Solid-liquid mixture chemical composition on-line detection system and method based on laser multispectral analysis

Abstract

The invention relates to a solid-liquid mixture chemical composition on-line detection system and method based on laser multispectral analysis, and the system comprises: (1) a stable solid-liquid two-phase flow generation module which provides a uniform and stable solid-liquid two-phase flow liquid column for a measurement module; (2) a laser-induced breakdown spectroscopy measurement module for detecting an atomic emission spectrum; (3) a laser Raman spectrum measurement module for detecting a molecular emission spectrum; (4) a granularity analysis module for representing granularity information; and (5) a multi-sensor synchronous control and multi-source data fusion processing module which is responsible for time synchronization of a laser, a spectrum detector and a data storage and processing unit in the system, analysis and processing of multi-source heterogeneous spectrum data, and outputting a final chemical composition online detection result. The system designed by the invention is used for carrying out online detection on the solid-liquid mixture, particle size distribution, elements and mineral composition in the solid-liquid mixture can be synchronously analyzed, and higher analysis precision can be obtained through fusion analysis of multi-source heterogeneous data.

Description

Technical Field

[0001] The invention belongs to the field of physics-measurement-automatic analysis, and specifically relates to an online detection system and method for the chemical composition of a solid-liquid mixture based on laser multispectral analysis. Background Art

[0002] Solid-liquid mixtures are widely present in the natural environment and industrial production. As a very important material carrier and transportation form, they are used in many fields such as mineral production and transportation, petrochemical industry, hydrometallurgy, and food and drug production. The chemical composition of solid-liquid mixtures is an important characterization parameter for studying their internal state, transmission mechanism and change law. Solid-liquid mixtures are a rather complex form of matter, involving the mixture of two different phases of solid particles and liquid. The measurement of their chemical composition is affected by many factors such as particle size distribution, concentration, flow rate and mineral composition.

[0003] Existing online analytical instruments are usually only applicable to situations where the interference parameters remain unchanged or are only within a very small range. Once the interference changes, the measurement accuracy will decrease. Therefore, they are only used for trend tracking in some production fields. In order to obtain higher measurement accuracy, it is often necessary to go through complex pre-treatment sample preparation processes such as sampling, separation, drying, and grinding to separate various interferences, and then perform chemical analysis. The whole process takes up to several hours, or even measured in days. Since solid-liquid mixtures often involve complex chemical reactions and material transfer processes, this large-lag information feedback method is difficult to reveal the dynamic state change process within it, making it difficult for related production fields to achieve fine modeling and regulation of complex physical and chemical reaction processes, and has become a major bottleneck in the technological development of multiple process fields such as mineral separation, pharmaceuticals, and environmental engineering.

[0004] In summary, existing analytical instruments and technologies are unable to achieve high-precision online analysis of chemical composition under the measurement conditions of complex changes in the physical and chemical matrix, which restricts technological innovation and industrial upgrading in the fields of mineral processing, control, and environmental engineering. Summary of the invention

[0005] In view of the above-mentioned deficiencies of existing analytical instruments and technologies, the purpose of the present invention is to provide a solid-liquid mixture chemical composition analysis system and method that can be applied online, and to obtain high-precision chemical component content information through laser multi-spectral measurement of solid-liquid two-phase flow and fusion analysis of multi-source heterogeneous data.

[0006] The technical solution adopted by the present invention to achieve the above-mentioned purpose is: an online detection system for the chemical composition of a solid-liquid mixture based on laser multispectral analysis, comprising:

[0007] The stable solid-liquid two-phase flow generation module includes an online sampling and steady flow structure and a split flow sampling structure; it is used to obtain stable solid-liquid two-phase flow samples to be tested, and introduce the samples to be tested into different measurement modules through the split flow sampling structure;

[0008] The laser induced breakdown spectroscopy measurement module includes a pulsed laser, a LIBS spectrometer, and a laser excitation and plasma luminescence detection optical system; the module is used to focus the pulsed laser onto the surface of the solid-liquid two-phase flow through the laser excitation and plasma luminescence detection optical system to generate plasma, and to collect the LIBS spectrum through the laser excitation and plasma luminescence detection optical system using the LIBS spectrometer;

[0009] The laser Raman spectroscopy measurement module includes a continuous laser, a Raman spectrometer, and a laser emission and Raman scattering light detection optical system. The laser is focused onto the surface of the solid-liquid two-phase flow through the laser emission and Raman scattering light detection optical system to generate Raman scattering, and the Raman spectrometer is used to collect the Raman spectrum.

