CN112700822A - Laser-induced breakdown spectroscopy concentration extraction method for online monitoring of trace gas impurities - Google Patents
Laser-induced breakdown spectroscopy concentration extraction method for online monitoring of trace gas impurities Download PDFInfo
- Publication number
- CN112700822A CN112700822A CN202011399646.3A CN202011399646A CN112700822A CN 112700822 A CN112700822 A CN 112700822A CN 202011399646 A CN202011399646 A CN 202011399646A CN 112700822 A CN112700822 A CN 112700822A
- Authority
- CN
- China
- Prior art keywords
- line
- characteristic
- concentration
- data
- formula
- 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.)
- Pending
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 47
- 239000012535 impurity Substances 0.000 title claims abstract description 33
- 238000002536 laser-induced breakdown spectroscopy Methods 0.000 title claims abstract description 29
- 238000000605 extraction Methods 0.000 title claims abstract description 11
- 230000003595 spectral effect Effects 0.000 claims abstract description 128
- 238000001228 spectrum Methods 0.000 claims abstract description 30
- 238000009795 derivation Methods 0.000 claims abstract description 26
- 238000004364 calculation method Methods 0.000 claims abstract description 22
- 238000001914 filtration Methods 0.000 claims abstract description 18
- 238000007781 pre-processing Methods 0.000 claims abstract description 14
- 238000000034 method Methods 0.000 claims description 43
- 238000012795 verification Methods 0.000 claims description 36
- 238000012360 testing method Methods 0.000 claims description 24
- 238000012549 training Methods 0.000 claims description 22
- 238000001514 detection method Methods 0.000 claims description 21
- 238000004458 analytical method Methods 0.000 claims description 15
- 230000015556 catabolic process Effects 0.000 claims description 9
- 238000011156 evaluation Methods 0.000 claims description 9
- 230000005284 excitation Effects 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 238000005259 measurement Methods 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 8
- 239000000203 mixture Substances 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 238000012790 confirmation Methods 0.000 claims description 4
- 239000002245 particle Substances 0.000 claims description 4
- 238000000295 emission spectrum Methods 0.000 claims description 3
- 230000005281 excited state Effects 0.000 claims description 3
- 238000003705 background correction Methods 0.000 claims description 2
- 238000009499 grossing Methods 0.000 claims description 2
- 238000013441 quality evaluation Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 13
- 239000007789 gas Substances 0.000 description 78
- 239000000523 sample Substances 0.000 description 18
- 239000000446 fuel Substances 0.000 description 12
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 10
- 239000011261 inert gas Substances 0.000 description 10
- 230000002285 radioactive effect Effects 0.000 description 10
- 230000004927 fusion Effects 0.000 description 9
- 239000001257 hydrogen Substances 0.000 description 8
- 229910052739 hydrogen Inorganic materials 0.000 description 8
- 238000010200 validation analysis Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 229910052734 helium Inorganic materials 0.000 description 5
- UFHFLCQGNIYNRP-VVKOMZTBSA-N Dideuterium Chemical compound [2H][2H] UFHFLCQGNIYNRP-VVKOMZTBSA-N 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000001069 Raman spectroscopy Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 238000004817 gas chromatography Methods 0.000 description 3
- 239000003758 nuclear fuel Substances 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000005855 radiation Effects 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 239000001307 helium Substances 0.000 description 2
- SWQJXJOGLNCZEY-UHFFFAOYSA-N helium atom Chemical compound [He] SWQJXJOGLNCZEY-UHFFFAOYSA-N 0.000 description 2
- 238000004949 mass spectrometry Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000000691 measurement method Methods 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 238000001303 quality assessment method Methods 0.000 description 2
- 239000002901 radioactive waste Substances 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 239000002912 waste gas Substances 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 239000000443 aerosol Substances 0.000 description 1
- 238000001636 atomic emission spectroscopy Methods 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 238000000889 atomisation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000007707 calorimetry Methods 0.000 description 1
- 239000012159 carrier gas Substances 0.000 description 1
- 238000003776 cleavage reaction Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004992 fission Effects 0.000 description 1
- 229910052732 germanium Inorganic materials 0.000 description 1
- GNPVGFCGXDBREM-UHFFFAOYSA-N germanium atom Chemical compound [Ge] GNPVGFCGXDBREM-UHFFFAOYSA-N 0.000 description 1
- 229910052736 halogen Inorganic materials 0.000 description 1
- 150000002367 halogens Chemical class 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000000746 purification Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 239000000941 radioactive substance Substances 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000007017 scission Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000002459 sustained effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000012384 transportation and delivery Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/20—Identification of molecular entities, parts thereof or of chemical compositions
-
- 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
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/90—Programming languages; Computing architectures; Database systems; Data warehousing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Theoretical Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Crystallography & Structural Chemistry (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Optics & Photonics (AREA)
- Biochemistry (AREA)
- Software Systems (AREA)
- Plasma & Fusion (AREA)
- Databases & Information Systems (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Analytical Chemistry (AREA)
- Spectroscopy & Molecular Physics (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
A laser-induced breakdown spectroscopy concentration extraction method for online monitoring of trace gas impurities comprises the following steps: 1) collecting spectral data to form a spectral data set; 2) carrying out data preprocessing to obtain the wavelength and the peak intensity value of the characteristic spectral line of the target element; 3) establishing a special database of target element characteristic spectral lines; 4) matching calculation is carried out on the wavelength of the element characteristic spectral line and a database, and element attribution information of the characteristic spectral line is confirmed; 5) establishing an element calibration model; 6) the invention is different from the multivariate statistical regression technical means, combines the two advantageous technologies of sectional average smooth spectrum noise filtering and derivation peak searching, can fully correct the spectrum background while retaining the original spectrum characteristic variable information to the maximum extent, and can quickly calculate and obtain the wavelength and the net peak height intensity value of the element characteristic peak.
Description
Technical Field
The invention relates to the technical field of atomic emission spectroscopy analysis, in particular to a laser-induced breakdown spectroscopy concentration extraction method for online monitoring of trace gas impurities.
Background
The hydrogen production by electrolyzing water is an important hydrogen production way in the development and utilization of new energy. As the main fuel for fuel engines and fuel cells, hydrogen purity is critical to the performance of the fuel engine, fuel cell efficiency, and the sustained maintenance of electrode activity. The stored hydrogen gas may be contaminated before use by trace impurity components such as N2, O2, and Ar from air leakage or introduced during hydrogen production. The purity of the hydrogen in the storage tank and the delivery pipeline is accurately monitored in time, which is a necessary premise for ensuring the efficient utilization of the hydrogen energy.
