procedure for micro X-ray fluorescence spectrometers
Fu et al. A fast variable selection method for quantitative analysis of soils using laser-induced breakdown spectroscopy
CN106814061A (en) A kind of method for improving LIBS overlap peak accuracy of quantitative analysis
CN105718749B (en) A kind of analysis of coal nature characteristics method based on large database concept identification
CN104483292B (en) A kind of method that use multiline ratio method improves laser microprobe analysis accuracy
JP5150620B2 (en) A method to determine the feasibility of the proposed structural analysis process
Müller et al. Enhanced spectral analysis of C-TOF Aerosol Mass Spectrometer data: Iterative residual analysis and cumulative peak fitting
CN112834485B (en) Non-calibration method for quantitative analysis of laser-induced breakdown spectroscopy elements
Brocchieri et al. Estimation of Ag coating thickness by different methods using a handheld XRF instrument
Klus et al. Effect of experimental parameters and resulting analytical signal statistics in laser-induced breakdown spectroscopy
Tang et al. Spectral interference elimination and self-absorption reduction in laser-induced breakdown spectroscopy assisted with laser-stimulated absorption
Kanngießer Quantification procedures in micro X-ray fluorescence analysis
Yu et al. Response surface methodology for optimizing LIBS testing parameters: A case to conduct the elemental contents analysis in soil
Gu et al. A data preprocessing method based on matrix matching for coal analysis by laser-induced breakdown spectroscopy
CN109030467B (en) Self-absorption effect correction method for laser breakdown spectroscopy
Xu et al. High spectral stability and quantitative accuracy of LIBS by optimization of a combined atomic and ionic line algorithm
CN103792214A (en) Method for improving carbon content measuring accuracy in steel
CN105717094B (en) A kind of metal element content analysis method based on large database concept identification
CN105717093B (en) A kind of cement characteristics analysis method based on large database concept identification
CN107941758A (en) A kind of coal quality measurement of caloric value method based on Pretreated spectra

Legal Events

Date Code Title Description
PB01 Publication
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