[0010] The particle size analysis module includes a dilution and dispersion structure and a He-Ne laser light source and an optical detector respectively arranged on both sides thereof; the module is used to irradiate the solid-liquid two-phase flow passing through the dilution and dispersion structure with the He-Ne laser light source, measure Mie scattering with the optical detector and obtain solid phase particle size distribution data;

[0011] The multi-sensor synchronous control and multi-source data fusion processing module includes a host computer and a timing generator; the host computer realizes the timing control of the above modules through the timing generator, synchronously saves the spectrum and particle size data measured by each module, performs analysis and calculation, and outputs the chemical composition analysis results.

[0012] The laser induced breakdown spectroscopy measurement module is configured according to specific application requirements: 1) with a single or two pulsed lasers, the laser wavelength of which can be 355nm, 532nm or 1064nm, as the excitation light source; 2) with a spectrometer with different detection spectrum bands or with multiple spectrometers spliced ​​together to detect LIBS spectral signals within the extreme ultraviolet-near infrared wavelength range; 3) with the laser excitation and plasma luminescence detection optical systems being coaxial or non-coaxial optical systems.

[0013] The laser Raman spectrum measurement module is configured according to specific application requirements: 1) a laser with a wavelength of 532nm, 785nm or 1064nm and a spectrometer with a spectrum band corresponding to the laser wavelength; 2) the laser emission and Raman scattered light detection optical system used for laser incidence and Raman spectrum collection is a coaxial or non-coaxial optical system.

[0014] The laser excitation and plasma luminescence detection optical system and the laser emission and Raman scattering light detection optical system are two completely independent optical systems, or a comprehensive optical system that reuses some components.

[0015] The dilution and dispersion structure comprises a mixing barrel, on which a solid-liquid two-phase flow injection pipeline, a clean water injection pipeline, a measurement output pipeline and a drainage pipeline are arranged, an agitator is arranged at the bottom of the barrel, a shading ratio detector is arranged on the barrel wall, and the measurement output pipeline is connected to the measurement window.

[0016] The on-line detection method of the chemical composition of solid-liquid mixture based on laser multispectral analysis, through the following measurement and analysis steps, collects spectral data, establishes a multispectral fusion analysis model, and performs on-line detection, is used for on-line real-time analysis and measurement of different solid-liquid mixtures, and the method includes the following steps:

[0017] Step 1: The host computer of the multi-sensor synchronous control and multi-source data fusion processing module controls the timing generator to output multi-channel trigger signals through different delays, and controls the laser induced breakdown spectroscopy measurement module, laser Raman spectroscopy measurement module, and particle size analysis module to collect multi-source heterogeneous data of the same time sequence for each solid-liquid two-phase flow sample in the modeling sample set: LIBS spectrum, Raman spectrum, and solid phase particle size distribution data;

[0018] Step 2: The host computer integrates multi-source heterogeneous data to perform chemical composition modeling and analysis, including taking the LIBS spectra, Raman spectra and particle size data of all solid-liquid two-phase flow samples in the modeling sample set as input, and taking the contents of each element and compound in the solid-liquid two-phase flow as output to build a chemical composition analysis model;

[0019] Step 3: Collect multi-source heterogeneous data of the solid-liquid two-phase flow sample under the same time sequence in real time, input the chemical composition analysis model for solution, and output the content of the elements and compounds to be detected in real time.

[0020] The collecting of multi-source heterogeneous data includes:

[0021] Step 11: The laser induced breakdown spectroscopy measurement module receives the trigger instruction, controls the pulse laser to output pulse laser, focuses on the surface of the solid-liquid two-phase flow through the laser excitation and plasma luminescence detection optical system to generate plasma, and controls the LIBS spectrometer to collect LIBS spectrum through the laser excitation and plasma luminescence detection optical system;

[0022] Step 12: The laser Raman spectrum measurement module receives the trigger instruction, controls the continuous laser to output laser, focuses on the surface of the solid-liquid two-phase flow through the laser emission and Raman scattered light detection optical system to generate Raman scattering, and controls the Raman spectrometer to collect Raman spectrum through the laser emission and Raman scattered light detection optical system;

[0023] Step 13: The particle size analysis module receives the trigger instruction, controls the He-Ne laser light source to irradiate the solid-liquid two-phase flow passing through the dilution and dispersion structure, collects the Mie scattering spectrum information obtained on the optical detector, and calculates the solid phase particle size distribution data.