In the development and utilization of clean energy nuclear energy, real-time and rapid monitoring of gas components is also very important. Radioactive waste generated in the form of radioactive waste gas and the like during operation of a nuclear power plant contains radioactive particles or nuclides such as inert gas isotopes, halogens, aerosols and the like generated by fission, neutron irradiation and burning of a fuel element, and is released to the environment. Although the total cumulative emission of radionuclides during these conventional releases is large, the concentration of airborne radionuclides generated by nuclear power plants is generally low, so that the radioactivity in the environmental medium is below the detection limit of instrumental monitoring, and the irradiation dose cannot be calculated by environmental monitoring alone. Therefore, in the safety management of nuclear power plants, the emission of radioactive inert gases in effluents is mainly counted at present through detection lower limits from the effluent monitoring of each nuclear power plant, namely, according to the effluent monitoring data of the nuclear power plant, the amount of radioactive substances released by the nuclear power plant is obtained by adopting a proper environment model, and finally basic data for the radiation environment influence evaluation of the nuclear power plant is formed, and the basis for evaluating the environment quality and the dose received by the public is established. However, since the emission amount of radioactive inert gas actually counted with the data below the detection limit is almost equal to the total emission amount of gaseous effluent, it is difficult to estimate the target monitoring amount below the detection limit using the experimentally analyzed data. Especially in the case of not clearly defining the measurement detection limit of the radioactive nuclide of the nuclear power plant effluent and the key nuclide therein, the statistics of the emission amount is greatly different due to the influence of different sampling modes, different monitoring instruments, different monitoring methods and different monitoring conditions between different measurement mechanisms and the monitoring capability and detection limit of the radioactive inert gas with low concentration level, so that the monitoring number from different channels is greatly differentAccording to the effluent discharge evaluation, the comparability is lacked, the reliability of the measurement result evaluation is poor, and the scientificity of the effluent discharge evaluation of the nuclear power plant is finally influenced. Current isotopes for radioactive inert gases such as Xe and Kr, as well as He, Ar41Most of the monitoring methods for major monitoring indexes are radioactivity activity detection methods, such as a beta measurement method and a measurement method using a high-purity germanium spectrometer, which are more prone to off-line detection characteristics, the sampling amount of inert gas is usually several liters, the measurement time is also ten minutes to several hours, and due to the lack of a mature concentration gas sampling method, the monitoring result of each nuclide of radioactive inert gas is often smaller than the detection limit of an instrument. In view of this, in order to meet the requirement of radioactive inert gas monitoring in nuclear power plants, it is necessary to improve the technical sensitivity of radionuclide monitoring and the effectiveness of analytical methods during normal operation of nuclear power plants to ensure that the statistical results of radioactive inert gas emissions are comparable in a wide range. And a new technology and a new method are developed in time, the detection capability of effluent process monitoring is improved, and the concentration information of the target element is rapidly acquired, so that the important technical premise of objective evaluation of radioactive inert gas emission is formed.
When the fusion energy is oriented to the future, the heavy hydrogen and the super heavy hydrogen in the fusion reactor are used as fuel, the operated core plasma interacts with the wall material, and the generated impurities are exhausted along with the fuel which is not burnt. The burnt nuclear fuel is finely processed, impurity gas in the nuclear fuel is removed, heavy hydrogen and super heavy hydrogen are quickly recovered and then injected into a fusion reactor, and efficient circulation and utilization of the nuclear fuel can be realized. Therefore, impurities in the fuel cycle exhaust gas of the fusion reactor can be rapidly and accurately monitored, and the method has great significance for reducing the utilization cost of fusion energy fuel. At present, in the design of plasma waste gas integrated purification treatment systems at home and abroad, laser Raman spectroscopy, mass spectrometry, gas chromatography, calorimetry, ionization chamber monitoring and other technologies are used for analyzing waste gas samples according to respective advantages and disadvantages. However, He and Ar are not sensitive to raman effect, and cannot be monitored well by laser raman spectroscopy, and they cannot be detected by gas chromatography when they are used as carrier gases, and are very difficult to be analyzed by mass spectrometry due to their overlapping interference by gas species, cleavage products, and the like. A new efficient technical means is developed based on the technology and the application characteristics, and the method becomes a new way for meeting the requirement of rapidly monitoring the components of the circulating exhaust ash gas of the fusion device.
Laser Induced Breakdown Spectroscopy (LIBS), as a well-known analytical tool, has the technical advantages of being real-time and online and requiring no sample preparation, etc. The system has the advantages that the pulse laser can be introduced into a sample to be detected only by forming the optical path system by using the optical components, so that the system has non-invasiveness and good remote monitoring performance, is flexible in assembly, small in size and strong in application environment adaptability, and can realize atomization excitation of the sample by directly acting the high-energy-density pulse laser on the sample, so that the analysis speed is very high, and the system is particularly suitable for online instantaneous acquisition of target element information in extreme environments including a radiation environment.
However, the application of LIBS technology is currently more focused on analysis of solid samples. The method is influenced by factors such as matrix characteristics of the solid sample, and the element calibration by utilizing the LIBS technology is mainly based on the comprehensive application of the multivariate statistical regression technology, so that the required data volume is large, the calculation cost is high, and the real-time analysis characteristics of the method are reduced to a certain extent. The matrix composition characteristics and the element excitation characteristics of a gas sample are considered, a proper LIBS calibration analysis technology is developed, and a laser-induced breakdown spectroscopy concentration extraction method applicable to online monitoring of trace gas impurities under multiple environmental conditions is established, so that the method has important technical and application values for rapid and accurate monitoring of hydrogen production purity of electrolyzed water and detection of typical nuclide content with specific radiation indication characteristics and measurement sensitivity in different types of nuclear facilities. In the fusion energy field, the technical advantages of remote, rapid, non-invasive, high-sensitivity and the like of the LIBS technology are fully exerted, effective information of trace impurity components in the process of exhausting ash gas, storing fuel and conveying fuel of the super-heavy hydrogen recovery system is obtained on line in a quantitative mode, and the fusion energy utilization method has important technical value and technical innovation for improving the economy and sustainable development of fusion energy utilization. In view of the above, the invention provides a laser-induced breakdown spectroscopy concentration extraction method for online monitoring of trace gas impurities.
Disclosure of Invention
The invention aims to provide a laser-induced breakdown spectroscopy concentration extraction method for online monitoring of trace gas impurities, so as to solve the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme:
a laser-induced breakdown spectroscopy concentration extraction method for online monitoring of trace gas impurities comprises the following steps:
1) collecting spectral data to form a spectral data set;
2) carrying out data preprocessing to obtain the wavelength and the peak intensity value of the characteristic spectral line of the target element;
3) establishing a special database of target element characteristic spectral lines;
4) matching calculation is carried out on the wavelength of the element characteristic spectral line and a database, and element attribution information of the characteristic spectral line is confirmed;
5) establishing an element calibration model;
6) and applying the characteristic peak intensity value as an input set to an element calibration model to realize the rapid calibration and real-time output of the target element concentration.
As a further scheme of the invention: the step 1) specifically comprises the following steps:
1.1) under the optimized working condition, continuously exciting the flowing gas in the conveying pipeline by using a laser-induced breakdown spectroscopy analysis system, and collecting the plasma spectrum accumulated pulse data of gas components;
1.2) randomly and equally dividing the collected gas component spectrum data into three groups, wherein one group is used as a training set, the other group is used as a verification set, and the third group is a test set.
