CN116223482A - Water quality detection method and device based on LIBS and Raman spectrum combined machine learning - Google Patents
Water quality detection method and device based on LIBS and Raman spectrum combined machine learning Download PDFInfo
- Publication number
- CN116223482A CN116223482A CN202310501249.XA CN202310501249A CN116223482A CN 116223482 A CN116223482 A CN 116223482A CN 202310501249 A CN202310501249 A CN 202310501249A CN 116223482 A CN116223482 A CN 116223482A
- Authority
- CN
- China
- Prior art keywords
- laser
- water quality
- computer
- machine learning
- sample
- 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
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 53
- 238000001514 detection method Methods 0.000 title claims abstract description 35
- 238000010801 machine learning Methods 0.000 title claims abstract description 29
- 238000002536 laser-induced breakdown spectroscopy Methods 0.000 title claims abstract description 24
- 238000001237 Raman spectrum Methods 0.000 title claims description 10
- 239000003344 environmental pollutant Substances 0.000 claims abstract description 36
- 231100000719 pollutant Toxicity 0.000 claims abstract description 36
- 238000000034 method Methods 0.000 claims abstract description 23
- 238000001069 Raman spectroscopy Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 6
- 239000000523 sample Substances 0.000 claims description 65
- 239000013307 optical fiber Substances 0.000 claims description 18
- 230000005855 radiation Effects 0.000 claims description 16
- 238000013528 artificial neural network Methods 0.000 claims description 11
- 238000001228 spectrum Methods 0.000 claims description 7
- 239000004973 liquid crystal related substance Substances 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 230000009467 reduction Effects 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 3
- 230000015556 catabolic process Effects 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 230000001537 neural effect Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000007906 compression Methods 0.000 claims description 2
- 230000006835 compression Effects 0.000 claims description 2
- 230000010365 information processing Effects 0.000 claims description 2
- 238000000513 principal component analysis Methods 0.000 claims description 2
- 238000012372 quality testing Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 16
- 230000003595 spectral effect Effects 0.000 abstract description 7
- 239000000126 substance Substances 0.000 abstract description 5
- 238000010521 absorption reaction Methods 0.000 abstract description 2
- 230000007704 transition Effects 0.000 abstract description 2
- 230000003287 optical effect Effects 0.000 description 5
- 239000000356 contaminant Substances 0.000 description 4
- 239000000835 fiber Substances 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000004737 colorimetric analysis Methods 0.000 description 1
- 238000013144 data compression Methods 0.000 description 1
- 230000005518 electrochemistry Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000002262 irrigation Effects 0.000 description 1
- 238000003973 irrigation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000002957 persistent organic pollutant Substances 0.000 description 1
- 238000011202 physical detection method Methods 0.000 description 1
- 230000005180 public health Effects 0.000 description 1
- 239000010453 quartz Substances 0.000 description 1
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N silicon dioxide Inorganic materials O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 1
- 238000002798 spectrophotometry method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/65—Raman scattering
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A20/00—Water conservation; Efficient water supply; Efficient water use
- Y02A20/20—Controlling water pollution; Waste water treatment
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Pathology (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Biochemistry (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Immunology (AREA)
- Mathematical Physics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Artificial Intelligence (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Computational Linguistics (AREA)
- Plasma & Fusion (AREA)
- Optics & Photonics (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The invention discloses a water quality detection method and a device based on LIBS and Raman spectroscopy combined with machine learning, wherein the water quality detection is carried out by combining a laser-induced breakdown spectroscopy technology and a Raman spectroscopy technology and combining a machine learning technology, and qualitative and quantitative detection of water quality pollution is realized by expanding a database and comparing data; the method utilizes various electromagnetic wave wavelengths released by electron energy level transition after energy absorption of different structural molecules of different pollutants, and the spectral intensities corresponding to the wavelengths, and the Raman scattered light with different wavelengths scattered by different molecular structures, and identifies the pollutant types in water to detect water quality through machine learning algorithm processing of spectral information such as wavelength, light intensity data, spectral line number, frequency shift and the like, and can learn infinitely by using a laser-induced breakdown spectroscopy technology and a Raman spectroscopy technology, and has the capability of distinguishing similar substances.
Description
Technical Field
The invention relates to the field of water quality detection, in particular to a water quality detection method and device based on LIBS and Raman spectrum combined machine learning.
