CN115219472B - Method and system for quantitatively identifying multiple pollution sources of mixed water body - Google Patents
Method and system for quantitatively identifying multiple pollution sources of mixed water body Download PDFInfo
- Publication number
- CN115219472B CN115219472B CN202210968262.1A CN202210968262A CN115219472B CN 115219472 B CN115219472 B CN 115219472B CN 202210968262 A CN202210968262 A CN 202210968262A CN 115219472 B CN115219472 B CN 115219472B
- Authority
- CN
- China
- Prior art keywords
- pollution source
- fluorescence spectrum
- spectrum data
- dimensional fluorescence
- mixed water
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 91
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000002189 fluorescence spectrum Methods 0.000 claims abstract description 101
- 238000012549 training Methods 0.000 claims abstract description 30
- 238000002156 mixing Methods 0.000 claims abstract description 19
- 238000005259 measurement Methods 0.000 claims abstract description 11
- 238000012795 verification Methods 0.000 claims description 21
- 238000012360 testing method Methods 0.000 claims description 14
- 238000007781 pre-processing Methods 0.000 claims description 11
- 238000011156 evaluation Methods 0.000 claims description 8
- 238000001069 Raman spectroscopy Methods 0.000 claims description 7
- 238000013527 convolutional neural network Methods 0.000 claims description 7
- 229910021642 ultra pure water Inorganic materials 0.000 claims description 4
- 239000012498 ultrapure water Substances 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 2
- 239000011521 glass Substances 0.000 claims description 2
- 239000012528 membrane Substances 0.000 claims description 2
- 238000004321 preservation Methods 0.000 claims description 2
- 239000000203 mixture Substances 0.000 claims 1
- 230000007613 environmental effect Effects 0.000 abstract description 3
- 239000010865 sewage Substances 0.000 description 10
- 238000004364 calculation method Methods 0.000 description 6
- 238000001228 spectrum Methods 0.000 description 4
- 238000000862 absorption spectrum Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000004927 fusion Effects 0.000 description 3
- 239000013535 sea water Substances 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000007635 classification algorithm Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000005284 excitation Effects 0.000 description 2
- 238000004811 liquid chromatography Methods 0.000 description 2
- 244000144972 livestock Species 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 244000144977 poultry Species 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000003911 water pollution Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000009395 breeding Methods 0.000 description 1
- 230000001488 breeding effect Effects 0.000 description 1
- 238000005352 clarification Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001506 fluorescence spectroscopy Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000009364 mariculture Methods 0.000 description 1
- 239000008239 natural water Substances 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000002133 sample digestion Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000006228 supernatant Substances 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/64—Fluorescence; Phosphorescence
- G01N21/6428—Measuring fluorescence of fluorescent products of reactions or of fluorochrome labelled reactive substances, e.g. measuring quenching effects, using measuring "optrodes"
- G01N21/643—Measuring fluorescence of fluorescent products of reactions or of fluorochrome labelled reactive substances, e.g. measuring quenching effects, using measuring "optrodes" non-biological material
-
- 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
- G01N21/64—Fluorescence; Phosphorescence
- G01N21/6402—Atomic fluorescence; Laser induced fluorescence
-
- 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
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/18—Water
-
- 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
- G01N21/64—Fluorescence; Phosphorescence
- G01N2021/6417—Spectrofluorimetric devices
-
- 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
- G01N21/64—Fluorescence; Phosphorescence
- G01N21/6428—Measuring fluorescence of fluorescent products of reactions or of fluorochrome labelled reactive substances, e.g. measuring quenching effects, using measuring "optrodes"
- G01N2021/6439—Measuring fluorescence of fluorescent products of reactions or of fluorochrome labelled reactive substances, e.g. measuring quenching effects, using measuring "optrodes" with indicators, stains, dyes, tags, labels, marks
- G01N2021/6441—Measuring fluorescence of fluorescent products of reactions or of fluorochrome labelled reactive substances, e.g. measuring quenching effects, using measuring "optrodes" with indicators, stains, dyes, tags, labels, marks with two or more labels
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2201/00—Features of devices classified in G01N21/00
- G01N2201/12—Circuits of general importance; Signal processing
- G01N2201/129—Using chemometrical methods
- G01N2201/1296—Using chemometrical methods using neural networks
-
- 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
- Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Immunology (AREA)
- General Health & Medical Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Physics & Mathematics (AREA)
- Pathology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Optics & Photonics (AREA)
- Molecular Biology (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Engineering & Computer Science (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The invention discloses a method and a system for quantitatively identifying multiple pollution sources of a mixed water body, wherein the method comprises the following steps: obtaining a single pollution source sample and performing three-dimensional fluorescence spectrum measurement to obtain single pollution source three-dimensional fluorescence spectrum data; mixing the three-dimensional fluorescence spectrum data of the single pollution source to obtain a three-dimensional fluorescence spectrum data set of the pollution source; training a pre-constructed mixed water body multi-pollution source quantitative identification model by utilizing a pollution source three-dimensional fluorescence spectrum data set to obtain a mixed water body multi-pollution source quantitative identification model; and carrying out quantitative identification on the sample to be detected based on the mixed water body multi-pollution source quantitative identification model to obtain a quantitative identification result of the sample to be detected. The system comprises: the device comprises a measuring module, a mixing module, a training module and a quantitative identification module. By using the method, a plurality of pollution sources in the mixed water body can be rapidly identified, and the duty ratio conditions of various pollution sources can be given. The method and the system for quantitatively identifying the mixed water body multi-pollution sources can be widely applied to the technical field of environmental supervision.