As a further scheme of the invention: the step 2) comprises the following specific steps:
2.1) respectively carrying out noise filtering pretreatment on the spectral data of the training set, the verification set and the test set to obtain corresponding noise-filtered spectral data;
2.2) noise filtering pretreatment is calculated by adopting a piecewise average smoothing algorithm, and the formula (1) is a BOX _ ave function;
2.3) further preprocessing the data after noise filtering by adopting a primary and secondary derivation method;
2.4) the derivation method adopts a first derivation and a second derivation algorithm, and the formula (2) is a dev function;
2.5) obtaining the wavelength W of the spectral line through the calculation of the derivation algorithm in the step 2.4) and the positive and negative changes of the slopeλ(ii) a Obtaining the peak height intensity value H of the element spectral lineλ；
2.7) H in step 2.5)λRelating to the XX coefficient in formula (2) in step 2.4). Since the derivative calculation simultaneously achieves the background correction of the spectrum, HλIs the net peak height intensity value, which is expressed as a count of the peak height intensity of the spectral line.
As a further scheme of the invention: the step 3) specifically comprises the following steps:
3.1) establishing a special database of the characteristic spectral lines of the target elements, wherein the special database is the spectral data characteristics of gas component elements acquired according to working conditions and comprises spectral parameter information of the characteristic spectral lines of the target elements;
3.2) the spectral parameter information of the characteristic line of the target element in 3.1) at least comprises the wavelength lambda of the characteristic lineijThe element type (i) corresponding to the characteristic spectral line and the particle excited state type(s) corresponding to the characteristic spectral line;
3.3) characteristic line wavelength λ in said 3.2)ijWherein: i is the sort order of the elements in the database; j is the sequence of different characteristic spectral lines of the same element, and the sequence value is 1, 2, …
As a further scheme of the invention: the step 4) specifically comprises the following steps:
4.1) converting the spectral line wavelength W in the step 2.5)λAs input data, the characteristic spectral line wavelength λ in step 3.2) is compared withijCarrying out matching calculation;
4.2) the matching calculation in step 4.1) is performed according to equation (3):
Dk＝|Wλ-λij|……(3)
wherein: k is an ordinal value of 1, 2, …
4.3) passing through step 4.2), sorting the calculated values by DkMinimum as spectral line WλAnd (3) acquiring corresponding spectral parameter information of the element characteristic spectral line in the element characteristic spectral line special database in the step 3.1) according to the identification basis of element attribution, and realizing element characteristic spectral line identification and element attribution confirmation.
As a further scheme of the invention: the step 5) comprises the following specific steps:
5.1) aiming at the training set data, based on the results of element characteristic spectral line identification and element attribution confirmation in the step 2.5) and the step 4.3), and based on the characteristic spectral line W in the step 2.6)λPeak height intensity value of HλThe reference concentration (x) of the element i in the gas is assigned to the characteristic spectral line for the two-dimensional data input endi) Establishing a single-line calibration model for the other input end of the two-dimensional data, wherein the equation is specifically expressed as formula (4-1):
yi＝ki·xi+bi…(4-1)
wherein: y isiAnd xiAre respectively characteristic spectral line WλPeak height intensity value of HλAnd the concentration of the element i to which it belongs in the gas; coefficient kiAnd biThe slope and intercept of the single line calibration model are respectively;
5.2) for the same element i, selecting a plurality of characteristic spectral lines, respectively establishing a single-line scaling equation according to the step 5.1), and obtaining different equation coefficient combinations kijAnd bijThe scaling equation formula can be expressed as:
yij＝kij·xi+bij…(4-2)；
5.3) the one-line scaling equation of 5.2) is a linear curve model based on least square linear fitting, wherein the least square fitting of the linear curve model determines a coefficient Rij 2；
5.4) concentration x of the element to which the characteristic line belongs in formula (4-1) in said step 5.1) and in formula (4-2) in said step 5.2)iIs the volume concentration of the element in the gas component orMolar concentration;
5.5) further, in order to better implement the invention, the following arrangement is adopted in particular: in the step 5.1), establishing a formula (4-1) and in the step 5.3), establishing a formula (4-2) to calculate a single-line scaling equation least square fitting measurement coefficient Rij 2On the basis, the calibration method also comprises robustness evaluation and verification, namely, the robustness of the single-line calibration model is verified by using verification set data, and the reliability of the calibration result is improved by selecting different characteristic spectral lines is judged and selected.
5.6) the robustness assessment and verification in step 5.5) is specifically as follows: the element characteristic spectral line W in the verification set data obtained in the step 2.6) is usedλPeak height intensity value of HλThe formula (4-1) or the formula (4-2) is input as an input end, and the concentration predicted value of the corresponding element of the characteristic spectral line in the verification set is calculated and obtained
As an output terminal of the formula (4-1) or the formula (4-2), and calculates a predicted value
And a reference concentration xiRoot mean square error RMSEP in between;
5.7) scaling equation coefficients in equation (4-1) in said 5.1) and equation (4-2) in said 5.3), and RMSEP in 5.6) are related to the working optimization conditions, gas composition characteristics, and element excitation characteristics, having element line characteristics;
5.8) determining the coefficient R according to a least squares fit in step 5.2) and step 5.3)ij 2And 5.6) the calculated value of RMSEP, and evaluating the quality of the scaling equation of the formula (4-2). With Rij 2The maximum value and the minimum value of RMSEP are used as the optimal characteristic spectral line of the determined target element, and the optimal equation coefficient k in the formula (4-2) is determinedijAnd bijAccording to the method, an optimal single-line scaling equation of the element i concentration is established, and an optimal single-line scaling equation formula (5) is established:
yi＝ki0·xi+bi0……(5)。
as a further scheme of the invention: the step 6) specifically comprises the following steps:
6.1) determining the net peak intensity value H of the corresponding element spectral line of the test set of step 2.6) according to said optimal single line scaling equation formula (5) established in step 5.8)λObtaining the optimal detection result of the concentration of the target element in the gas component as the input end data of the formula (5) of the scaling equation
6.2) according to the detection result of the concentration of the test set element i in the step 6.1)
And a reference value x for the concentration of element iiCalculating the root mean square error RMSE predicted by the optimal calibration equation, and further judging the generalization capability of the element optimal characteristic spectral line selected in the step 5.7) to improve the calibration result;
6.3) the optimal detection result of the concentration of the target element i in the gas component in the step 6.1) has the real-time output characteristic.
The overall process of the laser-induced breakdown spectroscopy concentration extraction method for on-line monitoring of trace gas impurities comprises the following steps:
WF1 utilizes a laser-induced breakdown spectroscopy analysis system to carry out pulse laser excitation on gas components in the conveying pipeline to generate plasma, and plasma emission spectrum data are collected and recorded;
the WF2 randomly and averagely divides the acquired data into three parts, one part is used as a training set, the other part is a verification set, and the rest third part is a test set;
WF3 respectively performs noise filtering and derivation peak searching preprocessing on the spectral data of the training set, the verification set and the test set to obtain the wavelength of an element characteristic peak and the count value of the net peak high intensity in a specific spectral range;
WF4 creates a specialized spectral line database of gas component elements;
matching the characteristic peak wavelength of the target element with the data of the special spectral line database by the WF5, and confirming the element attribution of the selected characteristic peak;
and the WF6 respectively uses the characteristic peak net peak height value of the confirmed element attribution in the training set and the reference concentration of the attributive element in the gas as two input end variables of the calibration model to establish a single-line calibration model, verifies the effectiveness and optimization of the single-line calibration model by using the two input end variables corresponding to the verification set, and finally calculates and obtains the calibration result and the quality evaluation data of the target element concentration of the test set according to the established single-line calibration optimization model.