Background
The water quality monitoring has great significance in protecting public health, environment and economy, agricultural irrigation, industrial manufacture and the like as the requirement of human beings and natural systems. The currently commonly used water quality detection methods include chemical detection methods: and judging the water quality by detecting the concentration of specific chemical substances in the water sample. The method comprises spectrophotometry, electrochemistry, colorimetry and the like. However, this method requires a large amount of chemical reagents, is complicated to operate, and is difficult to detect organic pollutants. Biological detection method: the water quality is judged by detecting the number and the type of organisms in water. The method comprises a biological index method and a biological sensor method. However, this method requires the use of biological specimens, requires a certain skill and experience for the operation, and may be disturbed by environmental factors. Physical detection method: the water quality is judged by detecting physical characteristics of the water sample, such as temperature, color, turbidity, pH value and the like. The method is simple and easy to implement, but is difficult to detect some chemical pollutants. In conclusion, the existing methods have the defects of high cost, complex sample treatment, long detection period and the like.
Disclosure of Invention
The invention aims to: in order to overcome the defects in the prior art, the invention provides a method and a device for qualitatively and quantitatively detecting water quality pollution by combining a laser-induced breakdown spectroscopy technology and a Raman spectroscopy technology and combining a machine learning technology and by expanding a database and comparing data.
In order to achieve the above purpose, the invention adopts the following technical scheme: a water quality detection method based on LIBS and Raman spectrum combined machine learning utilizes a Raman spectrum detection system to collect scattered light signals of laser irradiation sample water quality, then utilizes the LIBS detection system to collect electromagnetic wave data obtained by the laser irradiation sample water quality, inputs the two-time spectrum data into a computer, establishes a machine learning classification recognition network by utilizing a back propagation error neural algorithm, builds a local database, and then realizes the recognition of a detected sample by fusion and comparison analysis of data information after detection of unknown pollution to be detected, and specifically comprises the following steps:
step 1, a first laser is controlled to emit laser to a sample with known pollutants, a scattered light signal is received through an optical fiber probe, and the scattered light signal is converted into an electric signal and transmitted to a computer;
step 4, continuously replacing marked common pollutants, repeating the steps 1-3, preprocessing data information, establishing a machine learning classification recognition network by using a back propagation error neural network algorithm, and constructing a local database;
and 7, comparing and analyzing pollutant components by the computer according to a local database and a neural network algorithm.
As a preferred embodiment of the present invention: in the step 3, normalization processing is performed on each channel of each sampling result of the input spectrum data, and the normalized data is input to a machine learning recognition algorithm in a computer as input data of a corresponding wavelength.
As a preferred embodiment of the present invention: in the step 4, the data preprocessing includes principal component analysis, linear discrimination dimension reduction, and data information processing and compression.
As a preferred embodiment of the present invention: and in the step 7, the computer judges the pollutants through a local database and a linear discriminant dimension reduction and back propagation error neural network.
In another aspect, an apparatus for a water quality testing method based on LIBS and raman spectroscopy combined with machine learning, includes:
a shell, wherein a carrying turntable for placing water quality samples is arranged in the shell, a case for placing a computer is arranged on one side of the carrying turntable, and a reflecting lens for refracting the water quality samples onto the carrying turntable is arranged above the inside of the shell;
the first laser is fixedly arranged on the side wall of the inner part of the shell, and the laser emission end of the first laser faces the object carrying turntable;
the second laser is fixedly arranged on the top wall in the shell, and a reflecting lens and a focusing lens are sequentially arranged along the direction of the light path;
the optical fiber probe is arranged on the case and faces the carrying turntable and is used for receiving scattered light of the first laser after passing through the sample and electromagnetic waves scattered by the second laser after breaking down sample pollutants;
the spectrometer is arranged in the chassis, the input end of the spectrometer is connected with the optical fiber probe, and the output end of the spectrometer is connected with the computer and is used for converting scattered light after the first laser irradiates the sample and electromagnetic waves scattered by the second laser breakdown pollutant into electric signals and transmitting the electric signals into the computer;
the liquid crystal display is arranged on the shell and used for personnel to interact with the whole device;
the switch door is arranged at the position of the shell corresponding to the object carrying rotary table, and the opening and closing direction of the switch door is outwards.