Description
Technical Field
The invention relates to the technical field of environmental supervision, in particular to a method and a system for quantitatively identifying multiple pollution sources of a mixed water body.
Background
The water pollution tracing of the estuary area is always a hot spot and a difficult point in the field of environmental supervision. The area is difficult to separate the mixed water pollution sources in the estuary area due to continuous mixing of seawater, incoming water upstream of the river, various industrial enterprises, domestic sewage, livestock and poultry cultivation, mariculture, agriculture and forestry drainage, atmospheric rainfall and the like under the tidal and runoff effects, and the accurate prevention and control of sea area pollution are not facilitated.
Chinese patent document No. CN111426668A discloses a method for tracing, classifying and identifying polluted water by utilizing three-dimensional fluorescence spectrum characteristic information, classifying sample pollution types by a neural network, and obtaining suspected tracing information by similarity matching; chinese patent document No. CN113311081a discloses a pollution source identification method and apparatus based on three-dimensional liquid chromatography fingerprint, the method utilizes self-organizing neural network to realize automatic comparison and identification of three-dimensional liquid chromatography fingerprint, thereby identifying pollution source; the Chinese patent document No. CN113033623A discloses a pollution source identification method and a system based on ultraviolet-visible absorption spectrum, the method comprises the steps of collecting a pollution source sample and preprocessing the pollution source sample, performing ultraviolet-visible absorption spectrum test on the preprocessed pollution source sample to obtain spectrum data of the pollution source sample, preprocessing the spectrum data, performing standard normal transformation on the preprocessed spectrum data, establishing a pollution source identification model according to the spectrum data after standard normal transformation and a classification algorithm, training the pollution source identification model, and performing pollution source identification through the trained pollution source identification model; chinese patent document No. CN113011478A discloses a pollution source identification method and a system based on data fusion, wherein the method is used for carrying out pollution index test on a pollution source sample after pretreatment to obtain conventional water quality data, ultraviolet-visible absorption spectrum data and three-dimensional fluorescence spectrum data, carrying out feature extraction on the pretreated test data, splicing the extracted feature data to construct fusion data, establishing a pollution source identification model according to the fusion data and a classification algorithm, training, and carrying out pollution source identification through the trained pollution source identification model; the methods can well identify pollution sources, but cannot distinguish various pollution sources of the mixed water body, and cannot obtain the duty ratio of the multiple pollution sources.
The method can only provide the approximate pollution source duty ratio in a survey river, but can not provide the pollution source duty ratio of a specific station although the method can provide the duty ratio of various pollution sources based on the isotope tracing method, and the method relates to isotope detection and is extremely complex in measurement.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a method and a system for quantitatively identifying multiple pollution sources of a mixed water body, which can rapidly identify multiple pollution sources in the mixed water body and provide the duty ratio conditions of various pollution sources.
The first technical scheme adopted by the invention is as follows: a method for quantitatively identifying multiple pollution sources of a mixed water body, comprising the following steps:
obtaining a single pollution source sample and performing three-dimensional fluorescence spectrum measurement to obtain single pollution source three-dimensional fluorescence spectrum data;
mixing the three-dimensional fluorescence spectrum data of the single pollution source to obtain a three-dimensional fluorescence spectrum data set of the pollution source;
training a pre-constructed mixed water body multi-pollution source quantitative identification model by utilizing a pollution source three-dimensional fluorescence spectrum data set to obtain a mixed water body multi-pollution source quantitative identification model;
and carrying out quantitative identification on the sample to be detected based on the mixed water body multi-pollution source quantitative identification model to obtain a quantitative identification result of the sample to be detected.
Further, pretreatment of the single pollution source sample is also included, wherein the pretreatment comprises the steps of filtering the single pollution source sample through a 0.22 mu m filter membrane and filling the single pollution source sample into a brown glass bottle for light-shielding preservation.