Compared with the prior art, the invention has the following important technical advantages and effects:
(1) the method is different from a multivariate statistical regression technical means, combines two advantageous technologies of noise filtering and derivation peak searching of a segmented average smooth spectrum, can fully correct the spectrum background while retaining original spectrum characteristic variable information to the maximum extent, can quickly calculate and obtain the wavelength and the net peak height intensity value of an element characteristic peak, and provides an instant and reliable input variable for single line calibration, thereby being particularly suitable for the on-line monitoring requirement of gas components. And compared with the multivariate statistical regression technology, the calculation cost is obviously reduced.
(2) The method provided by the invention can realize the calibration of the concentration of the target element in the gas component only by using a single element characteristic spectral line (or a plurality of spectral lines), and compared with the conventional method, the method only uses a few data calculation dimensions, saves the calculation cost and is beneficial to the real-time output of data information.
(3) Compared with the multivariate statistical regression technology, the data preprocessing algorithm adopted by the invention avoids the complex calculation of high-dimensional data and can ensure the stability of data preprocessing and the accuracy of the calibration result.
(4) The calibration model provided by the invention is remarkably improved in stability and reliability test and generalization capability application, is used for analyzing gas component elements, and can completely meet the requirement of real-time calibration monitoring on the concentration of the target element in an actual application scene.
(5) The invention can realize the rapid extraction of the concentration information of trace gas impurities on-line monitoring by matching with a specially designed analysis system, especially under the multi-field coupling extreme environment.
(6) The invention fully utilizes the technical advantages of laser-induced breakdown spectroscopy, does not need separation and preparation of a sample, can realize excitation and detection of non-invasive spectroscopy by only utilizing the combination of optical components and lenses, and has the advantages of good detection limit, low cost, rapidness and high efficiency.
Drawings
FIG. 1 is a schematic diagram of a laser-induced breakdown spectroscopy analysis system;
FIG. 2 is a laser induced breakdown spectrum of hydrogen gas containing trace impurity components;
FIG. 3 is a graph of a characteristic peak noise filtering effect;
FIG. 4a shows the first derivation of characteristic Ar I696.45 nm;
FIG. 4b is the second derivative of characteristic Ar I696.45 nm;
FIG. 4c shows the peak search result of characteristic Ar I696.45 nm;
FIG. 5 is a graph showing the comparison of the net peak intensity of characteristic peaks of different elements calculated by deriving peak finding and the high intensity of Gaussian fitting peaks;
FIG. 6 gas reference concentrations of helium, oxygen, nitrogen and helium are related to their net peak intensities;
in the way: the device comprises a pulse laser 1, a control circuit 2, a gas cylinder 3, an air inlet system 4, a light path system 5, a functional gas chamber 6, a vacuum pump system 7, a spectrometer 8 and a computer 9.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1:
a method for extracting concentration of a laser-induced breakdown spectrum for online monitoring of trace gas impurities comprises the steps of collecting data of the laser-induced breakdown spectrum, preprocessing the spectrum, extracting characteristic peak intensity, establishing a special database of element characteristic spectral lines, identifying attribution of elements of the characteristic spectral lines, modeling and optimizing by single-line calibration, and rapidly extracting and outputting concentration information of the online monitoring of trace gas impurities in real time, and relates to the following steps:
1) collecting spectral data to form a spectral data set;
2) preprocessing the spectral data set, and extracting the wavelength and the peak intensity of the characteristic peak of the element;
3) establishing an element spectral line special database;
4) matching calculation is carried out on the wavelength information of the characteristic spectral line and an element spectral line special database, and the element attribution of the characteristic spectral line is confirmed;
5) establishing a single-line calibration equation by using the element characteristic spectral line peak height intensity value and optimizing;
6) and obtaining the concentration information of the target element to be measured by using the optimal scaling equation model.
Example 2:
the embodiment is further optimized on the basis of the embodiment 1, and in order to better realize the invention, the following setting mode is particularly adopted: the step 1) specifically comprises the following steps:
1.1) carrying out pulse laser continuous excitation on a gas component sample in a conveying pipeline by using a laser-induced breakdown spectroscopy analysis system, and collecting accumulated data of a plasma spectrum under the action of continuous pulse laser;
1.2) randomly and averagely dividing the acquired spectral data into three groups, wherein one group is used as a training set, the other group is used as a verification set, and the third group is a test set; the training set data comprises spectrum data of the target element in the maximum concentration range; the validation set data and the test set data are each independent of the training set data.
Example 3:
the embodiment is further optimized on the basis of any one of the above embodiments, and in order to better implement the invention, the following setting mode is particularly adopted: the step 2) comprises the following specific steps:
2.1) respectively carrying out noise filtering pretreatment on the spectral data of the training set, the verification set and the test set;
2.2) the noise filtering process adopts a segmented quasi-gaussian fitting function BOX _ ave function, and the formula (1) is the BOX _ ave function:
wherein:
2.3) carrying out further derivation pretreatment on the data after noise filtering of the training set, the verification set and the test set through the step 2.2);
2.4) the derivation pretreatment sequentially comprises primary derivation and secondary derivation, the functions are respectively a formula (2-1) and a formula (2-2), and the formula (2) is a dev function;
2.5) extracting the element characteristic peak wavelength W obtained after derivation in the step 2.4)λAnd peak height intensity value HλInformation;
2.6) characteristic Peak wavelength W in step 2.5)λWhen the derivation algorithm is calculated, the calculation result is obtained through the positive and negative changes of the slope and is related to the coefficient AAA in the formula (2); hλRelated to the BBB coefficient in equation (2);
step 2.4) adopts the formula (2-1) and the formula (2-2) to respectively carry out derivation calculation, so that the automatic correction of the spectrum background can be realized, and therefore, H in the step 2.5) isλIs the value of the net peak high intensity of the element characteristic peak;
2.7) the characteristic peak wavelength W extracted in the step 2.5)λThe attribute identification of the characteristic spectral line elements is used as a basis for confirming the attribute identification of the characteristic spectral line elements; its corresponding net peak height intensity value HλAs input data for the single-line scaling model.
Example 4:
the present embodiment is re-optimized on the basis of any of the above embodiments, and in order to further better implement the present invention, the following setting modes are particularly adopted: the step 3) comprises the following steps:
3.1) establishing a special target element characteristic spectral line database containing element characteristic peak spectral parameter information by contrasting an NIST database according to the element composition and the element excitation characteristics of gas components under the working condition;
3.2) elemental characteristic line specific database includes characteristic line wavelengths (λ)ij) The element type (i) corresponding to the characteristic spectral line, the particle excited state type(s) corresponding to the characteristic spectral line and other spectral parameter information.