As a preferred embodiment of the present invention: two USB interfaces are arranged on the computer, and openings are arranged on the shell and the case corresponding to the USB interfaces.
As a preferred embodiment of the present invention: the carrying turntable is provided with a groove to be tested, and is driven to rotate by a motor.
As a preferred embodiment of the present invention: the reflecting lens is provided with a reflecting lens adjusting knob, and the focusing lens is provided with a focusing lens adjusting knob.
As a preferred embodiment of the present invention: the laser of the first laser has a laser wavelength of 532 nm; the second laser is an Nd-YAG laser with the laser wavelength of 1065 nm.
Compared with the prior art, the invention has the following beneficial effects: the invention relates to a water quality detection method and a device based on a laser-induced breakdown spectroscopy technology, a Raman spectroscopy technology and a machine learning technology. The method is characterized in that the types of pollutants in water are identified by utilizing various electromagnetic wave wavelengths released by electronic energy level transition of different structural molecules of different pollutants after energy absorption and the spectral intensities corresponding to the wavelengths, and Raman scattered lights with different wavelengths and scattered by different molecular structures, and the types of the pollutants in water are identified by machine learning algorithm processing spectral information such as wavelength, light intensity data, spectral line number, frequency shift and the like so as to detect water quality. The laser-induced breakdown spectroscopy and the Raman spectroscopy can be used for infinite learning, and can distinguish similar substances.
By combining the laser-induced breakdown spectroscopy technology and the Raman spectroscopy technology and combining the machine learning technology to detect water quality, and expanding a database and comparing data, qualitative and quantitative detection of water quality pollution is realized. The Raman spectrum monitoring technology also has the advantages of non-contact and nondestructive detection, short detection time, small sample consumption and the like. LIBS combines with Raman spectrum technology to increase reliability of detection result.
Drawings
FIG. 1 is a schematic flow chart of the present invention;
FIG. 2 is a schematic view of the apparatus of the present invention;
FIG. 3 is a schematic elevational view of the apparatus of the present invention;
fig. 4 is a schematic view of the back of the device structure of the present invention.
Reference numerals: the laser comprises a reflective lens adjusting knob 1, a 532nm laser 2, a case 3, a trigger 4, a carrying turntable 5, a motor 6, a focal lens 7, a liquid crystal display 8, a shell 9, a reflective lens 10, an Nd-YAG laser 11, a focal lens adjusting knob 12, an optical fiber probe 13, an Nd-YAG laser radiating hole 14, a trigger radiating hole 15, a switch door 16, a first USB interface 17, a second USB interface 18 and a power socket 19.
Description of the embodiments
The present invention is further illustrated in the accompanying drawings and detailed description which are to be understood as being merely illustrative of the invention and not limiting of its scope, and various equivalent modifications to the invention will fall within the scope of the appended claims to the skilled person after reading the invention.
As shown in fig. 1, in a water quality detection method combining LIBS and raman spectroscopy with machine learning, a raman spectroscopy detection system is used to collect scattered light signals of laser irradiation sample water quality, then an LIBS detection system is used to collect electromagnetic wave data obtained by the laser irradiation sample water quality, the two-time spectrum data are input into a computer, a machine learning classification recognition network is built by using a back propagation error neural algorithm, a local database is built, and then recognition of a detected sample is realized by fusion and comparison analysis of data information after detection of unknown pollution to be detected, which specifically comprises the following steps:
step 1, a first laser is controlled to emit laser to a sample with known pollutants, a scattered light signal is received through an optical fiber probe 13, and the scattered light signal is converted into an electric signal and transmitted to a computer;
step 4, continuously replacing marked common pollutants, repeating the steps 1-3, preprocessing data information, establishing a machine learning classification recognition network by using a back propagation error neural network algorithm, and constructing a local database;
and 7, comparing and analyzing pollutant components by the computer according to a local database and a neural network algorithm.