Further, the step of mixing the three-dimensional fluorescence spectrum data of the single pollution source to obtain a three-dimensional fluorescence spectrum data set of the pollution source specifically comprises the following steps:
preprocessing the single pollution source three-dimensional fluorescence spectrum data to obtain preprocessed single pollution source three-dimensional fluorescence spectrum data;
classifying the preprocessed single pollution source three-dimensional fluorescence spectrum data according to different types to obtain different types of single pollution source three-dimensional fluorescence spectrum data;
mixing two or more kinds of single pollution source three-dimensional fluorescence spectrum data according to different proportions to obtain multi-pollution source three-dimensional fluorescence spectrum data;
and integrating the single pollution source three-dimensional fluorescence spectrum data and the multi-pollution source three-dimensional fluorescence spectrum data to obtain a pollution source three-dimensional fluorescence spectrum data set.
Further, the pretreatment of the single pollution source three-dimensional fluorescence spectrum data comprises subtracting an ultrapure water blank, converting the fluorescence intensity into a raman unit (r.u.), and removing the rayleigh scattering and the raman scattering.
Further, the step of training the pre-constructed mixed water body multi-pollution source quantitative identification model by utilizing the pollution source three-dimensional fluorescence spectrum data set to obtain the mixed water body multi-pollution source quantitative identification model specifically comprises the following steps:
dividing a pollution source three-dimensional fluorescence spectrum data set into a training set and a verification set;
training a pre-constructed mixed water body multi-pollution source quantitative recognition model by using a training set, outputting a pollution source type recognition result by taking cross entropy as a loss function, and outputting a pollution source duty ratio result by taking mean square error as the loss function to obtain a trained mixed water body multi-pollution source quantitative recognition model;
performing accuracy verification on the trained mixed water multi-pollution-source quantitative identification model by using a verification set to obtain an accuracy verification result;
and selecting the trained mixed water body multi-pollution source quantitative identification model with the highest accuracy according to the accuracy verification result to obtain the mixed water body multi-pollution source quantitative identification model.
Further, the method also comprises the step of evaluating the model performance of the mixed water body multi-pollution source quantitative identification model, and specifically comprises the following steps:
acquiring a test set and testing a mixed water body multi-pollution source quantitative identification model to obtain model performance evaluation parameters;
and evaluating the quantitative identification model of the multiple pollution sources of the mixed water body according to the model performance evaluation parameters to obtain a model performance evaluation result.
The second technical scheme adopted by the invention is as follows: a system for quantitatively identifying multiple sources of pollution in a mixed body of water, comprising:
the measuring module is used for acquiring a single pollution source sample and carrying out three-dimensional fluorescence spectrum measurement to obtain single pollution source three-dimensional fluorescence spectrum data;
the mixing module is used for mixing the three-dimensional fluorescence spectrum data of the single pollution source to obtain a three-dimensional fluorescence spectrum data set of the pollution source;
the training module is used for training the pre-constructed mixed water body multi-pollution source quantitative identification model by utilizing the pollution source three-dimensional fluorescence spectrum data set to obtain the mixed water body multi-pollution source quantitative identification model;
and the quantitative recognition module is used for quantitatively recognizing the sample to be detected based on the mixed water body multi-pollution source quantitative recognition model to obtain a quantitative recognition result of the sample to be detected.
The method and the system have the beneficial effects that: firstly, the single pollution source sample is preprocessed, so that the construction of a quantitative identification model of the mixed water body multi-pollution source can be reduced, and the accuracy of the single pollution source sample test is ensured; secondly, preprocessing the three-dimensional fluorescence spectrum data of the single pollution source in order to enable the three-dimensional fluorescence spectrum data of the single pollution source to be more accurate and visual; finally, dividing the pollution source three-dimensional fluorescence spectrum data set into a training set, a verification set and a test set, training the model by adopting the training set, verifying the model accuracy by adopting the verification set, selecting a model with small model error and highest accuracy as an optimal model, testing the model by adopting the test set, evaluating the model performance, and ensuring the accuracy of the construction of the mixed water body multi-pollution source quantitative identification model; by using the three-dimensional fluorescence spectrum technology, the method has the advantages of simple and efficient sampling, pretreatment and measurement, accurate and sensitive pollution source identification, and can quickly and accurately identify the pollution source and the duty ratio of the mixed water body by combining the multi-label and multi-task convolutional neural network model, thereby having low cost, high timeliness and strong operability, being beneficial to large-scale popularization and having important significance on pollution tracing of the mixed water body.
Drawings
FIG. 1 is a flow chart of the steps of a method for quantitatively identifying multiple pollution sources of a mixed water body according to the present invention;
FIG. 2 is a block diagram of a system for quantitatively identifying multiple sources of pollution in a body of water in accordance with the present invention.