[0024] The particle size analysis module collects solid phase particle size distribution data including the following steps:

[0025] Step 13a: The host computer controls a portion of the original solid-liquid two-phase flow diverted by the split flow sampling structure to be injected into the mixing barrel through the solid-liquid two-phase flow injection pipeline, controls the agitator to stir evenly, and controls the clean water injection pipeline to inject clean water for dilution, thereby adjusting the shading ratio of the solid-liquid mixture in the mixing barrel;

[0026] Step 13b: When the shading ratio detector detects that the shading ratio value is suitable for particle size measurement, the injection of the solid-liquid two-phase flow injection pipeline and the clean water injection pipeline is stopped, and the measurement output pipeline is controlled to open to introduce the solid-liquid mixture into the measurement window to perform laser particle size measurement;

[0027] Step 13c: After the measurement is completed, the drainage pipeline and the clean water injection pipeline are controlled to open, the solid-liquid two-phase flow is emptied and the mixing barrel and the measurement window are cleaned.

[0028] The chemical composition analysis model is:

[0029] C x =f(I LIBS ,I Raman ,I Mie )

[0030] Where x is an atom (element) or molecule (compound), I LIBS ,I Raman and I Mie are LIBS spectrum, Raman spectrum and Mie scattering spectrum data respectively. f(·) is the quantitative analysis model of chemical composition established by supervised learning method. The modeling samples are solid-liquid two-phase flow samples with known chemical composition. The training data are multi-spectral measurement data of each modeling sample. The response label is the chemical composition content information.

[0031] The LIBS spectrum and Raman spectrum data I LIBS and I Raman , which is the original spectral data collected by the spectrometer, or the spectral feature data after data preprocessing and feature extraction;

[0032] The Mie scattering spectrum data I Mie , is the scattering spectrum intensity data at different positions collected by the spectral detector, or the particle size distribution data calculated by the laser particle size analyzer based on the scattering spectrum.

[0033] The present invention has the following advantages and beneficial effects:

[0034] 1. The chemical composition online analysis system proposed in the present invention can synchronously measure LIBS spectrum, Raman spectrum and Mie scattering spectrum. By configuring lasers with different wavelengths, spectrometers with different spectral bands and detectors with different angles, it can realize the physical and chemical composition characterization of different substances at three scales: atomic, molecular and particle. It can be widely used in the online analysis of the chemical composition of various solid-liquid mixtures.

[0035] 2. The present invention can realize the simultaneous analysis of the atomic and molecular composition of solid-liquid mixtures by fusing and analyzing heterogeneous data from multiple sensors, and can eliminate the complex influence of the atomic, molecular composition and particle size changes of solid-liquid mixtures on spectral signals during online measurement, thereby realizing high-precision online analysis of chemical composition. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 Schematic diagram of an online detection system for the chemical composition of a solid-liquid mixture based on laser multispectral analysis;

[0037] Figure 2 It is a schematic diagram of the online sampling, steady flow and flow diversion structure;

[0038] Figure 3 This is a schematic diagram of the optical system and spectrum splicing of the laser induced breakdown spectroscopy measurement module;

[0039] Figure 4 This is a schematic diagram of the dilution and dispersion structure of the laser particle size analysis module;

[0040] Figure 5 This is a schematic diagram of modeling and application of multi-spectral fusion analysis model;

[0041] Among them, 1 is a stable solid-liquid two-phase flow generation module, 11 is an online sampling and steady flow structure, 1101 is a feed component, 1102 is a stirring barrel, 1103 is a stirrer, 1104 is a control valve, 1105 is a steady flow box, 1106 is a solid-liquid two-phase flow outlet, 12 is a split flow sampling structure, 2 is a laser induced breakdown spectroscopy measurement module, 21 is a pulsed laser, 22 is a LIBS spectrometer, 23 is a laser excitation and plasma luminescence detection optical system, 2301 is a laser beam expansion optical path, 2302 is a plasma luminescence detection optical path, 2303 is a coaxial coupling focusing optical path, 3 is a laser Raman spectroscopy measurement module, 31 is Continuous laser, 32 is a Raman spectrometer, 33 is a laser emission and Raman scattered light detection optical system, 4 is a laser particle size analysis module, 41 is a dilution and dispersion structure, 4101 is a mixing barrel, 4102 is a solid-liquid two-phase flow injection pipeline, 4103 is a clean water injection pipeline, 4104 is a measurement output pipeline, 4105 is a drainage pipeline, 4106 is an agitator at the bottom of the barrel, 4107 is a shading ratio detector at the barrel wall, 4108 is a connection measurement window, 42 is a helium-neon laser light source, 43 is an optical detector, 5 is a multi-sensor synchronous control and multi-source data fusion processing module, 51 is a host computer, and 52 is a timing generator. DETAILED DESCRIPTION

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation method of the present invention is described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the invention, so the present invention is not limited by the specific implementation disclosed below.