The structure of the water quality detection device based on LIBS and Raman spectrum combined machine learning is shown in figures 2, 3 and 4. Comprising the following steps:
a shell 9, wherein a carrying turntable 5 for placing water quality samples is arranged in the shell, a case 3 for placing a computer is arranged on one side of the carrying turntable 5, a reflecting lens 10 for refracting onto the carrying turntable 5 is arranged above the inside, two USB interfaces are arranged on the computer, and openings are arranged on the shell 9 and the case 3 corresponding to the USB interfaces; the carrying turntable 5 is provided with a groove to be tested, used for placing a container carrying a sample, and capable of placing a container such as a quartz cuvette, and the carrying turntable 5 is driven to rotate by a motor 6, the outer side of the shell 9 is provided with a power socket 19, and the power socket 19 is electrically connected with a device arranged in the shell;
the first laser is fixedly arranged on the inner side wall of the shell 9, the laser emitting end of the first laser faces the carrying turntable 5, and the first laser is a laser 2 with the laser wavelength of 532 nm; the method comprises the steps of carrying out a first treatment on the surface of the
The second laser is fixedly arranged on the top wall in the shell 9, and a reflecting lens 10 and a focusing lens 7 are sequentially arranged along the light path direction; the reflecting lens 10 is provided with a reflecting lens 10 adjusting knob 1, the focal lens 7 is provided with a focal lens adjusting knob 12, and the second laser is an Nd-YAG laser 11 with the laser wavelength of 1065 nm. The reflecting lens 10 and the focusing lens 7 form an optical path system, and can be used for adjusting the focus.
The optical fiber probe 13 is arranged on the case 3 and faces the carrying turntable 5, and is used for receiving scattered light of the first laser after passing through the sample and electromagnetic waves scattered after the second laser breaks down sample pollutants;
the spectrometer is arranged in the case 3, the input end of the spectrometer is connected with the optical fiber probe 13, and the output end of the spectrometer is connected with the computer and is used for converting scattered light after the first laser irradiates the sample and electromagnetic waves scattered by the second laser breakdown pollutant into electric signals and transmitting the electric signals into the computer;
a liquid crystal display 8 provided on the housing 9 for personnel to interact with the whole device;
the switch door is arranged at the position of the shell 9 corresponding to the object carrying rotary table 5, and the opening and closing direction of the switch door is outwards.
The computer controls the two lasers to detect in a time sharing mode, the 532nm laser 2 emits laser with power of 100mW under the control of the computer, then the computer controls the trigger 4, the Nd-YAG laser 11 is started to emit laser with duration of 6 nanoseconds and energy of about 260mJ at the frequency of 10Hz, and the laser wavelength is 1064nm. The optical fiber probe 13 collects electromagnetic wave signals. The optical signal is transmitted via an optical fiber to a spectrometer located within the chassis 3. The motor 6 is driven by the control of the computer to rotate the position of the groove to be detected on the carrying turntable 5 to the laser detection position, and the outward opening and closing door 16 is used for conveniently placing or taking out samples during the operation of the instrument, and also for conveniently taking out and cleaning dirt on the container wall, wherein an Nd-YAG laser radiating hole 14 is formed in an Nd-YAG laser, and a trigger radiating hole 15 is also formed in a trigger of the Nd-YAG laser for radiating heat during the operation.
The reflective lens 10 and the focal lens 1 are responsible for adjusting and correcting the optical path. Three small buttons can realize the fine adjustment of the lens in three dimensions.
As shown in fig. 1. Firstly, in the process of establishing a database, a computer controls two lasers to perform time-sharing detection, a 532nm laser 2 emits laser to a sample, a fiber probe 13 receives scattered light data, a Nd-YAG laser 11 emits laser to break down the sample with known pollutants to obtain radiation electromagnetic wave data, and the data is converted into an electric signal by a spectrometer coupled to a case 3 through the fiber probe 13 and transmitted to the computer in the case 3. Data containing common contaminants is collected and provided to a computer, which creates a local database based thereon. And (3) carrying out normalization processing on each channel of each sampling result of the input spectrum data, and inputting the normalized data as input data with corresponding wavelength into a machine learning recognition algorithm in a computer.