Detailed Description
The invention will now be described in further detail with reference to the drawings and to specific examples. The step numbers in the following embodiments are set for convenience of illustration only, and the order between the steps is not limited in any way, and the execution order of the steps in the embodiments may be adaptively adjusted according to the understanding of those skilled in the art.
Referring to fig. 1, the present invention provides a method for quantitatively identifying multiple pollution sources of a mixed water body, comprising the following steps:
s1, acquiring a single pollution source sample and performing three-dimensional fluorescence spectrum measurement to obtain single pollution source three-dimensional fluorescence spectrum data;
s1.1, acquiring a single pollution source sample and preprocessing the single pollution source sample to obtain a preprocessed single pollution source sample;
specifically, firstly, in order to ensure the integrity and accuracy of a quantitative identification model of multiple pollution sources of a mixed water body, an acquired single pollution source sample comprises typical representative samples of all possible pollution source types in an acquisition area, a target area selects a natural water body of a coastal river as a research object, and the possible pollution sources of the water body are respectively: 4 types of sewage are discharged from seawater backward flowing facilities and town sewage treatment facilities, agricultural rural area source sewage is discharged from rainwater and the like, wherein the agricultural rural area source sewage is further divided into rural living direct sewage, livestock and poultry breeding sewage, agriculture and forestry drainage and the like, and 20 typical representative samples of each type are collected in different places in a river basin, and the total number of the representative samples is 80.
Secondly, the pretreatment method of the single pollution source sample is carried out in a plurality of ways, a natural clarification method can be adopted, the supernatant liquid of 2/3 of the upper layer is used for analysis and determination, and a water sample digestion method can also be adopted.
S1.2, performing three-dimensional fluorescence spectrum measurement on the pretreated single pollution source sample to obtain single pollution source three-dimensional fluorescence spectrum data.
Specifically, the scanning range of the excitation wavelength and the emission wavelength of the three-dimensional fluorescence spectrum adopted by the three-dimensional fluorescence spectrum measurement is 200-600 nm, and the scanning interval range is 1-10 nm.
S2, mixing the three-dimensional fluorescence spectrum data of the single pollution source to obtain a three-dimensional fluorescence spectrum data set of the pollution source;
s2.1, preprocessing the single pollution source three-dimensional fluorescence spectrum data to obtain preprocessed single pollution source three-dimensional fluorescence spectrum data;
specifically, in order to make the three-dimensional fluorescence spectrum data of the single pollution source more accurate and visual, the obtained three-dimensional fluorescence spectrum data of the single pollution source also needs to be preprocessed, and as a preferable scheme of the embodiment, the preprocessing of the three-dimensional fluorescence spectrum data of the single pollution source comprises subtracting the blank of ultrapure water, and the fluorescence intensity of the original fluorescence fingerprint is converted into a raman unit (r.u.) by utilizing the integral of the raman scattering intensity of the ultrapure water with the excitation wavelength of 350nm and the emission wavelength of 381-426 nm, so that the rayleigh scattering and raman scattering are removed.
S2.2, classifying the preprocessed single pollution source three-dimensional fluorescence spectrum data according to different types to obtain different types of single pollution source three-dimensional fluorescence spectrum data;
specifically, the types of the pretreated single pollution source three-dimensional fluorescence spectrum data are classified according to four categories of seawater backflow, urban sewage treatment facility sewage discharge, agricultural rural area source sewage discharge and rainwater, and the serial numbers are S, W, N and R in sequence.
S2.3, mixing two or more kinds of single pollution source three-dimensional fluorescence spectrum data according to different proportions to obtain multi-pollution source three-dimensional fluorescence spectrum data;
specifically, as a preferred scheme of this embodiment, the three-dimensional fluorescence spectrum data of different types of single pollution sources are mixed in a mode of combining two by two, and the ratio is selected from 2:8, 3:7 and 4:6, so that the fluorescence data of 14400 mixed samples are obtained in total, and the process needs to ensure that the mixed water sample contains all combinations of pollution source types.
S2.4, integrating the single pollution source three-dimensional fluorescence spectrum data and the multi-pollution source three-dimensional fluorescence spectrum data to obtain a pollution source three-dimensional fluorescence spectrum data set.
S3, training a pre-constructed mixed water body multi-pollution source quantitative identification model by utilizing a pollution source three-dimensional fluorescence spectrum data set to obtain a mixed water body multi-pollution source quantitative identification model;
specifically, as a preferred scheme of the embodiment, the mixed water body multi-pollution source quantitative recognition model adopts a multi-label multi-task convolutional neural network model, adopts the same convolutional neural network to simultaneously complete two tasks, one is recognition of pollution source types, one is calculation of various pollution source duty ratios, cross entropy is adopted as a loss function for the pollution source type recognition, mean square error is adopted as a loss function for the pollution source duty ratios, and the total loss function of the mixed water body multi-pollution source quantitative recognition model adopts the sum of the two; multiple tags, meaning that each pollution source corresponds to a tag, for a mixed water sample to be detected, the pollution source type may be one or more.