Example 5:
in this embodiment, re-optimization is performed on the basis of any one of the above embodiments, and in order to further better implement the present invention, the following setting manner is particularly adopted, and the step 4) includes the following specific steps:
4.1) extracting the characteristic peak wavelength W of the element extracted in the step 2.5)λAs input data, the characteristic spectral line wavelength data (lambda) in the element characteristic spectral line special database in the step 3.2) is compared withij) Matching calculation is carried out on the spectral parameters;
4.2) the matching calculation is carried out according to the formula (3):
Dk＝|Wλ-λij|……(3)
wherein: i is the kind of the element in the database; j is a different characteristic spectral line of the same element with the sequence number 1, 2, …
4.3) sorting the calculated values according to step 4.2) by DkThe minimum value is used as a judgment basis for identifying the characteristic spectral line elements, and attribution information of the elements of the characteristic spectral line in the step 3.2) is obtained;
example 6:
in this embodiment, re-optimization is performed on the basis of any one of the above embodiments, and in order to further better implement the present invention, the following setting is particularly adopted, and the step 5) includes the following specific steps:
5.1) selecting the corresponding characteristic spectral line W of the target element in the training set in the step 2.5) according to the element attribution identification result of the characteristic spectral line in the step 4.3)λThe net peak height intensity value H is calculatedλWith the concentration (x) of the element in the gas componenti) Linear combination is carried out, a single-line calibration model is established, and the equation is specifically expressed as a formula (4-1):
yi＝kixi+bi
wherein: y isiAnd xiRespectively corresponding to characteristic spectral lines WλPeak height intensity value of HλAnd the concentration of the element to which it belongs in the gas; coefficient kiAnd biThe slope and intercept, respectively, of the single line calibration model.
5.2) according to the element attribution characteristics of element spectral lines, a plurality of single-line calibration model equations can be established for the same element, and the equations can be specifically expressed as a formula (4-2);
5.3) carrying out least square linear fitting on the plurality of calibration equations established in the step 5.2), and determining a coefficient R according to the fitting2Evaluating to determine the element characteristic spectral line W of the optimal single-line scaling equation in formula (4-2)λThe optimal one-line scaling equation is expressed as formula (5);
yi0＝ki0xi+bi0 (5)
and 5.4) carrying out robustness evaluation by using verification set data according to the established optimal single-line calibration model, and using the data as the basis for evaluating the calibration quality of the optimal single-line calibration equation. Respectively calculating the Root Mean Square Error (RMSEP) and the Mean Absolute Error (MAE) between the predicted value of the target element content of the verification set and the reference value thereof, and further evaluating the prediction analysis performance of the optimal single-line scaling equation;
synthesis of R fitted with a Linear scaling equation in 5.4)2And RMSEP parameter as basis for comprehensive evaluation of optimal single-line scaling equation, specifically maximum R2The value and the minimum RMSEP value are used as indicators.
Example 7:
the present embodiment is re-optimized on the basis of any of the above embodiments, and in order to further better implement the present invention, the following setting modes are particularly adopted: the step 6) comprises the following specific steps:
6.1) according to the optimal single-line scaling equation established in the step 5.1) and the step 5.3), corresponding characteristic spectral lines W of the target elements in the test set obtained in the step 2.5)λPeak height intensity value of HλAnd inputting an optimal scaling equation to obtain a test concentration value of the target element.
6.2) verifying the generalization ability of the single-line calibration model by using the test set data. Specifically, the concentration content of the elements in the test set and the known reference content are used to calculate the root mean square error RMSE and the average absolute error MAE, and the generalization ability of the element concentration calibration by the optimal single-line calibration equation established in the steps 5.1) and 5.3) is evaluated.
6.3) wherein the tested concentration of the target element in the gas component in the step 6.1) can be output in real time.
To further illustrate the operational procedures of the present invention and to demonstrate the effectiveness of the method for extracting the concentration of a laser-induced breakdown spectrum for online monitoring of trace gas impurities, the following examples are further illustrated in conjunction with the accompanying fig. 1-6, tables 1-2 and the following examples:
TABLE 1 calibration model quality assessment results
TABLE 1-1 concentration of He in mock gas
Remarking: verification samples; na is invalid prediction data
TABLE 1-2 concentration of O2 in the simulated gas
Remarking: verification samples; the predicted content of the element O in the table
TABLE 1-3 concentration of N2 in the simulated gas
Remarking: to verify the samples: the predicted content of the elements in the table is based on the N element
TABLE 1-4. concentration of Ar in simulated gas
Remarking: verification samples; the data in the table are the predicted content of Ar element
Wherein, 7 samples of the concentration gradient are prepared for establishing a calibration equation, and 1 sample is prepared as a verification sample for evaluating and verifying the calibration quality of the calibration equation.
Table 2 data preprocessing combined single line scaling method gas component target element content prediction graph
Example 8:
the laser-induced breakdown spectroscopy analysis system comprises a pulse laser 1, a control circuit 2, a gas cylinder 3, an air inlet system 4, a light path system 5, a functional gas chamber 6, a vacuum pump system 7, a spectrometer 8 and a computer 9, wherein the computer 9 is electrically connected with the pulse laser 1 through the control circuit 2, the computer 9 is connected with the spectrometer 8, the output end of the pulse laser 1 is connected with the input end of the functional gas chamber 6 through the light path system 5, the air inlet of the functional gas chamber 6 passes through the two air inlet systems 4 and the output end of the gas cylinder 3, the exhaust hole of the functional gas chamber 6 is connected with the exhaust end of the vacuum pump system 7, and the detection probe of the spectrometer 8 is arranged in the functional gas chamber 6 as shown in fig. 1.
The energy of the pulse laser is 150mJ, the laser wavelength is 1064nm, the pulse action frequency is 10Hz, the pulse delay is 0.8 mus, and the spectrometer integration time is 50 mus.
This example utilizes a fine gas distribution system to generate 14 simulated gas samples: the 1# to 14# simulated gases are composed of hydrogen as a main component (volume ratio is 90%), and other trace impurity components of O2, N2, Ar and He in different concentration levels. Table 1 shows the reference content of trace impurity gas components in 14 gas distributions calibrated by gas chromatography. The mixed gas is controlled by a flowmeter to enter a sample chamber in a gas pipeline, and the gas pressure is 1 standard atmosphere (1.01E5 Pa). The gas components in the sample chamber are continuously excited by pulse laser by using a laser-induced breakdown spectroscopy analysis system, plasma emission spectra are collected by using a lens group, and spectral data of different simulated gases are obtained by recording through a spectrometer system, as shown in fig. 2.
When spectrum data are collected aiming at simulated gases with different gas ratios, each gas sample is excited by continuous pulse laser, an accumulated measurement spectrum is obtained by every 100 pulses, and each sample collects 4 independent spectra. A total of 56 spectra were collected for 14 samples.
The spectral data were randomly divided into three groups, one group of data was a training set (7 samples), the other group of data was a validation set (4 samples), and the rest were prediction set samples (3 in total). The validation set data will be used to evaluate the robustness and reliability of the single-line scaling model. And carrying out noise filtering and derivation preprocessing on the collected sample spectral data of the training set, the verification set and the prediction set, and extracting the characteristic peak intensity value of the target element. The specific flow is shown in fig. 3 and 4.
The method comprises the steps of taking the peak height value of a characteristic peak of a target element extracted from training spectral data and the concentration of the element in the gas component as two-dimensional data input ends, and establishing a single-line scaling equation of the element, namely yi＝kix+bi(4-1), the same element can select a plurality of characteristic spectral lines to establish a plurality of scaling equations yij＝kijx+bij(4-2)。
Least squares fit determination coefficient R for calculating single line scaling equationij 2With Rij 2Based on the minimum value, the optimal scaling equation of trace impurity elements is established, yi0＝ki0xi+bi0(5). Wherein the optimal scaling equations established for He, O2, N2, Ar are shown in fig. 5; table 2 shows the quality assessment result parameters for the best scaling model.