The computer uses an LDA-BPANN neural network algorithm to analyze the class of the normalized and pretreated water quality pollutants. The LDA linear discrimination dimension reduction algorithm is used as a supervised learning method, and can realize good data compression and data dimension reduction effects. The BPANN back propagation error neural network consists of three layers, namely an input layer, a hidden layer and an output layer. The spectrum data of each common water quality pollutant is divided into a training group and a testing group, the data in the training group are used for training the neural network parameters, and the data in the testing group is used for checking the quality of the results. The network is continuously trained through errors until the network variable parameters enable the identification to achieve the optimal effect. In the identification process, the two lasers transmit laser in a time-sharing mode to obtain electromagnetic wave data, and the electromagnetic wave data are transmitted to a computer in the form of electric signals through the optical fiber probe 13 and the spectrometer. The following data processing procedure is the same as the previous preprocessing procedure. By comparing the spectral signals of unknown water quality contaminants with the established local database, the output layer will generate a recognition result of the water quality contaminants based on each set of measured input data.
As shown in fig. 2, 3 and 4, a water sample is placed in a container and then placed on a rotary table, a motor 6 works under the control of a computer, the sample to be detected is conveyed to a position to be detected, the computer controls time-sharing detection, a 532nm laser 2 firstly emits 100mW laser, the laser detection sample with integration time being set to 200ms, raman scattered light passing through the sample is received by an optical fiber probe 13 and coupled to a spectrometer in a case 3, after a period of time is collected, the computer controls a trigger 4, the nd-YAG laser 11 emits laser with 10Hz, pulse duration 6ns, energy 260mJ and wavelength of 1064nm, the laser breaks down pollutants in the water sample through a reflecting lens 10 and a focal lens 7, the optical fiber probe 13 receives scattered electromagnetic waves, and the result is transmitted to the spectrometer in the case 3. The spectrometer converts the optical signal into an electrical signal and transmits the electrical signal to a computer in the chassis 3, and the computer compares the previously established or imported local database, determines by using an LDA-BPANN algorithm and displays the result on the liquid crystal display screen 8.
The container loaded with the sample can be placed on the turntable 5 through the switch door 16, the computer controls the motor 6 to rotate the position of the groove to be measured to the laser measuring position, and the sample can be taken out conveniently through the switch door 16, so that the sample can be treated later and the container can be cleaned conveniently.
As shown in fig. 4, a database of established water quality contaminants can be imported to the computer through the first USB interface 17 and the second USB interface 18.
Before being installed and used, a user adjusts and calibrates the optical path by adjusting the reflective lens 10 adjusting knob 1 and the focal lens adjusting knob 12 so as to ensure that the position of the focal point can detect the sample and that the energy attenuation is within a set range near the fiber probe 13. After the power is on, the laser is preheated for about 30 minutes, and the laser can be started for real-time measurement after the preheating. An operator can interact with the water quality detection device through the liquid crystal display 8.
The foregoing is only a preferred embodiment of the invention, it being noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the present invention, and such modifications and adaptations are intended to be comprehended within the scope of the invention.
Claims (9)
1. A water quality detection method based on LIBS and Raman spectrum combined machine learning is characterized in that a Raman spectrum detection system is used for collecting scattered light signals of laser irradiation sample water quality, then an LIBS detection system is used for collecting electromagnetic wave data obtained by the laser irradiation sample water quality, the two-time spectrum data are input into a computer, a machine learning classification recognition network is built by using a back propagation error neural algorithm, a local database is built, and then recognition of a detected sample is realized by fusion and comparison analysis of data information after detection of unknown pollution to be detected, and the method specifically comprises the following steps:
step 1, a first laser is controlled to emit laser to a sample with known pollutants, a scattered light signal is received through an optical fiber probe, and the scattered light signal is converted into an electric signal and transmitted to a computer;
step 2, controlling a second laser to break down electromagnetic waves at a sample radiation position with known pollutants, obtaining radiation electromagnetic wave data through a signal collector, and converting the radiation electromagnetic wave data into electric signals to be transmitted to a computer;
step 3, the computer learns the two times of data obtained in the step 1 and the step 2;
step 4, continuously replacing marked common pollutants, repeating the steps 1-3, preprocessing data information, establishing a machine learning classification recognition network by using a back propagation error neural network algorithm, and constructing a local database;
step 5, after the local database is established, testing samples of unknown pollutants, wherein a 532nm knob laser works first, the samples of the unknown pollutants are emitted with laser, scattered light signals are received through an optical fiber probe, and the scattered light signals are converted into electric signals and transmitted to a computer;
step 6, the second laser excites the sample to radiate electromagnetic waves, radiation electromagnetic wave data are obtained through the signal collector, and the radiation electromagnetic wave data are converted into electric signals to be transmitted to a computer;
and 7, comparing and analyzing pollutant components by the computer according to a local database and a neural network algorithm.
2. The method for detecting water quality based on LIBS and Raman spectroscopy combined with machine learning according to claim 1, wherein in the step 3, normalization processing is performed on each channel of each sampling result of the input spectrum data, and the normalized data is input into a machine learning recognition algorithm in a computer as input data of a corresponding wavelength.
3. The method for detecting water quality based on LIBS and Raman spectroscopy combined with machine learning according to claim 1, wherein in the step 4, the data preprocessing comprises principal component analysis, linear discrimination dimension reduction and data information processing and compression.
4. The method for detecting water quality based on LIBS and Raman spectroscopy combined machine learning according to claim 1, wherein the computer in the step 7 judges the pollutants through a local database and a linear discrimination dimension reduction and back propagation error neural network.
5. An apparatus for a machine learning based water quality testing method based on LIBS and raman spectroscopy according to any one of claims 1 to 4, comprising:
a shell, wherein a carrying turntable for placing water quality samples is arranged in the shell, a case for placing a computer is arranged on one side of the carrying turntable, and a reflecting lens for refracting the water quality samples onto the carrying turntable is arranged above the inside of the shell;
the first laser is fixedly arranged on the side wall of the inner part of the shell, and the laser emission end of the first laser faces the object carrying turntable;
the second laser is fixedly arranged on the top wall in the shell, and a reflecting lens and a focusing lens are sequentially arranged along the direction of the light path;
the optical fiber probe is arranged on the case and faces the carrying turntable and is used for receiving scattered light of the first laser after passing through the sample and electromagnetic waves scattered by the second laser after breaking down sample pollutants;
the spectrometer is arranged in the chassis, the input end of the spectrometer is connected with the optical fiber probe, and the output end of the spectrometer is connected with the computer and is used for converting scattered light after the first laser irradiates the sample and electromagnetic waves scattered by the second laser breakdown pollutant into electric signals and transmitting the electric signals into the computer;
the liquid crystal display is arranged on the shell and used for personnel to interact with the whole device;
the switch door is arranged at the position of the shell corresponding to the object carrying rotary table, and the opening and closing direction of the switch door is outwards.
6. The device of claim 5, wherein two USB interfaces are provided on the computer, and openings are provided on the housing and the case corresponding to the USB interfaces.
7. The device of claim 5, wherein the carrying turntable is provided with a groove to be detected, and the carrying turntable is driven to rotate by a motor.
8. The device of claim 5, wherein the reflective lens is provided with a reflective lens adjusting knob, and the focal lens is provided with a focal lens adjusting knob.
9. The device of the water quality detection method based on LIBS and Raman spectroscopy combined with machine learning according to claim 5, wherein the laser wavelength of the first laser is 532 nm; the second laser is an Nd-YAG laser with the laser wavelength of 1065 nm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310501249.XA CN116223482A (en) | 2023-05-06 | 2023-05-06 | Water quality detection method and device based on LIBS and Raman spectrum combined machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310501249.XA CN116223482A (en) | 2023-05-06 | 2023-05-06 | Water quality detection method and device based on LIBS and Raman spectrum combined machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116223482A true CN116223482A (en) | 2023-06-06 |
Family
ID=86577186
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310501249.XA Pending CN116223482A (en) | 2023-05-06 | 2023-05-06 | Water quality detection method and device based on LIBS and Raman spectrum combined machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116223482A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101493416A (en) * | 2009-03-10 | 2009-07-29 | 中国海洋大学 | Underwater laser Raman spectrum/laser-induced breakdown spectroscopy combined detection device and method |
| CN104897624A (en) * | 2015-04-28 | 2015-09-09 | 四川大学 | Laser-induced breakdown spectroscopy and Raman spectroscopy combination system |