[0043] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0044] The on-line detection system for the chemical composition of solid-liquid mixture based on laser multispectral analysis specifically includes: a stable solid-liquid two-phase flow generation module 1, a laser induced breakdown spectroscopy measurement module 2, a laser Raman spectroscopy measurement module 3, a laser particle size analysis module 4, and a multi-sensor synchronous control and multi-source data fusion processing module 5. The stable solid-liquid two-phase flow generation module 1 generates a stable solid-liquid two-phase flow sample to be tested through an online sampling and steady flow structure 11, and introduces the sample to be tested into different measurement modules 2, 3, and 4 through a split flow sampling structure 12.

[0045] The stable solid-liquid two-phase flow generation module 1 includes an online sampling and stabilizing flow structure 11 and a shunt sampling structure 12. The online sampling and stabilizing flow structure 11 adopts a slurry automatic stirring and distribution feeding system in patent CN115155385A, and the slurry automatic stirring and distribution feeding system includes: a feeding component 1101, a stirring barrel 1102, an agitator 1103, a slurry valve 1104, a stabilizing flow box 1105, and an outlet 1106; the present invention adds a shunt sampling structure 12 at the lower end of the outlet 1106 to generate two paths, one of which introduces part of the solid-liquid two-phase flow into the particle size analysis module 4 through a pipeline; the other retains most of the remaining solid-liquid two-phase flow to flow vertically downward, and is measured by the laser induced breakdown spectroscopy measurement module 2 and the laser Raman spectroscopy measurement module 3. The online sampling and flow stabilization structure 11 makes the solid-liquid mixture form a uniform and stable solid-liquid two-phase flow only through operations such as stirring and stabilizing the flow. The diverter sampling structure 2 adopts a diverter valve to proportionally introduce the solid-liquid two-phase flow into the particle size measurement module 4 and the spectral measurement modules 2 and 3. While providing a representative solid-liquid mixture sample for measurement to the particle size measurement module 4, it provides a vertically flowing stable liquid column for the laser induced breakdown spectroscopy measurement module 2 and the Raman spectroscopy measurement module 3 directly for spectral measurement, without the need for sample pre-treatment such as sampling, reduction, and drying before measurement.

[0046] The laser induced breakdown spectroscopy measurement module 2 includes a pulse laser 21, a LIBS spectrometer 22, and a laser excitation and plasma luminescence detection optical system 23. In the laser induced breakdown spectroscopy measurement module 2, i) the pulse laser 21 can be configured with a single or two pulse lasers according to specific application requirements, and the laser wavelength can be 355nm, 532nm or 1064nm; ii) the LIBS spectrometer 22 can select a single spectrometer with different detection spectral bands or multiple spectrometers spliced ​​together to detect LIBS spectral signals within the extreme ultraviolet-near infrared wavelength range. Splicing multiple spectrometers refers to introducing plasma luminescence into spectrometers with different spectral bands through a one-to-many optical fiber, and splicing the spectra detected by each spectrometer into a spectrum according to the wavelength. Figure 3 iii) The laser excitation and spectral signal collection of the laser excitation and plasma luminescence detection optical system 23 is realized by a coaxial or non-coaxial optical system. The laser excitation and plasma luminescence detection optical system 23 includes: Figure 3 As shown, the laser beam expansion optical path 2301, the plasma luminescence detection optical path 2302 and the coaxial coupling focusing optical path 2303, the pulse laser 21 focuses the laser onto the surface of the solid-liquid two-phase flow to generate plasma through the laser excitation and the laser beam expansion optical path 2301 and the coaxial coupling focusing optical path 2302 of the plasma luminescence detection optical system 23, and the LIBS spectrometer 22 collects the LIBS spectrum through the laser excitation and the plasma luminescence detection optical path 2302 and the coaxial coupling focusing optical path 2303 of the plasma luminescence detection optical system 23.

[0047] The laser Raman spectrum measurement module 3 includes a continuous laser 31, a Raman spectrometer 32, and a laser emission and Raman scattered light detection optical system 33. The laser emitted by the continuous laser 31 is introduced into the solid-liquid two-phase flow surface by the laser emission and Raman scattered light detection optical system 33, and the Raman scattered spectrum is collected by the Raman spectrometer 32. In the laser Raman spectrum measurement module 3, i) the continuous laser 31 can be configured with a laser of 532nm, 785nm or 1064nm wavelength according to specific application requirements; ii) the Raman spectrometer 33 is selected according to the spectrum corresponding to the laser wavelength; iii) the laser emission and Raman scattered light detection optical system 33 The laser incidence and spectrum collection are realized by a coaxial or non-coaxial optical system.