Carrying out gas component target element concentration verification by combining data preprocessing with a single line calibration method, wherein the characteristic spectral line peak height intensity value of the target element extracted from the verification set is substituted into an optimal calibration equation, and the verification concentration of the target element in the verification set is obtained through calculation; the optimal scaling equations established for He, O2, N2, Ar in the validation set samples (e.g., sample No. 8) were validated, and the elemental concentration results obtained are shown in the bottom row of table 1.
The robustness and reliability of the optimal single-line scaling equation can be evaluated by calculating the Root Mean Square Error (RMSEP) or the Mean Absolute Error (MAE) of the target element concentration predictions of the validation set. As shown in table 1, the MAE values of He, O2, N2, and Ar in sample No. 8 as the verification set are 3.99%, 3.41, 5.27%, and 0.86%, respectively, which verifies that the established calibration model has good stability.
For comparison, selecting the characteristic peak of the optimal single-line scaling equation to perform Gaussian fitting on the training set and the verified spectral data after noise filtering, and acquiring the Gaussian net peak intensity value H of the characteristic peakGλAnd the net peak intensity value H of the element characteristic peak obtained after the derivative data processingλA linear correlation is established as shown in fig. 6. Determining the coefficient R from a least squares fit2The method can judge that the net peak intensity value of the element characteristic peak obtained after the derivative data is processed and the Gaussian fitting peak height intensity value show good linear correlation, R2The value is close to 1, and the high intensity of the net peak obtained by utilizing the derivation peak searching is further proved to have good reliability and effectiveness.
For the validation set samples No. 8-11, the concentrations of Ar, He, N2 and O2 in the mixed gas are predicted by using the established optimal scaling equation, and RMSEP and MAE are compared on the prediction results, as shown in tables 3 and 4.
Table 3: predicted outcome RMSEP comparison
Table 4: prediction outcome MAE comparison
Comparing the results in tables 3 and 4, it can be seen that the method for rapidly extracting the laser-induced breakdown spectroscopy target information for on-line monitoring of trace gas impurities provided by the invention can realize real-time monitoring of the concentration of the target element of the gas component under the working condition, shows good robustness, and has small precision and relative error and obvious technical advantages for calibration prediction of a verification set.
The embodiment proves that the single-line calibration performance of the concentration of the target element in the gas component is reliable by adopting noise filtering and derivation for spectrum pretreatment, and the technical advantages of the single-line calibration and the laser-induced breakdown spectroscopy for real-time and rapid calibration of trace gas impurity components can be exerted by combining the single-line calibration.
Example 9:
2 parts of simulated mixed gas samples (S1 and S2) with different element compositions are prepared, and 4 spectral data of S1 and S2 are respectively collected as test set data according to the same experimental working conditions for establishing an optimal single-line calibration equation.
The data of the S1 and S2 laser-induced breakdown spectroscopy were subjected to model quantitative analysis according to the procedure shown in FIG. 6. The results of the calibration of the elemental concentrations in the resulting S1 and S2 samples are shown in fig. 6 and table 5:
table 5: comparison of different elements of prediction results
The results in table 5 show that the method of single-line calibration by using noise filtering and derivation pretreatment and using the net peak height value has good generalization capability, and the quantitative accuracy provides a new effective spectroscopic detection method for online monitoring of trace gas impurities under multi-environment conditions.
The foregoing is merely some embodiments of the invention and is not intended to limit the invention in any manner. Any simple modifications and equivalent changes made to the above-described embodiments according to the technical spirit of the present invention are within the scope of the present invention.
Claims (8)
1. A laser-induced breakdown spectroscopy concentration extraction method for on-line monitoring of trace gas impurities is characterized by comprising the following steps:
1) collecting spectral data to form a spectral data set;
2) carrying out data preprocessing to obtain the wavelength and the peak intensity value of the characteristic spectral line of the target element;
3) establishing a special database of target element characteristic spectral lines;
4) matching calculation is carried out on the wavelength of the element characteristic spectral line and a database, and element attribution information of the characteristic spectral line is confirmed;
5) establishing an element calibration model;
6) and applying the characteristic peak intensity value as an input set to an element calibration model to realize the rapid calibration and real-time output of the target element concentration.
2. The method for extracting the concentration of the laser-induced breakdown spectrum for the online monitoring of trace gas impurities according to claim 1, wherein the step 1) specifically comprises the following steps:
1.1) under the optimized working condition, continuously exciting the flowing gas in the conveying pipeline by using a laser-induced breakdown spectroscopy analysis system, and collecting the plasma spectrum accumulated pulse data of gas components;
1.2) randomly and equally dividing the collected gas component spectrum data into three groups, wherein one group is used as a training set, the other group is used as a verification set, and the third group is a test set.
3. The method for extracting the concentration of the laser-induced breakdown spectrum for the online monitoring of trace gas impurities according to claim 1 or 2, wherein the step 2) comprises the following specific steps:
2.1) respectively carrying out noise filtering pretreatment on the spectral data of the training set, the verification set and the test set to obtain corresponding noise-filtered spectral data;
2.2) noise filtering pretreatment is calculated by adopting a piecewise average smoothing algorithm, and the formula (1) is a BOX _ ave function;
2.3) further preprocessing the data after noise filtering by adopting a primary and secondary derivation method;
2.4) the derivation method adopts a first derivation and a second derivation algorithm, and the formula (2) is a dev function;
2.5) obtaining the wavelength W of the spectral line through the calculation of the derivation algorithm in the step 2.4) and the positive and negative changes of the slopeλ(ii) a To obtain the elementPeak height intensity value H of spectral lineλ；
2.7) H in step 2.5)λRelating to the XX coefficient in formula (2) in step 2.4). Since the derivative calculation simultaneously achieves the background correction of the spectrum, HλIs the net peak height intensity value, which is expressed as a count of the peak height intensity of the spectral line.
4. The method for extracting the concentration of the laser-induced breakdown spectrum for the online monitoring of trace gas impurities according to claim 3, wherein the step 3) specifically comprises the following steps:
3.1) establishing a special database of the characteristic spectral lines of the target elements, wherein the special database is the spectral data characteristics of gas component elements acquired according to working conditions and comprises spectral parameter information of the characteristic spectral lines of the target elements;
3.2) the spectral parameter information of the characteristic line of the target element in 3.1) at least comprises the wavelength lambda of the characteristic lineijThe element type (i) corresponding to the characteristic spectral line and the particle excited state type(s) corresponding to the characteristic spectral line;
3.3) characteristic line wavelength λ in said 3.2)ijWherein: i is the sort order of the elements in the database; j is the order of different characteristic spectral lines of the same element, with the order values 1, 2, ….