The number of labels of the task is 4, and four categories S, W, N, R are respectively represented by 0,1,2 and 3.
S3.1, dividing a pollution source three-dimensional fluorescence spectrum data set into a training set and a verification set;
specifically, the three-dimensional fluorescence spectrum data of the pollution source is combined into 14400 groups, 90% of sample data is randomly selected as a training set, and the rest 10% of sample data is used as a verification set.
S3.2, training the pre-constructed mixed water body multi-pollution source quantitative identification model by utilizing a training set to obtain a trained mixed water body multi-pollution source quantitative identification model;
s3.3, performing accuracy verification on the trained mixed water body multi-pollution source quantitative identification model by using a verification set to obtain an accuracy verification result;
specifically, the accuracy verification result includes the accuracy of the pollution source category identification and the error of the pollution source duty ratio calculation result.
The correct rate of the pollution source category identification describes the proportion of the number of correctly classified samples to the total number of classified samples, and the calculation formula is specifically as follows:
in the above, A TP To correctly classify the number of samples, A TN For incorrectly classifying the number of samples, A total For classifying the total number of samples.
The error of the pollution source duty ratio calculation result adopts a mean square error, and the expression is specifically as follows:
in the above formula, m is the total number of samples, x n Is the true value of the nth sample, y n Is the predicted value of the nth sample.
S3.4, selecting the trained mixed water body multi-pollution source quantitative identification model with the highest accuracy according to the accuracy verification result, and obtaining the mixed water body multi-pollution source quantitative identification model.
S4, carrying out quantitative identification on the sample to be detected based on the mixed water body multi-pollution source quantitative identification model to obtain a quantitative identification result of the sample to be detected.
The method further comprises the steps of performing model performance evaluation on the mixed water body multi-pollution source quantitative identification model, dividing the pollution source three-dimensional fluorescence spectrum data set in the step S3 into a training set, a verification set and a test set according to the ratio of 18:1:1, and testing the mixed water body multi-pollution source quantitative identification model obtained through the training set and the verification set by using the test set to obtain the accuracy of model performance evaluation parameter pollution source type identification and the error of pollution source duty ratio calculation result, wherein the accuracy of pollution source type identification is about 88%, so that the mixed water body multi-pollution source quantitative identification model can distinguish water samples of single pollution sources and can distinguish water samples mixed by multiple pollution sources; the error of the pollution source duty ratio calculation result is about 0.0052, so that the mixed water body multi-pollution source quantitative identification model is accurate and reliable, and has important reference value for pollution tracing of coastal areas entering sea and river.
As shown in fig. 2, a system for quantitatively identifying multiple pollution sources of a mixed water body comprises:
the measuring module is used for acquiring a single pollution source sample and carrying out three-dimensional fluorescence spectrum measurement to obtain single pollution source three-dimensional fluorescence spectrum data;
the mixing module is used for mixing the three-dimensional fluorescence spectrum data of the single pollution source to obtain a three-dimensional fluorescence spectrum data set of the pollution source;
the training module is used for training the pre-constructed mixed water body multi-pollution source quantitative identification model by utilizing the pollution source three-dimensional fluorescence spectrum data set to obtain the mixed water body multi-pollution source quantitative identification model;
and the quantitative recognition module is used for quantitatively recognizing the sample to be detected based on the mixed water body multi-pollution source quantitative recognition model to obtain a quantitative recognition result of the sample to be detected.
The content in the method embodiment is applicable to the system embodiment, the functions specifically realized by the system embodiment are the same as those of the method embodiment, and the achieved beneficial effects are the same as those of the method embodiment.
While the preferred embodiment of the present invention has been described in detail, the invention is not limited to the embodiment, and various equivalent modifications and substitutions can be made by those skilled in the art without departing from the spirit of the invention, and these modifications and substitutions are intended to be included in the scope of the present invention as defined in the appended claims.