[0048] The laser excitation and plasma luminescence detection optical system 23 and the laser emission and Raman scattered light detection optical system 33 described in the laser induced breakdown spectroscopy measurement module 2 and the laser Raman spectroscopy measurement module 3 can be two completely independent optical systems or a comprehensive optical system that reuses some components.

[0049] The laser particle size analysis module 4 includes a dilution and dispersion structure 41 and a helium-neon laser light source 42 and an optical detector 43 respectively arranged on both sides thereof. Figure 4 As shown, it includes: a mixing barrel 4101, on which a solid-liquid two-phase flow injection pipeline 4102, a clean water injection pipeline 4103, a measurement output pipeline 4104, and a drainage pipeline 4105 are provided, an agitator 4106 is provided at the bottom of the barrel, a shading ratio detector 4107 is provided on the barrel wall, and the measurement output pipeline 4104 is connected to a measurement window 4108. A portion of the original solid-liquid two-phase flow drained by the diversion sampling structure 12 is injected into the mixing barrel 4101 through the solid-liquid two-phase flow injection pipeline 4102, and is stirred by the agitator 4106 to keep it uniform. Clean water is injected through the clean water injection pipeline 4103 to dilute it, thereby adjusting the shading ratio of the solid-liquid mixture in the mixing barrel 4101. When the shading ratio detector 4107 detects that the shading ratio value is suitable for particle size measurement, the injection of the solid-liquid two-phase flow injection pipeline 4102 and the clean water injection pipeline 4103 is stopped, and the measurement output pipeline 4104 is opened to guide the solid-liquid mixture into the measurement window 4108 for laser particle size measurement. After the measurement, the drain pipeline 4105 and the clean water injection pipeline 4103 are opened to drain the solid-liquid two-phase flow and clean the mixing barrel 4101 and the measurement window 4108. During laser particle size measurement, the helium-neon laser light source 42 irradiates the solid-liquid two-phase flow sample to be measured through the measurement window 4108, and the optical detector 43 detects and collects the Mie scattering spectrum after the helium-neon laser light source 42 passes through 41 on the other side of the measurement window 4108 to obtain particle size information. The particle size data output obtained by the optical detector 43 can be the original Mie scattering spectrum data or the particle size distribution data calculated using the Mie scattering spectrum.

[0050] The multi-sensor synchronous control and multi-source data fusion processing module 5 includes a host computer 51 and a timing generator 52. The host computer 51 realizes the timing control of other modules through the timing generator 52, synchronously measures and saves the spectrum data of each module, performs analysis and calculation, and outputs the chemical composition analysis results. Among them, in the multi-sensor synchronous control and multi-source data fusion processing module 5, the timing generator 52 can provide a multi-channel trigger signal, and while measuring the particle size information, the Raman laser, Raman spectrometer, LIBS laser and LIBS spectrometer are triggered in sequence through different delays to ensure that the laser spectrum measurements of different modules are carried out synchronously, and the collected LIBS and Raman spectra correspond to the solid-liquid two-phase flow samples passing through the system in the same time period.

[0051] The laser induced breakdown spectroscopy measurement module 2 and the laser Raman spectroscopy measurement module 3 are only responsible for collecting spectral data, and do not perform chemical composition analysis based on LIBS or Raman spectroscopy; the quantitative analysis of the atomic and molecular composition in the solid-liquid two-phase flow is realized by the multi-spectral fusion analysis model integrated in the host computer 52 described in the multi-sensor synchronous control and multi-source data fusion processing module 5. The multi-spectral fusion analysis model, with LIBS spectrum, Raman spectrum and Mie scattering spectrum as input, outputs the content of each element and compound in the solid-liquid two-phase flow, namely:

[0052] C x =f(I LIBS ,I Raman ,I Mie )

[0053] Where x is an atom (element) or a molecule (compound); LIBS and I Raman It can be the original spectral data collected by the spectrometer, or it can be the spectral feature data after data preprocessing and feature extraction; Mie It can be the scattering spectrum intensity data at different positions collected by the spectral detector, or it can be the particle size distribution data calculated based on the scattering spectrum. f(·) is the chemical composition quantitative analysis model established by the supervised learning method. The modeling samples are solid-liquid two-phase flow samples with known chemical composition. The training data are the multi-spectral measurement data of each modeling sample, and the response label is the chemical composition content information.