5. The method for extracting the laser-induced breakdown spectroscopy concentration for on-line monitoring of trace gas impurities according to claim 4, wherein the step 4) specifically comprises the following steps:
4.1) converting the spectral line wavelength W in the step 2.5)λAs input data, the characteristic spectral line wavelength λ in step 3.2) is compared withijCarrying out matching calculation;
4.2) the matching calculation in step 4.1) is performed according to equation (3):
Dk＝|Wλ-λij|……(3)
wherein: k is an ordinal value of 1, 2, …
4.3) passing through step 4.2), sorting the calculated values by DkMinimum value asSpectral line WλAnd (3) acquiring corresponding spectral parameter information of the element characteristic spectral line in the element characteristic spectral line special database in the step 3.1) according to the identification basis of element attribution, and realizing element characteristic spectral line identification and element attribution confirmation.
6. The method for extracting the laser-induced breakdown spectroscopy concentration of trace gas impurities on-line monitoring according to claim 5, wherein the step 5) comprises the following specific steps:
5.1) aiming at the training set data, based on the results of element characteristic spectral line identification and element attribution confirmation in the step 2.5) and the step 4.3), and based on the characteristic spectral line W in the step 2.6)λPeak height intensity value of HλThe reference concentration (x) of the element i in the gas is assigned to the characteristic spectral line for the two-dimensional data input endi) Establishing a single-line calibration model for the other input end of the two-dimensional data, wherein the equation is specifically expressed as formula (4-1):
yi＝ki·xi+bi…(4-1)
wherein: y isiAnd xiAre respectively characteristic spectral line WλPeak height intensity value of HλAnd the concentration of the element i to which it belongs in the gas; coefficient kiAnd biThe slope and intercept of the single line calibration model are respectively;
5.2) for the same element i, selecting a plurality of characteristic spectral lines, respectively establishing a single-line scaling equation according to the step 5.1), and obtaining different equation coefficient combinations kijAnd bijThe scaling equation formula can be expressed as:
yij＝kij·xi+bij…(4-2)；
5.3) the one-line scaling equation of 5.2) is a linear curve model based on least square linear fitting, wherein the least square fitting of the linear curve model determines a coefficient Rij 2；
5.4) concentration x of the element to which the characteristic line belongs in formula (4-1) in said step 5.1) and in formula (4-2) in said step 5.2)iIs the volume concentration or mole of the element in the gas componentConcentration;
5.5) establishing a formula (4-1) in the step 5.1) and establishing a formula (4-2) in the step 5.3) to calculate a single line scaling equation least square fitting measurement coefficient Rij 2On the basis, the calibration method also comprises robustness evaluation and verification, namely, the robustness of the single-line calibration model is verified by using verification set data, and the reliability of the calibration result is improved by selecting different characteristic spectral lines is judged;
5.6) the robustness assessment and verification in step 5.5) is specifically as follows: the element characteristic spectral line W in the verification set data obtained in the step 2.6) is usedλPeak height intensity value of HλThe formula (4-1) or the formula (4-2) is input as an input end, and the concentration predicted value of the corresponding element of the characteristic spectral line in the verification set is calculated and obtained
As an output terminal of the formula (4-1) or the formula (4-2), and calculates a predicted value
And a reference concentration xiRoot mean square error RMSEP in between;
5.7) scaling equation coefficients in equation (4-1) in said 5.1) and equation (4-2) in said 5.3), and RMSEP in 5.6) are related to the working optimization conditions, gas composition characteristics, and element excitation characteristics, having element line characteristics;
5.8) determining the coefficient R according to a least squares fit in step 5.2) and step 5.3)ij 2And the calculated value of RMSEP in the step 5.6), the quality of the scaling equation of the formula (4-2) is evaluated; with Rij 2The maximum value and the minimum value of RMSEP are used as the optimal characteristic spectral line of the determined target element, and the optimal equation coefficient k in the formula (4-2) is determinedijAnd bijAccording to the method, an optimal single-line scaling equation of the element i concentration is established, and an optimal single-line scaling equation formula (5) is established:
yi＝ki0·xi+bi0……(5)。
7. the method for extracting the concentration of the laser-induced breakdown spectrum for the online monitoring of trace gas impurities according to claim 8, wherein the step 6) specifically comprises the following steps:
the step 6) specifically comprises the following steps:
6.1) determining the net peak intensity value H of the corresponding element spectral line of the test set of step 2.6) according to said optimal single line scaling equation formula (5) established in step 5.8)λObtaining the optimal detection result of the concentration of the target element in the gas component as the input end data of the formula (5) of the scaling equation
6.2) according to the detection result of the concentration of the test set element i in the step 6.1)
And a reference value x for the concentration of element iiCalculating the root mean square error RMSE predicted by the optimal calibration equation, and further judging the generalization capability of the element optimal characteristic spectral line selected in the step 5.7) to improve the calibration result;
6.3) the optimal detection result of the concentration of the target element i in the gas component in the step 6.1) has the real-time output characteristic.
8. The method for extracting the concentration of the laser-induced breakdown spectrum for the online monitoring of trace gas impurities according to claim 7, wherein the whole process comprises the following steps:
WF1 utilizes a laser-induced breakdown spectroscopy analysis system to carry out pulse laser excitation on gas components in the conveying pipeline to generate plasma, and plasma emission spectrum data are collected and recorded;
the WF2 randomly and averagely divides the acquired data into three parts, one part is used as a training set, the other part is a verification set, and the rest third part is a test set;
WF3 respectively performs noise filtering and derivation peak searching preprocessing on the spectral data of the training set, the verification set and the test set to obtain the wavelength of an element characteristic peak and the count value of the net peak high intensity in a specific spectral range;
WF4 creates a specialized spectral line database of gas component elements;
matching the characteristic peak wavelength of the target element with the data of the special spectral line database by the WF5, and confirming the element attribution of the selected characteristic peak;
and the WF6 respectively uses the characteristic peak net peak height value of the confirmed element attribution in the training set and the reference concentration of the attributive element in the gas as two input end variables of the calibration model to establish a single-line calibration model, verifies the effectiveness and optimization of the single-line calibration model by using the two input end variables corresponding to the verification set, and finally calculates and obtains the calibration result and the quality evaluation data of the target element concentration of the test set according to the established single-line calibration optimization model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011399646.3A CN112700822A (en) | 2020-12-03 | 2020-12-03 | Laser-induced breakdown spectroscopy concentration extraction method for online monitoring of trace gas impurities |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011399646.3A CN112700822A (en) | 2020-12-03 | 2020-12-03 | Laser-induced breakdown spectroscopy concentration extraction method for online monitoring of trace gas impurities |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112700822A true CN112700822A (en) | 2021-04-23 |