Claims (6)
1. The method for quantitatively identifying the multi-pollution sources of the mixed water body is characterized by comprising the following steps of:
obtaining a single pollution source sample and performing three-dimensional fluorescence spectrum measurement to obtain a single pollution source three-dimensional fluorescence spectrum data set;
mixing the three-dimensional fluorescence spectrum data of the single pollution source to obtain a three-dimensional fluorescence spectrum data set of the pollution source;
training a pre-constructed mixed water body multi-pollution source quantitative identification model by utilizing a pollution source three-dimensional fluorescence spectrum data set to obtain a mixed water body multi-pollution source quantitative identification model;
the mixed water body multi-pollution source quantitative recognition model adopts a multi-label multi-task convolutional neural network model, adopts the same convolutional neural network to simultaneously complete two tasks, one is for recognizing pollution source types, the other is for calculating the duty ratio of various pollution sources, cross entropy is adopted as a loss function for the pollution source type recognition, mean square error is adopted as the loss function for the pollution source duty ratio, and the total loss function of the mixed water body multi-pollution source quantitative recognition model adopts the sum of the two;
quantitatively identifying the sample to be detected based on the mixed water body multi-pollution source quantitative identification model to obtain a quantitative identification result of the sample to be detected;
the step of mixing the three-dimensional fluorescence spectrum data of the single pollution source to obtain a three-dimensional fluorescence spectrum data set of the pollution source comprises the following steps of;
preprocessing the single pollution source three-dimensional fluorescence spectrum data to obtain preprocessed single pollution source three-dimensional fluorescence spectrum data;
classifying the preprocessed single pollution source three-dimensional fluorescence spectrum data according to different types to obtain different types of single pollution source three-dimensional fluorescence spectrum data;
mixing two or more kinds of single pollution source three-dimensional fluorescence spectrum data according to different proportions to obtain multi-pollution source three-dimensional fluorescence spectrum data;
and integrating the single pollution source three-dimensional fluorescence spectrum data and the multi-pollution source three-dimensional fluorescence spectrum data to obtain a pollution source three-dimensional fluorescence spectrum data set.
2. The method for quantitatively identifying multiple pollution sources of a mixed water body according to claim 1, further comprising the step of preprocessing a single pollution source sample, wherein the preprocessing comprises the steps of filtering the single pollution source sample through a 0.22 μm filter membrane and filling the single pollution source sample into a brown glass bottle for light-shielding preservation.
3. The method for quantitatively identifying multiple pollution sources in a mixed water body according to claim 1, wherein the pretreatment of the three-dimensional fluorescence spectrum data of the single pollution source comprises subtracting an ultrapure water blank, converting fluorescence intensity into raman units (r.u.), and removing rayleigh scattering and raman scattering.
4. The method for quantitatively identifying the multiple pollution sources of the mixed water body according to claim 1, wherein the step of training the pre-constructed quantitative identification model of the multiple pollution sources of the mixed water body by utilizing the three-dimensional fluorescence spectrum data set of the pollution sources to obtain the quantitative identification model of the multiple pollution sources of the mixed water body specifically comprises the following steps:
dividing a pollution source three-dimensional fluorescence spectrum data set into a training set and a verification set;
training a pre-constructed mixed water body multi-pollution source quantitative recognition model by using a training set, outputting a pollution source type recognition result by taking cross entropy as a loss function, and outputting a pollution source duty ratio result by taking mean square error as the loss function to obtain a trained mixed water body multi-pollution source quantitative recognition model;
performing accuracy verification on the trained mixed water multi-pollution-source quantitative identification model by using a verification set to obtain an accuracy verification result;
and selecting the trained mixed water body multi-pollution source quantitative identification model with the highest accuracy according to the accuracy verification result to obtain the mixed water body multi-pollution source quantitative identification model.
5. The method for quantitatively identifying the multiple pollution sources of the mixed water body according to claim 1, further comprising the step of evaluating model performance of a quantitative identification model of the multiple pollution sources of the mixed water body, and specifically comprising the following steps:
acquiring a test set and testing a mixed water body multi-pollution source quantitative identification model to obtain model performance evaluation parameters;
and evaluating the quantitative identification model of the multiple pollution sources of the mixed water body according to the model performance evaluation parameters to obtain a model performance evaluation result.