[0054] The online detection method of the chemical composition of solid-liquid mixture based on laser multispectral analysis is as follows:

[0055] First, use Figure 1 The online detection system shown in the figure performs synchronous measurement of LIBS spectrum, Raman spectrum and Mie scattering spectrum on the modeling sample with known chemical composition, and uses the measurement data to establish a quantitative analysis model of chemical composition. The specific process is as follows:

[0056] 1. Perform the following steps S1-S4 on each modeling sample to obtain the Raman spectrum, LIBS spectrum and particle size data of each sample;

[0057] S1: import the modeled sample into the online sampling and steady flow structure 11 to generate a top-down steady solid-liquid two-phase flow;

[0058] S2-1: The host computer 53 controls the laser Raman spectroscopy measurement module 3 through the timing generator 52, and vertically focuses the 785nm laser emitted by the continuous laser 31 onto the surface of the solid-liquid two-phase flow through the laser emission and Raman scattered light detection optical system 33, and uses the Raman spectrometer 32 to collect 200cm -1 ~3500cm -1 The Raman scattering spectrum is denoted as Spc Raman0 , save to the host computer 51;

[0059] S2-2: Call the Raman characteristic peak recognition algorithm on the host computer 51 to find Spc Raman0 The Raman characteristic peaks in the image are calculated and the peak area is saved as I Raman0 ;

[0060] S3-1: The host computer 51 controls the laser induced breakdown spectroscopy measurement module 2 through the timing generator 52, and vertically focuses the 1064nm laser emitted by the pulse laser 21 onto the surface of the solid-liquid two-phase flow to generate plasma through the laser excitation and plasma luminescence detection optical system 23, and uses the LIBS spectrometer 22 to collect the LIBS spectrum in the wide spectral range of 250 to 800nm, which is recorded as Spc LIBS0 , save to the host computer 51;

[0061] S3-2: Call the LIBS characteristic spectrum line recognition algorithm on the host computer 51 to find Spc LIB0S Calculate the LIBS characteristic lines in the spectrum and save them as I LIBS0 ;

[0062] S4: The sample to be tested is introduced into the laser particle size analysis module 4 through the split flow sampling structure 12. Figure 4 The dilution and dispersion structure 41 shown obtains a sample to be tested with an appropriate shading ratio, introduces it into the sample measurement window 4108, records the Mie scattering spectrum formed on the optical detector 43 after the helium-neon laser light source 42 passes through the sample measurement window 4108, obtains the particle size distribution information D0 and saves it to the host computer 51;

[0063] 2. Modeling: Use the Raman spectrum, LIBS spectrum and particle size data of each sample to iteratively learn and train the model to obtain the ideal model;

[0064] S5: Align I by timestamp LIBS0,I Raman0 and D0, and integrated into the modeling data training set feature matrix Features0;

[0065] S6: Through supervised learning method, the quantitative analysis model f(·) of chemical component content is trained with feature matrix Features0 as input data and known chemical component content as reference labels.

[0066] 3. Use the ideal model to make actual predictions;

[0067] After that, you can use Figure 1 The online detection system and the established quantitative analysis model are used to perform online analysis on the chemical composition of the solid-liquid mixture to be tested. The specific process is as follows:

[0068] T1: Introduce the sample to be tested into the online sampling and steady flow structure 11 to obtain a stable solid-liquid two-phase flow from top to bottom;

[0069] T2-1: The host computer 51 controls the laser Raman spectroscopy measurement module 3 through the timing generator 52, and focuses the 785nm laser emitted by the continuous laser 31 vertically onto the surface of the solid-liquid two-phase flow through the laser emission and Raman scattered light detection optical system 33, and collects 200cm -1 ~3500cm -1 The Raman scattering spectrum is denoted as Spc Raman , save to the host computer 51;

[0070] T2-2: Call the Raman characteristic peak recognition algorithm on the host computer 51 to find Spc Raman The Raman characteristic peaks in the image are calculated and the peak area is saved as I Raman ;

[0071] T3-1: The host computer 51 controls the laser induced breakdown spectroscopy measurement module 2 through the timing generator 52, and vertically focuses the 1064nm laser emitted by the pulse laser 21 onto the surface of the solid-liquid two-phase flow to generate plasma through the laser excitation and plasma luminescence detection optical system 23, and uses the LIBS spectrometer 22 to collect the LIBS spectrum in the wide spectral range of 250 to 800nm, which is recorded as Spc LIBS , save to the host computer 51;

[0072] T3-2: Call the LIBS characteristic spectrum line recognition algorithm on the host computer 51 to find Spc LIBS Calculate the LIBS characteristic lines in the spectrum and save them as I LIBS ;

[0073] T4: The sample to be tested is introduced into the laser particle size analysis module 4 through the shunt sampling structure 12, and the sample to be tested with appropriate light shielding ratio is obtained through the dilution and dispersion structure 41, and laser particle size analysis is performed to obtain particle size distribution information D and save it to the host computer 51;

[0074] T5: I LIBS ,I Raman And D are integrated into data feature vector Features;

[0075] T6: Invoke the quantitative analysis model f(·) of chemical composition content in the host computer 51, substitute the feature vector Features, and calculate the chemical composition content C of the sample to be tested. x =f(Features).