| US20160169805A1 (en) * | 2014-12-16 | 2016-06-16 | Thermo Scientific Portable Analytical Instruments Inc. | Combined raman spectroscopy and laser-induced breakdown spectroscopy |
| CN110196246A (en) * | 2018-02-26 | 2019-09-03 | 成都艾立本科技有限公司 | A kind of laser-induced breakdown-Raman spectrum combined system |
| CN110312486A (en) * | 2016-12-22 | 2019-10-08 | 先进截骨工具 -Aot股份公司 | Laser equipment and tissue characterization's method |
| CN115753740A (en) * | 2022-11-18 | 2023-03-07 | 中国海洋大学 | Underwater ore in-situ detection method and system based on combination of LIBS and pulse Raman |
-
2023
- 2023-05-06 CN CN202310501249.XA patent/CN116223482A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101493416A (en) * | 2009-03-10 | 2009-07-29 | 中国海洋大学 | Underwater laser Raman spectrum/laser-induced breakdown spectroscopy combined detection device and method |
| US20160169805A1 (en) * | 2014-12-16 | 2016-06-16 | Thermo Scientific Portable Analytical Instruments Inc. | Combined raman spectroscopy and laser-induced breakdown spectroscopy |
| CN104897624A (en) * | 2015-04-28 | 2015-09-09 | 四川大学 | Laser-induced breakdown spectroscopy and Raman spectroscopy combination system |
| CN110312486A (en) * | 2016-12-22 | 2019-10-08 | 先进截骨工具 -Aot股份公司 | Laser equipment and tissue characterization's method |
| CN110196246A (en) * | 2018-02-26 | 2019-09-03 | 成都艾立本科技有限公司 | A kind of laser-induced breakdown-Raman spectrum combined system |
| CN115753740A (en) * | 2022-11-18 | 2023-03-07 | 中国海洋大学 | Underwater ore in-situ detection method and system based on combination of LIBS and pulse Raman |
Non-Patent Citations (1)
| Title |
|---|
| 聂卫东等: "BP 神经网络算法在图像预测物体重量中的应用", 《湖南工业职业技术学院学报》, pages 22 - 24 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6731804B1 (en) | Thermal luminescence liquid monitoring system and method | |
| CN109827946B (en) | Soil component in-situ analysis system and method | |
| US7760354B2 (en) | Spectroscopic method and apparatus for identification of a substance using a tunable light source | |
| CN101995395B (en) | Method for online detecting material by laser induction multiple spectrum united fingerprint network | |
| JP2009505070A (en) | System and method for quantification and identification of biological samples suspended in liquid | |
| CN107064084A (en) | Microminiature laser fluorescence spectrum instrument and spectral method of detection | |
| KR20030001892A (en) | gas identification device | |
| CN111562252A (en) | Raman detection system based on coaxial dual-wavelength fluorescence elimination | |
| JP3780646B2 (en) | Multiple component determination method and apparatus for sewage | |
| CN112730383A (en) | Optical fiber array LIBS detection system for online detection | |
| CN109520967A (en) | The detection system and its detection method of trace heavy metal in a kind of food | |
| CN214472708U (en) | Absorption breakdown spectrum combined measurement system based on three-dimensional adjustable multi-optical-path structure | |
| CN114835241A (en) | Real-time monitoring and intelligent ozone adding control method for refractory organic matters in sewage | |
| CN106338499A (en) | Element laser detection and analysis instrument and mineral element analysis method | |
| CN116223482A (en) | Water quality detection method and device based on LIBS and Raman spectrum combined machine learning | |
| CN111175260B (en) | Ocean TOC sensor based on ultraviolet three-dimensional fluorescence and using method | |
| CN113189084A (en) | Quasi-online detection device and method for atmospheric particulates and heavy metals in sewage | |
| CN113267464B (en) | Method and device for detecting multi-component heavy metals in edible oil based on near infrared combined colorimetric sensor array | |
| EP4047364A1 (en) | System and method for measuring water characteristics in a water facility | |
| CN211741079U (en) | Breakdown spectrum and absorption spectrum combined measuring system | |
| CN111272736B (en) | Breakdown spectrum and absorption spectrum combined measurement system and method | |
| CN108982389B (en) | Automatic background and reference correction system and method for visible and near infrared spectrums | |
| CN111413318A (en) | Serum detection system and method based on Raman spectrum | |
| CN113640272A (en) | Spectral library establishing and detecting method based on portable Raman spectrometer and application | |
| CN112666103A (en) | Water quality monitoring system based on spectral characteristics |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20230606 |
|
| RJ01 | Rejection of invention patent application after publication |