6. A system for quantitatively identifying multiple sources of pollution in a body of water in a mixture, comprising:
the measuring module is used for acquiring a single pollution source sample and carrying out three-dimensional fluorescence spectrum measurement to obtain a single pollution source three-dimensional fluorescence spectrum data set;
the mixing module is used for mixing the three-dimensional fluorescence spectrum data of the single pollution source to obtain a three-dimensional fluorescence spectrum data set of the pollution source;
the training module is used for training the pre-constructed mixed water body multi-pollution source quantitative identification model by utilizing the pollution source three-dimensional fluorescence spectrum data set to obtain the mixed water body multi-pollution source quantitative identification model;
the mixed water body multi-pollution source quantitative recognition model adopts a multi-label multi-task convolutional neural network model, adopts the same convolutional neural network to simultaneously complete two tasks, one is for recognizing pollution source types, the other is for calculating the duty ratio of various pollution sources, cross entropy is adopted as a loss function for the pollution source type recognition, mean square error is adopted as the loss function for the pollution source duty ratio, and the total loss function of the mixed water body multi-pollution source quantitative recognition model adopts the sum of the two;
the quantitative identification module is used for quantitatively identifying the sample to be detected based on the mixed water body multi-pollution source quantitative identification model to obtain a quantitative identification result of the sample to be detected;
the step of mixing the three-dimensional fluorescence spectrum data of the single pollution source to obtain a three-dimensional fluorescence spectrum data set of the pollution source comprises the following steps of;
preprocessing the single pollution source three-dimensional fluorescence spectrum data to obtain preprocessed single pollution source three-dimensional fluorescence spectrum data;
classifying the preprocessed single pollution source three-dimensional fluorescence spectrum data according to different types to obtain different types of single pollution source three-dimensional fluorescence spectrum data;
mixing two or more kinds of single pollution source three-dimensional fluorescence spectrum data according to different proportions to obtain multi-pollution source three-dimensional fluorescence spectrum data;
and integrating the single pollution source three-dimensional fluorescence spectrum data and the multi-pollution source three-dimensional fluorescence spectrum data to obtain a pollution source three-dimensional fluorescence spectrum data set.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210968262.1A CN115219472B (en) | 2022-08-12 | 2022-08-12 | Method and system for quantitatively identifying multiple pollution sources of mixed water body |
| NL2034211A NL2034211B1 (en) | 2022-08-12 | 2023-02-23 | Method and system for quantitatively identifying multi-pollution sources of mixed water body |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210968262.1A CN115219472B (en) | 2022-08-12 | 2022-08-12 | Method and system for quantitatively identifying multiple pollution sources of mixed water body |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115219472A CN115219472A (en) | 2022-10-21 |
| CN115219472B true CN115219472B (en) | 2023-05-12 |
Family
ID=83615500
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210968262.1A Active CN115219472B (en) | 2022-08-12 | 2022-08-12 | Method and system for quantitatively identifying multiple pollution sources of mixed water body |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN115219472B (en) |
| NL (1) | NL2034211B1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115494047B (en) * | 2022-11-17 | 2023-05-12 | 广东博创佳禾科技有限公司 | Detection method and system for agricultural pollutants in water environment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5424959A (en) * | 1993-07-19 | 1995-06-13 | Texaco Inc. | Interpretation of fluorescence fingerprints of crude oils and other hydrocarbon mixtures using neural networks |
| CN102661939A (en) * | 2012-05-15 | 2012-09-12 | 北京化工大学 | Method for rapidly tracing to water pollution source |
| CN106018359A (en) * | 2016-05-12 | 2016-10-12 | 常州市排水管理处 | Sewage plant water-quality monitoring early-warning method and system |
| CN107576641A (en) * | 2017-09-11 | 2018-01-12 | 中国科学技术大学 | A kind of decomposition method and device of three-dimensional fluorescence spectrum data |
| CN111426668A (en) * | 2020-04-28 | 2020-07-17 | 华夏安健物联科技(青岛)有限公司 | Method for tracing, classifying and identifying polluted water body by using three-dimensional fluorescence spectrum characteristic information |
| CN114878528A (en) * | 2022-04-21 | 2022-08-09 | 宁波大学 | Method for quickly tracing surface water surface floating oil based on three-dimensional fluorescence spectroscopy |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7843556B2 (en) * | 2004-05-13 | 2010-11-30 | Nartest As | Portable device and method for on-site detection and quantification of drugs |
| WO2019232450A1 (en) * | 2018-06-01 | 2019-12-05 | Orb Xyz, Inc. | Detection of biological substances |
| IL262298A (en) * | 2018-10-11 | 2020-04-30 | The State Of Israel Ministry Of Agriculture & Rural Development Agricultural Res Organization Aro Vo | System and method for quantification of bacteria in water using fluorescence spectra measurements and machine-learning |