[0076] The present invention is described above by way of example in conjunction with the accompanying drawings. The present invention is not limited to the above embodiments. As long as various non-substantial improvements are made using the method concept and technical solution of the present invention, or the concept and technical solution of the invention are directly applied to other occasions without improvement, they are all within the protection scope of the present invention.

Claims

1. An online detection system for the chemical composition of solid-liquid mixture based on laser multispectral analysis, characterized in that: include: A stable solid-liquid two-phase flow generation module (1) includes an online sampling and flow stabilization structure (11) and a flow splitting sampling structure (12); It is used to obtain stable solid-liquid two-phase flow samples to be tested, and introduce the samples to be tested into different measurement modules through the split flow sampling structure; A laser induced breakdown spectroscopy measurement module (2) comprises a pulse laser (21), a LIBS spectrometer (22), and a laser excitation and plasma luminescence detection optical system (23); the module is used to focus the pulse laser onto the surface of the solid-liquid two-phase flow through the laser excitation and plasma luminescence detection optical system (23) to generate plasma, and to collect the LIBS spectrum through the laser excitation and plasma luminescence detection optical system (23) using the LIBS spectrometer (22); The laser Raman spectrum measurement module (3) comprises a continuous laser (31), a Raman spectrometer (32), and a laser emission and Raman scattered light detection optical system (33); the laser is focused onto the surface of the solid-liquid two-phase flow through the laser emission and Raman scattered light detection optical system (33) to generate Raman scattering, and the Raman spectrum is collected using the Raman spectrometer (32); The particle size analysis module (4) comprises a dilution and dispersion structure (41) and a helium-neon laser light source (42) and an optical detector (43) respectively arranged on both sides thereof; the module is used for irradiating the solid-liquid two-phase flow flowing through the dilution and dispersion structure (41) with the helium-neon laser light source (42), measuring Mie scattering with the optical detector (43) and obtaining solid phase particle size distribution data; The multi-sensor synchronous control and multi-source data fusion processing module (5) comprises a host computer (51) and a timing generator (52); the host computer (51) implements timing control of the above modules through the timing generator (52), synchronously stores the spectrum and particle size data measured by each module, performs analysis and calculation, and outputs the chemical composition analysis result.

2. The on-line detection system for chemical composition of solid-liquid mixture based on laser multispectral analysis according to claim 1, characterized in that: The laser induced breakdown spectroscopy measurement module (2) is configured according to specific application requirements 1) with one or two pulsed lasers, the laser wavelength of which can be 355nm, 532nm or 1064nm, as an excitation light source; 2) with a spectrometer of different detection spectrum bands or with a plurality of spectrometers spliced ​​together to detect LIBS spectrum signals within the extreme ultraviolet-near infrared wavelength range; 3) with the laser excitation and plasma luminescence detection optical system (23) being a coaxial or non-coaxial optical system.

3. The on-line detection system for chemical composition of solid-liquid mixture based on laser multispectral analysis according to claim 1, characterized in that: The laser Raman spectrum measurement module (3) is configured according to specific application requirements 1) with a laser of 532nm, 785nm or 1064nm wavelength and a spectrometer of a spectrum corresponding to the laser wavelength, and 2) with a laser emission and Raman scattered light detection optical system (33) used for laser incidence and Raman spectrum collection being a coaxial or non-coaxial optical system.

4. The on-line detection system for chemical composition of solid-liquid mixture based on laser multispectral analysis according to any one of claims 1 to 3, characterized in that: The laser excitation and plasma luminescence detection optical system (23) and the laser emission and Raman scattered light detection optical system (33) are two completely independent optical systems, or a comprehensive optical system that reuses some components.

5. The on-line detection system for chemical composition of solid-liquid mixture based on laser multispectral analysis according to claim 1, characterized in that: The dilution and dispersion structure (41) comprises a mixing barrel (4101), on which a solid-liquid two-phase flow injection pipeline (4102), a clean water injection pipeline (4103), a measurement output pipeline (4104), and a drainage pipeline (4105) are provided. An agitator (4106) is provided at the bottom of the barrel, a shading ratio detector (4107) is provided on the barrel wall, and the measurement output pipeline (4104) is connected to a measurement window (4108).

6. An online detection method for the chemical composition of a solid-liquid mixture based on laser multispectral analysis, characterized in that: Through the following measurement and analysis steps, spectral data are collected, a multi-spectral fusion analysis model is established, and online detection is performed to perform online real-time analysis and measurement of different solid-liquid mixed substances. The method includes the following steps: Step 1: The host computer (51) of the multi-sensor synchronous control and multi-source data fusion processing module (5) controls the timing generator (52) to output multi-channel trigger signals through different delays, and respectively controls the laser induced breakdown spectroscopy measurement module (2), the laser Raman spectroscopy measurement module (3), and the particle size analysis module (4) to collect multi-source heterogeneous data of the same time sequence for each solid-liquid two-phase flow sample in the modeling sample set: LIBS spectrum, Raman spectrum, and solid phase particle size distribution data; Step 2: The host computer (51) fuses multi-source heterogeneous data to perform chemical composition modeling and analysis, including taking the LIBS spectra, Raman spectra and particle size data of all solid-liquid two-phase flow samples in the modeling sample set as input, and taking the contents of each element and compound in the solid-liquid two-phase flow as output, to construct a chemical composition analysis model; Step 3: Collect multi-source heterogeneous data of the solid-liquid two-phase flow sample under the same time sequence in real time, input the chemical composition analysis model for solution, and output the content of the elements and compounds to be detected in real time.

7. The method for online detection of chemical composition of solid-liquid mixture based on laser multispectral analysis according to claim 6, characterized in that: The collecting of multi-source heterogeneous data includes: Step 11: The laser induced breakdown spectroscopy measurement module (2) receives a trigger instruction, controls the pulse laser (21) to output a pulse laser, focuses the laser on the surface of the solid-liquid two-phase flow through the laser excitation and plasma luminescence detection optical system (23) to generate plasma, and controls the LIBS spectrometer (22) to collect the LIBS spectrum through the laser excitation and plasma luminescence detection optical system (23); Step 12: The laser Raman spectrum measurement module (3) receives a trigger instruction, controls the continuous laser (31) to output laser light, focuses the laser light onto the surface of the solid-liquid two-phase flow through the laser emission and Raman scattered light detection optical system (33) to generate Raman scattering, and controls the Raman spectrometer (32) to collect Raman spectra through the laser emission and Raman scattered light detection optical system (33); Step 13: The particle size analysis module (4) receives a trigger instruction, controls the helium-neon laser light source (42) to irradiate the solid-liquid two-phase flow passing through the dilution and dispersion structure (41), collects Mie scattering spectrum information obtained on the optical detector (43), and calculates the solid phase particle size distribution data.

8. The method for online detection of chemical composition of solid-liquid mixture based on laser multispectral analysis according to claim 7, characterized in that: The particle size analysis module (4) collects solid phase particle size distribution data including the following steps: Step 13a: The host computer (51) controls a portion of the original solid-liquid two-phase flow diverted by the flow splitting sampling structure (12) to be injected into the mixing barrel (4101) through the solid-liquid two-phase flow injection pipeline (4102), controls the agitator (4106) to stir and maintain uniformity, and controls the clean water injection pipeline (4103) to inject clean water for dilution, thereby adjusting the shading ratio of the solid-liquid mixture in the mixing barrel (4101); Step 13b: When the shading ratio detector (4107) detects that the shading ratio value is suitable for particle size measurement, the injection of the solid-liquid two-phase flow injection pipeline (4102) and the clean water injection pipeline (4103) is stopped, and the measurement output pipeline (4104) is controlled to open to guide the solid-liquid mixture into the measurement window (4108) to perform laser particle size measurement; Step 13c: After the measurement is completed, the drainage pipeline (4105) and the clean water injection pipeline (4103) are controlled to open, the solid-liquid two-phase flow is emptied and the mixing barrel (4101) and the measurement window (4108) are cleaned.

9. The method for online detection of chemical composition of solid-liquid mixture based on laser multispectral analysis according to claim 6, characterized in that: The chemical composition analysis model is: C x =f(I LIBS ,I Raman ,I Mie ) Where x is an atom (element) or molecule (compound), I LIBS ,I Raman and I Mie are LIBS spectrum, Raman spectrum and Mie scattering spectrum data respectively. f(·) is the quantitative analysis model of chemical composition established by supervised learning method. The modeling samples are solid-liquid two-phase flow samples with known chemical composition. The training data are multi-spectral measurement data of each modeling sample. The response label is the chemical composition content information.

10. The method for online detection of chemical composition of solid-liquid mixture based on laser multispectral analysis according to claim 9, characterized in that: The LIBS spectrum and Raman spectrum data I LIBS and I Raman , which is the original spectral data collected by the spectrometer, or the spectral feature data after data preprocessing and feature extraction; The Mie scattering spectrum data I Mie , is the scattering spectrum intensity data at different positions collected by the spectral detector, or the particle size distribution data calculated by the laser particle size analyzer based on the scattering spectrum.

Images