Family
ID=75506181
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011399646.3A Pending CN112700822A (en) | 2020-12-03 | 2020-12-03 | Laser-induced breakdown spectroscopy concentration extraction method for online monitoring of trace gas impurities |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112700822A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113450883A (en) * | 2021-06-25 | 2021-09-28 | 中南大学 | Solution ion concentration detection method based on multispectral fusion |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2938066A1 (en) * | 2008-11-06 | 2010-05-07 | Centre Nat Rech Scient | SYSTEM AND METHOD FOR QUANTITATIVE ANALYSIS OF THE ELEMENTARY COMPOSITION OF LASER-INDUCED PLASMA SPECTROSCOPY MATERIAL (LIBS) |
| CN102262076A (en) * | 2011-07-26 | 2011-11-30 | 清华大学 | Laser-induced breakdown spectroscopy element concentration determination method based on spectral line combination |
| CN102313731A (en) * | 2010-07-09 | 2012-01-11 | 中国科学院沈阳自动化研究所 | Method for detecting content of component in unknown object on line |
| US20140168645A1 (en) * | 2012-12-13 | 2014-06-19 | Gwangju Institute Of Science And Technology | Quantitative analysis method for measuring target element in specimen using laser-induced plasma spectrum |
| CN104251846A (en) * | 2014-09-04 | 2014-12-31 | 清华大学 | Discriminant analysis combined laser-induced breakdown spectroscopy quantitative analysis method |
| CN106814061A (en) * | 2016-12-13 | 2017-06-09 | 华中科技大学 | A kind of method for improving LIBS overlap peak accuracy of quantitative analysis |
| US20170191940A1 (en) * | 2014-06-20 | 2017-07-06 | National Research Council Of Canada | Method for laser-induced breakdown spectroscopy and calibration |
| US20170219494A1 (en) * | 2016-02-01 | 2017-08-03 | Bwt Property, Inc. | Laser Induced Breakdown Spectroscopy (LIBS) Apparatus with Automatic Wavelength Calibration |
| US20190128811A1 (en) * | 2017-09-20 | 2019-05-02 | Worcester Polytechnic Institute | Molten metal inclusion testing |
| CN110646407A (en) * | 2019-11-20 | 2020-01-03 | 中国海洋大学 | Method for rapidly detecting content of phosphorus element in aquatic product based on laser-induced breakdown spectroscopy technology |
-
2020
- 2020-12-03 CN CN202011399646.3A patent/CN112700822A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2938066A1 (en) * | 2008-11-06 | 2010-05-07 | Centre Nat Rech Scient | SYSTEM AND METHOD FOR QUANTITATIVE ANALYSIS OF THE ELEMENTARY COMPOSITION OF LASER-INDUCED PLASMA SPECTROSCOPY MATERIAL (LIBS) |
| CN102313731A (en) * | 2010-07-09 | 2012-01-11 | 中国科学院沈阳自动化研究所 | Method for detecting content of component in unknown object on line |
| CN102262076A (en) * | 2011-07-26 | 2011-11-30 | 清华大学 | Laser-induced breakdown spectroscopy element concentration determination method based on spectral line combination |
| US20140168645A1 (en) * | 2012-12-13 | 2014-06-19 | Gwangju Institute Of Science And Technology | Quantitative analysis method for measuring target element in specimen using laser-induced plasma spectrum |
| US20170191940A1 (en) * | 2014-06-20 | 2017-07-06 | National Research Council Of Canada | Method for laser-induced breakdown spectroscopy and calibration |
| CN104251846A (en) * | 2014-09-04 | 2014-12-31 | 清华大学 | Discriminant analysis combined laser-induced breakdown spectroscopy quantitative analysis method |
| US20170219494A1 (en) * | 2016-02-01 | 2017-08-03 | Bwt Property, Inc. | Laser Induced Breakdown Spectroscopy (LIBS) Apparatus with Automatic Wavelength Calibration |
| CN106814061A (en) * | 2016-12-13 | 2017-06-09 | 华中科技大学 | A kind of method for improving LIBS overlap peak accuracy of quantitative analysis |
| US20190128811A1 (en) * | 2017-09-20 | 2019-05-02 | Worcester Polytechnic Institute | Molten metal inclusion testing |
| CN110646407A (en) * | 2019-11-20 | 2020-01-03 | 中国海洋大学 | Method for rapidly detecting content of phosphorus element in aquatic product based on laser-induced breakdown spectroscopy technology |
Non-Patent Citations (4)
| Title |
|---|
| GULAB SINGH MAURYA等: "Analysis of impurities on contaminated surface of the tokamak limiter using laser induced breakdown spectroscopy" * |
| KAI RONG等: "Experimental study on mercury content in flue gas of coal-fired units based on laser-induced breakdown spectroscopy" * |
| 杨文斌: "激光诱导击穿光谱技术在气体检测中的应用研究" * |
| 檀兵: "激光诱导击穿光谱谱峰元素的自动识别" * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113450883A (en) * | 2021-06-25 | 2021-09-28 | 中南大学 | Solution ion concentration detection method based on multispectral fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104251846B (en) | Discriminant analysis combined laser-induced breakdown spectroscopy quantitative analysis method | |
| Zhang et al. | LIBS analysis of the ash content, volatile matter, and calorific value in coal by partial least squares regression based on ash classification | |
| Okuda et al. | Rapid and simple determination of multi-elements in aerosol samples collected on quartz fiber filters by using EDXRF coupled with fundamental parameter quantification technique | |
| Wang et al. | Review of online source apportionment research based on observation for ambient particulate matter | |
| Zhang et al. | Quantitative analysis of coal quality by laser-induced breakdown spectroscopy assisted with different chemometric methods | |
| CN112700822A (en) | Laser-induced breakdown spectroscopy concentration extraction method for online monitoring of trace gas impurities | |
| Brown et al. | Comparison of ED-XRF and LA-ICP-MS with the European reference method of acid digestion-ICP-MS for the measurement of metals in ambient particulate matter | |
| Andrews et al. | Sensor technology for molten salt reactor off-gas systems | |
| Flower et al. | A laser-based technique to continuously monitor metal aerosol emissions | |
| CN105277531A (en) | Grading-based coal characteristic measurement method | |
| CN109387482A (en) | Isotope measure device | |
| Gao et al. | Ultra-repeatability measurement of calorific value of coal by NIRS-XRF | |
| CN114813635B (en) | Method for optimizing combustion parameters of coal stove and electronic equipment | |
| Charlton et al. | Operator declaration verification technique for spent fuel at reprocessing facilities | |
| CN208187913U (en) | Atmospheric particulates heavy metal and its isotope on-line water flushing device | |
| Mohan et al. | Calibration free laser ablation molecular isotopic spectrometry (CF-LAMIS) for boron isotopic composition determination | |
| Samson et al. | The photoionization cross sections of molecular chlorine | |
| Awasthi et al. | New Insights on Modern Age Coins by Calibration-Free Laser-Induced Breakdown Spectroscopy Method and Chemometric Approaches | |
| Brennetot et al. | Optimisation of the operating conditions of a quadrupole ICP-MS with hexapole collision/reaction cell for the analysis of selenium-79 in spent nuclear fuel using experimental designs | |
| Nejedlý et al. | Evaluation of elemental and black carbon measurements from the GAViM and IMPROVE networks | |
| Bishop | Opportunities for Computational Development in Laser Induced Breakdown Spectroscopy (LIBS) | |
| Lascola et al. | H-Canyon Dissolver Offgas Monitoring Using Raman Spectroscopy | |
| CN114659996B (en) | Hyperspectral oil detection method based on reflected light | |
| Andrews et al. | Establishing Isotopic Measurement Capabilities using Laser-Induced Breakdown Spectroscopy for the Molten Salt Reactor Campaign | |
| CN109813702A (en) | A method of based on Cr element in laser induced breakdown spectroscopy quantitative analysis tealeaves |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210423 |
|
| WD01 | Invention patent application deemed withdrawn after publication |