| CN109470667A (en) * | 2018-11-14 | 2019-03-15 | 华东理工大学 | A kind of combination water quality parameter and three-dimensional fluorescence spectrum carry out the method that pollutant is traced to the source |
| CN112084382A (en) * | 2020-09-04 | 2020-12-15 | 安徽思环科技有限公司 | Pretreatment method for three-dimensional fluorescence data of water quality of industrial park pollution source |
| CN113033623A (en) | 2021-03-05 | 2021-06-25 | 清华大学 | Pollution source identification method and system based on ultraviolet-visible absorption spectrum |
| CN113011478A (en) * | 2021-03-05 | 2021-06-22 | 清华大学 | Pollution source identification method and system based on data fusion |
| CN113051844B (en) * | 2021-03-12 | 2022-06-28 | 中国科学院烟台海岸带研究所 | Method for identifying and accurately tracing environmental pollution events of ocean and open water areas |
| CN113029994B (en) * | 2021-03-31 | 2023-03-24 | 扬州大学 | Microcystin concentration inversion method based on multi-source characteristic spectrum of extracellular organic matter |
| CN113311081B (en) | 2021-05-17 | 2023-08-11 | 清华大学 | Pollution source identification method and device based on three-dimensional liquid chromatography fingerprint |
| CN114563381A (en) * | 2022-01-14 | 2022-05-31 | 山东师范大学 | Water body pollution tracing method |
-
2022
- 2022-08-12 CN CN202210968262.1A patent/CN115219472B/en active Active
-
2023
- 2023-02-23 NL NL2034211A patent/NL2034211B1/en active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5424959A (en) * | 1993-07-19 | 1995-06-13 | Texaco Inc. | Interpretation of fluorescence fingerprints of crude oils and other hydrocarbon mixtures using neural networks |
| CN102661939A (en) * | 2012-05-15 | 2012-09-12 | 北京化工大学 | Method for rapidly tracing to water pollution source |
| CN106018359A (en) * | 2016-05-12 | 2016-10-12 | 常州市排水管理处 | Sewage plant water-quality monitoring early-warning method and system |
| CN107576641A (en) * | 2017-09-11 | 2018-01-12 | 中国科学技术大学 | A kind of decomposition method and device of three-dimensional fluorescence spectrum data |
| CN111426668A (en) * | 2020-04-28 | 2020-07-17 | 华夏安健物联科技(青岛)有限公司 | Method for tracing, classifying and identifying polluted water body by using three-dimensional fluorescence spectrum characteristic information |
| CN114878528A (en) * | 2022-04-21 | 2022-08-09 | 宁波大学 | Method for quickly tracing surface water surface floating oil based on three-dimensional fluorescence spectroscopy |
Non-Patent Citations (1)
| Title |
|---|
| 3D Fluorescence Spectroscopy and Its Applications;Nathalie Locquet et al.;《Electronic Absorption and Luminescence》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115219472A (en) | 2022-10-21 |
| NL2034211A (en) | 2024-02-16 |
| NL2034211B1 (en) | 2024-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Heiskanen et al. | Towards good ecological status of surface waters in Europe-interpretation and harmonisation of the concept | |
| Stubbington et al. | Biomonitoring of intermittent rivers and ephemeral streams in Europe: Current practice and priorities to enhance ecological status assessments | |
| Kaaya et al. | Tanzania River Scoring System (TARISS): a macroinvertebrate-based biotic index for rapid bioassessment of rivers | |
| CN110376343B (en) | Method and device for accurately diagnosing pollution source and electronic device | |
| CN109470667A (en) | A kind of combination water quality parameter and three-dimensional fluorescence spectrum carry out the method that pollutant is traced to the source | |
| CN1806501B (en) | Marine phytoplankton automatic distinguishing method and apparatus | |
| CN109784752A (en) | A kind of reservoir area of Three Gorges water eutrophication risk assessment early warning system and its analysis method | |
| Dubelaar et al. | High frequency monitoring reveals phytoplankton dynamics | |
| CN114943917B (en) | Algorithm for visually identifying aeration quantity of aerobic tank of sewage plant | |
| CN115219472B (en) | Method and system for quantitatively identifying multiple pollution sources of mixed water body | |
| CN116340723B (en) | Rural water pollution quick tracing method and system based on big data | |
| Bird et al. | Aquatic invertebrates as indicators of human impacts in South African wetlands | |
| CN113658640B (en) | Fresh water ecological system health evaluation method | |
| CN110033133A (en) | Silt source tracing method in a kind of river and lake system | |
| CN114417981A (en) | Intelligent river length patrol system | |
| Garmendia et al. | Testing the usefulness of a simple automatic method for particles abundance and size determination to derive cost-effective biological indicators in large monitoring networks | |
| CN113030049A (en) | Online detection and identification method and system for toxic red tide algae in fish | |
| CN101231229A (en) | Non-dyeing automatic counting method for liquid bacterium-containing quantity | |
| CN112630202A (en) | Method for identifying source of overflowing sewage in rainy days of urban drainage system | |
| de Azevedo et al. | Taxonomic and functional approaches to phytoplankton in ecosystems with different coverage of aquatic plants | |
| CN115656063A (en) | Target pollutant content analysis method and device based on hyperspectral data | |
| CN110097552A (en) | A kind of automatic division method of mouse prefrontal lobe neuron two-photon fluorescence image | |
| CN115082797A (en) | Early wheat scab monitoring and early warning method and system in field environment | |
| CN114186625A (en) | Wood identification method and system based on image feature fusion | |
| Maffezzoli et al. | Detection of ice core particles via deep neural networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |