A method for in situ inversion of kjeldahl nitrogen in a water body

Abstract

The present application belongs to the technical field of nitrogen content detection, and in particular relates to a method for in-situ inversion of Kjeldahl nitrogen in water. The method comprises the following steps: S1: sample collection; S2: chemical analysis; S3: three-dimensional fluorescence spectrum measurement and processing; and S4: model establishment. The method is simple in pretreatment, does not need to add any chemical reagent, is friendly to the environment, is rich in spectral information, is accurate and reliable, is simple, fast and convenient to operate, can be used for in-situ, rapid, accurate, efficient, non-destructive and pollution-free detection and monitoring of Kjeldahl nitrogen, and can be used for rapid judgment of the degree and trend of water organic pollution, thereby providing support for rapid environmental pollution prevention.

Description

Technical Field

[0001] This invention belongs to the field of nitrogen content detection technology, and specifically relates to a method for in-situ inversion of Kjeldahl nitrogen in water. Background Technology

[0002] Kjeldahl nitrogen in urban water bodies mainly includes ammonia nitrogen, proteins, peptides, peptones, amino acids, nucleic acids, urea, and other synthetic organic nitrogen compounds in the trivalent negative state. Nitrogen is one of the important indicators of water quality; excessive nitrogen can lead to problems such as eutrophication. Based on the correlation between nitrogen content and total organic matter, Kjeldahl nitrogen can be used to assess the nitrogen content of organic matter in water bodies, which is of great significance for assessing the degree of organic pollution and the sources of pollution in water bodies.

[0003] Currently, there are two main standard methods for determining Kjeldahl nitrogen in water: "Determination of Kjeldahl Nitrogen in Water" (GB 11891-89) and "Determination of Kjeldahl Nitrogen in Water - Gas Phase Molecular Absorption Spectrometry" (HJ / T 196-2005). While these two standard methods provide accurate and reliable results, they are cumbersome to operate, requiring pretreatment methods such as digestion and oxidation to convert organic nitrogen into ammonia nitrogen before detection. This results in long detection times and low efficiency, making them suitable for laboratory analysis but failing to meet the requirements for rapid, accurate, and efficient determination in polluted water bodies. Therefore, it is essential to provide a rapid in-situ determination technology for Kjeldahl nitrogen. Summary of the Invention

[0004] The purpose of this invention is to provide a method for in-situ inversion of Kjeldahl nitrogen in water bodies, so as to achieve accurate, efficient, non-destructive, pollution-free, and rapid in-situ determination of Kjeldahl nitrogen in urban water bodies.

[0005] Therefore, the present invention provides the following technical solution.

[0006] One aspect of the present invention provides a method for in-situ inversion of Kjeldahl nitrogen in water, the method comprising the following steps: S1: Sample collection: Collect water as the sample to be tested, and at the same time obtain the water temperature of the sample to be tested; S2: Chemical analysis: Obtain the laboratory determination value of Kjeldahl nitrogen in the sample to be tested; S3: Three-dimensional fluorescence spectroscopy measurement and processing: The sample to be tested is scanned using an online UV-induced fluorescence analyzer to obtain the original three-dimensional fluorescence spectrum of the sample to be tested. The original three-dimensional fluorescence spectra of the protein-like and humic acid-like samples to be tested are preprocessed. At the same time, the original three-dimensional fluorescence intensities of the protein-like and humic acid-like samples are Raman normalized to obtain the normalized fluorescence intensities of the protein-like and humic acid-like samples to be tested. S4: Model Establishment: A multiple linear regression model was established with the Kjeldahl nitrogen laboratory measurement value of the sample as the dependent variable, and the normalized fluorescence intensity of protein-like substances, the normalized fluorescence intensity of humic acid-like substances, and water temperature as independent variables.

[0007] In a preferred embodiment of the present invention, in step S1, the process of collecting the sample to be tested is as follows: a 2L water sample is collected at a depth of 50cm below the water surface using a vertical sampler, and after standing for 30min, 500ml~1L of supernatant is taken into a brown glass bottle as the sample to be tested and stored at 3~4℃.

[0008] In a preferred embodiment of the present invention, step S2 includes: detecting the Kjeldahl nitrogen value of the sample to be tested within 24 hours after collection, wherein the detection is in accordance with the "Determination of Kjeldahl Nitrogen in Water" (GB 11891-89).

[0009] In a preferred embodiment of the present invention, step S3 includes: S31: The original three-dimensional fluorescence spectrum of the sample to be tested is preprocessed using the Delaunay triangle interpolation method to remove Rayleigh scattering and Raman scattering; S32: Select the three-dimensional fluorescence characteristic peaks of protein-like and humic acid-like substances, and obtain the original three-dimensional fluorescence intensity of the protein-like and humic acid-like substances in the sample to be tested, respectively. S33: Subtract the blank value of ultrapure water from the original fluorescence intensity of the protein-like and humic acid-like samples, and then divide by the Raman scattering value of ultrapure water at Ex350nm / Em398nm to obtain the normalized fluorescence intensity of the protein-like and humic acid-like samples.

[0010] In a preferred embodiment of the present invention, in step S31, the original three-dimensional fluorescence spectrum data of the sample to be tested is determined by an online UV-induced fluorescence analyzer, and when the intensity of the original three-dimensional fluorescence spectrum exceeds the instrument's measurement range, it is diluted with ultrapure water.

[0011] In a preferred embodiment of the present invention, in step S32, the three-dimensional fluorescence characteristic peaks of the protein-like substance include: (i) Tryptophan-like characteristic peaks T1: Ex 275~285nm / Em 320~350nm and T2: Ex 215~237nm / Em 340~381nm; and (ii) Characteristic peaks of tyrosine-like compounds B1: Ex275~310 / Em305~320nm and B2: Ex220~237 / Em305~320nm; The three-dimensional fluorescence characteristic peaks of the humic acid-like acid include: (i) Characteristic peak A of fulvic acid-like substances: Ex 230~260 / Em 400~480 nm; and (ii) Humic acid characteristic peaks C: Ex320~360 / Em420~460nm and M: Ex290~310 / Em370~420nm.

[0012] In a preferred embodiment of the present invention, the original three-dimensional fluorescence intensity of the protein-like substance is the sum of the fluorescence intensities at the strongest points of the tryptophan-like characteristic peaks T1 and T2 and the tyrosine-like characteristic peaks B1 and B2, and the original three-dimensional fluorescence intensity of the humic acid-like substance is the sum of the fluorescence intensities at the strongest points of the fulvic acid-like characteristic peak A and the humic acid-like characteristic peaks C and M.

[0013] In a preferred embodiment of the present invention, in step S4, a multiple linear regression model is established using Origin software, wherein the regression model is: KTN (mg / L)=β0+β1*D+β2*F+β3*T+ε In the formula: KTN is the Kjeldahl nitrogen concentration (mg / L), β0 is the intercept, β1 is the protein-like fluorescence intensity coefficient, D is the normalized protein-like fluorescence intensity value, β2 is the humic acid-like fluorescence intensity coefficient, F is the normalized humic acid-like fluorescence intensity value, β3 is the water temperature coefficient, T is the water temperature (°C), and ε is the random error value.

[0014] In a preferred embodiment of the present invention, a model verification process is further included after step S4.

[0015] In a preferred embodiment of the present invention, the verification process includes: verifying the accuracy of the model by comparing the laboratory measurements of Kjeldahl nitrogen from additionally collected actual samples with the model's predicted values.

[0016] By employing the above technical solution, the present invention has at least the following advantages: This invention leverages the three-dimensional fluorescent fingerprint characteristics of proteins and humic acids. While Kjeldahl nitrogen lacks a three-dimensional fluorescent fingerprint, it exhibits a certain correlation with proteins and humic acids. Therefore, a multiple linear regression model for Kjeldahl nitrogen retrieval in urban water bodies is established by comparing water temperature, the three-dimensional fluorescence intensity of proteins and humic acids with laboratory Kjeldahl nitrogen measurements using standard methods. Verification shows that the model constructed in this invention possesses high sensitivity and accuracy, enabling in-situ, rapid, accurate, efficient, non-destructive, and pollution-free detection and monitoring of Kjeldahl nitrogen. This invention also provides technical support for rapidly determining the sources and trends of organic pollution in rivers and for decision-making regarding pollution incident control and resolution.

[0017] The method of this invention has a simple pretreatment process, requires no chemical reagents, is environmentally friendly, provides rich spectral information, is accurate and reliable, and is simple and fast to operate. It can be used to quickly determine the degree and trend of organic pollution in water bodies, and provide rapid support for environmental pollution prevention and control.

[0018] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below. Attached Figure Description

[0019] Figure 1 Three-dimensional fluorescence spectra of water quality samples from typical urban rivers in Yangzhou; Figure 2 For residual histograms; Figure 3 This is a residual-predicted value plot; Figure 4 This is a correlation diagram between the predicted values ​​from the Kjeldahl nitrogen inversion model and the laboratory measurements. Detailed Implementation

[0020] To make the technical means, creative features, achieved objectives, and effects of this invention readily understandable, the technical solutions in the embodiments of this invention will be clearly and completely described below in conjunction with the embodiments of this invention. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0021] Fluorescence technology is a rapidly developing new chemical analysis technique that has emerged in the last 30 years. It enables qualitative and quantitative analysis of organic compounds by measuring the spectral signals emitted by molecules with chemical structures such as double bonds, triple bonds, and conjugated π bonds under visible or ultraviolet light excitation. It is a convenient, easy-to-maintain, and environmentally friendly analytical technique. Three-dimensional fluorescence technology can acquire rich spectral information of organic matter in water bodies over a wide range of excitation and emission wavelengths, offering advantages such as speed, reliability, and practicality. In recent years, it has been widely used in chemical analysis and environmental monitoring.

[0022] In water bodies primarily polluted by domestic sources or biological metabolites, protein-like substances are free proteins, polypeptides, and amino acids formed from the excrement, secretions, and remains of organisms, whether from within or outside the water, through a series of biochemical degradation and transformations. In these water bodies, protein-like substances are among the main organic pollutants, maintaining a relatively fixed quantitative relationship with other initial organic pollutants. Through transformation and degradation, they form other organic pollutants. Humic acid-like substances are a large class of relatively stable organic acid complexes with various active groups such as hydroxyl and carboxyl groups, and varying molecular weights, formed from biomass and organic remains from within or outside the water through a series of complex microbial and environmental chemical processes. Their main structure is a hydrocarbon. In water bodies with severe domestic pollution or high nutrient levels, the content of humic acid-like substances increases. As stable degradation products of conventional organic pollutants in water, they are a major contributor to the concentration levels of conventional organic pollutants in the water. Aquatic proteins and humic acids are important components of the precursors and intermediate products of the biodegradation of organic pollutants in water bodies. Their content is correlated to varying degrees with a series of eutrophic organic pollution indicators, and can roughly reflect the concentration levels of degradation intermediate products. They also show a certain positive correlation with Kjeldahl nitrogen concentration levels. Proteins and humic acids possess three-dimensional fluorescent fingerprint characteristics, and are therefore often used to invert the concentrations of ammonia nitrogen (NH3-N) and chemical oxygen demand (COD) in water bodies. Cr ), permanganate index (COD) Mn The indicators include total nitrogen (TN) and five-day biochemical oxygen demand (BOD5). The content of Kjeldahl nitrogen (KTN) in water bodies is also correlated with protein-like and humic acid-like substances, but currently there is no three-dimensional fluorescence technology for retrieving KTN content in water bodies. Based on the theory that protein-like and humic acid-like substances have fluorescent properties and that their fluorescence intensity is positively correlated with KTN, this invention applies laser-induced three-dimensional fluorescence technology to the retrieval of KTN concentration in water bodies, thus providing a simple and rapid solution for obtaining KTN concentration levels in urban water bodies.

[0023] The Water Organic Pollutant Ultraviolet-Induced Fluorescence Online Analyzer (PFA), developed by the Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, is a three-dimensional fluorescence rapid monitoring instrument for common organic pollution indicators in water bodies. It features a built-in PC module for easy data editing, transmission, and system expansion. Its spectral excitation range is 239 nm to 400 nm, its spectral emission range is 260 nm to 560 nm, and its spectral resolution is 5 nm to 20 nm, covering the fluorescence response region of most biological organic pollutants in urban water bodies. The PFA is equipped with a GPS positioning device and fluorescence probes for water temperature and dissolved oxygen, enabling real-time recording of sampling point locations and water quality parameters. This invention utilizes the Water Organic Pollutant Ultraviolet-Induced Fluorescence Online Analyzer (PFA) to collect three-dimensional fluorescence spectral information of organic matter in water. Employing a combination of multiple linear regression and least squares methods, and using laboratory-analyzed Kjeldahl nitrogen as a benchmark, an inversion model is established with sample water temperature, protein-like substances, and humic acid-like substances three-dimensional fluorescence spectral data to achieve rapid inversion of Kjeldahl nitrogen in water. Compared with laboratory methods, this invention has the advantages of being simple and fast to operate. It only takes about 5 minutes to test a sample, which can quickly determine the source and degree of organic pollution in water bodies and play an important role in the distribution and investigation of organic pollution in urban rivers.

[0024] The following embodiments involve and mention: 1. Experimental apparatus The online UV-induced fluorescence analyzer (PFA) for organic pollutants in water, developed by the Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, has the following parameters: excitation source is a pulsed xenon lamp with an excitation wavelength (Ex) of 239–400 nm and a sampling interval ≤10 nm; emission wavelength (Em) is 260–560 nm with a sampling interval ≤20 nm; wavelength accuracy is ±0.5 nm; and single-band excitation response time is ≤30 s. The instrument receives and processes fluorescence signals via a photomultiplier tube (PMT), amplifier (AMP), and digital-to-analog converter (ADC) to obtain the three-dimensional fluorescence spectrum of the sample. Furthermore, the instrument is equipped with an automated sample introduction and cleaning system.

[0025] DR6000 spectrophotometer, HACH Corporation, USA.

[0026] Example 1: Model Establishment 1 Experimental Methods 1.1 Sample Source The samples used in the model were mainly from 53 sections of 44 major urban rivers in Yangzhou City, covering different seasons and water temperatures in spring, summer, autumn and winter. The pH value of the water quality ranged from 6.9 to 8.6, and the water quality categories covered Class III-V and light to severe black and odorous water in the "Surface Water Environmental Quality Standard" (GB 3838-2002) and the "Guidelines for the Treatment of Urban Black and Odorous Water Bodies".

[0027] 1.2 Sample Collection A 2L sample was collected at a depth of 50cm below the water surface, and the water temperature (T) and pH were measured simultaneously. After settling for 30 minutes, the sample tube of a water organic pollutant UV-induced fluorescence online analyzer (PFA) was inserted into the supernatant of the sample. The measurement was started, and the instrument automatically scanned the sample to obtain the original three-dimensional fluorescence spectrum. The instrument scanning parameters were: excitation wavelength Ex 239~400 nm, sampling interval ≤10 nm; emission wavelength Em 260~560 nm, sampling interval ≤20 nm. When the intensity of the original three-dimensional fluorescence spectrum exceeded the instrument's measurement range, it was diluted with ultrapure water; at the same time, 500ml~1L of the supernatant was taken as the sample and stored in a brown glass bottle at 3~4℃ for laboratory analysis of Kjeldahl nitrogen.

[0028] 1.3 Sample protein and humic acid three-dimensional fluorescence spectroscopy preprocessing like Figure 1 As shown, the three-dimensional fluorescence spectra of typical water samples from urban rivers in Yangzhou City generally exhibit seven DOM components and their corresponding main fluorescence characteristic peaks, primarily consisting of protein-like and humic acid-like substances. The protein-like substances include tryptophan-like characteristic peaks T1 (Ex 275~285nm / Em 320~350nm) and T2 (Ex 215~237nm / Em 340~381nm), and tyrosine-like characteristic peaks B1 (Ex 275~310 / Em 305~320nm) and B2 (Ex 220~237 / Em 305~320nm). The humic acid-like substances include fulvic acid-like characteristic peak A (Ex 230~260 / Em 400~480nm), and humic acid characteristic peaks C (Ex 320~360 / Em 420~460nm) and M (Ex 290~310 / Em 370~420nm). In this invention, the fluorescence intensity of protein-like peaks is the sum of the fluorescence intensities at the strongest points of the read tryptophan-like characteristic peaks T1 and T2, and the tyrosine-like characteristic peaks B1 and B2, and the fluorescence intensity of humic acid-like peaks is the sum of the fluorescence intensities at the strongest points of the read fulvic acid-like characteristic peaks A, C, and M.

[0029] First, the blank sample (ultrapure water) was scanned using a UV-induced fluorescence online analyzer (PFA). The original fluorescence intensity of the blank sample (ultrapure water) at the characteristic peaks of protein-like and humic acid-like substances, as well as the Raman scattering value of ultrapure water at Ex350nm / Em398nm, were measured. Next, after the collected sample was allowed to settle for 30 minutes, the sample inlet tube of the UV-induced fluorescence online analyzer (PFA) was inserted into the supernatant of the sample to be tested, and the instrument was activated to scan the three-dimensional fluorescence spectrum of the sample. If the fluorescence intensity of the sample exceeded the instrument's measurement range, it must be diluted with ultrapure water. The relative accuracy deviation of parallel spectral tests should be less than 2%, and the time interval between spectral analysis and chemical analysis of the same batch of water samples should not exceed 24 hours. The spectral measurement parameters were set as follows: Scanning range: excitation wavelength Ex 239~400 nm, sampling interval ≤10 nm; emission wavelength Em 260~560 nm, sampling interval ≤20 nm.

[0030] The original three-dimensional fluorescence spectrum of the sample contains Rayleigh scattering and Raman scattering. Strong Rayleigh and Raman scattering signals can affect the information analysis of the pollutant's three-dimensional fluorescence spectrum. To clearly reflect the three-dimensional fluorescence spectral characteristics of the pollutant, this embodiment further employs the Delaunay triangle interpolation method to preprocess the original three-dimensional fluorescence spectrum of the sample to eliminate the influence of Rayleigh and Raman scattering. To eliminate the influence of changes in the experimental environment and fluctuations in the spectrometer's light source, the original fluorescence intensity of the sample is normalized. Specifically, the original fluorescence intensity (Y1) of the protein-like and humic acid-like samples is subtracted from the fluorescence intensity blank value (Y0) of ultrapure water at the characteristic peaks of the protein-like and humic acid-like samples, respectively. Then, this is divided by the Raman scattering value (L) of ultrapure water at Ex350nm / Em398nm for Raman normalization. The normalized fluorescence intensity D of the protein-like sample and the normalized fluorescence intensity F of the humic acid-like sample are calculated according to the following formula: Y (normalized fluorescence intensity) = (Y1 - Y0) / L 1.4 Laboratory Analysis of Kjeldahl Nitrogen The Kjeldahl nitrogen (KTN) values ​​were obtained by analyzing the samples within 24 hours after collection, in accordance with the "Determination of Kjeldahl Nitrogen in Water" (GB 11891-89).

[0031] 1.5 Establishment of the Kjeldahl nitrogen inversion model A total of 138 data sets were obtained from the model sample. The Kjeldahl nitrogen values ​​of the tested samples ranged from 0.503 mg / L to 39.6 mg / L. After removing outliers, 128 valid data sets were obtained. Using the laboratory measured Kjeldahl nitrogen (KTN) value of the samples as the dependent variable, and the normalized fluorescence intensity value (D) of protein-like substances, the normalized fluorescence intensity value (F) of humic acid-like substances, and water temperature (T) as independent variables, a multiple linear regression model was constructed using Origin software. The regression equation model is as follows: KTN (mg / L)=β0+β1*D+β2*F+β3*T+ε In the formula: KTN is the Kjeldahl nitrogen concentration (mg / L), β0 is the intercept, β1 is the protein-like fluorescence intensity coefficient, D is the normalized protein-like fluorescence intensity value, β2 is the humic acid-like fluorescence intensity coefficient, F is the normalized humic acid-like fluorescence intensity value, β3 is the water temperature coefficient, T is the water temperature, and ε is the random error value.

[0032] 1.6 Model Diagnostic Methods The goodness of fit of the model and the significance of each independent variable are evaluated by using the regression coefficients, p-values, R² and other statistics of the inversion model. The variance inflation factor (VIF) is used to evaluate whether the model has severe multicollinearity. Residual analysis is used to evaluate whether the model has significant heteroscedasticity.

[0033] 2 Results and Analysis The coefficients and statistical parameters of each variable in the above equation are shown in Table 1.

[0034] Table 1 Note: " / " in the table indicates no data.

[0035] As shown in Table 1, the fluorescence intensity of protein-like molecules, humic acid-like molecules, and water temperature were all statistically significant (p<0.05), indicating that they all had significant effects on Kjeldahl nitrogen. Specifically, the fluorescence intensity of protein-like molecules and humic acid-like molecules had a positive correlation with Kjeldahl nitrogen, while water temperature had a negative correlation. The coefficient of determination R² = 0.760, and the adjusted R² = 0.754, indicating that the independent variables collectively explained 76% of the variation in Kjeldahl nitrogen, and the model had a high goodness of fit. Regarding multicollinearity, the variance inflation factor (VIF) for all independent variables was <5, indicating that there was no severe multicollinearity.

[0036] like Figure 2 and Figure 3 As shown, the residual analysis results indicate that the residuals basically conform to a normal distribution, the residual variance is basically stable, and there is no obvious heteroscedasticity.

[0037] Finally, the Delaunay triangle interpolation spectral preprocessing method, sample protein-like and humic acid-like normalized fluorescence intensity algorithms, and inversion model were incorporated into the UV-induced fluorescence online analyzer (PFA) to achieve in-situ Kjeldahl nitrogen inversion. The inversion model can be further optimized based on the addition of subsequent sample sources.

[0038] Example 2: Model Validation In this embodiment, the accuracy of the model constructed in Example 1 is verified by comparing the Kjeldahl nitrogen laboratory measurements of additionally collected actual samples with the model predictions. A UV-induced fluorescence online analyzer (PFA) with preprocessing techniques including Delaunay triangle interpolation, normalized fluorescence intensity algorithms for protein-like and humic acid-like samples, and a Kjeldahl nitrogen inversion model is used for in-situ Kjeldahl nitrogen retrieval. Simultaneously, the samples to be tested are brought back to the laboratory for Kjeldahl nitrogen analysis, and the model accuracy is verified using the laboratory measurements. Specifically, this includes: 1. Sample Collection Thirty-six cross-sections of 25 urban rivers in Yangzhou City, at a time period different from that in Example 1, were selected. 2L samples were collected at a depth of 50cm below the water surface, and water temperature and pH were measured simultaneously. After settling for 30 minutes, 500ml–1L of the supernatant was collected as the test sample and stored in a brown glass bottle at 3–4℃ for laboratory analysis of Kjeldahl nitrogen.

[0039] 2. In-situ inversion of Kjeldahl nitrogen First, the fluorescence intensity of the blank sample (ultrapure water) at the characteristic peaks of protein-like and humic acid-like substances was measured using a photoluminescence analyzer (PFA), and the fluorescence intensity of ultrapure water at Ex350nm / Em398nm was measured. Simultaneously, the water temperature of the sample to be tested was recorded in the PFA. Then, the fluorescence intensity of the characteristic peaks of protein-like and humic acid-like substances in the supernatant sample after 30 minutes of settling was measured using the PFA. The instrument automatically performed Delaunay triangle interpolation spectral preprocessing and Raman normalization, and substituted the results into the Kjeldahl nitrogen inversion model constructed in Example 1 to obtain the in-situ measured Kjeldahl nitrogen values ​​of the sample.

[0040] 3. Laboratory analysis of Kjeldahl nitrogen The Kjeldahl nitrogen values ​​were obtained by analyzing the samples within 24 hours after collection, in accordance with the "Determination of Kjeldahl Nitrogen in Water" (GB 11891-89).

[0041] 4. Validation of Kjeldahl nitrogen prediction results from the inversion model A total of 36 sets of Kjeldahl nitrogen validation data were obtained, with Kjeldahl nitrogen values ​​ranging from 1.79 mg / L to 39.3 mg / L. Linear regression was performed on the laboratory measured values ​​of Kjeldahl nitrogen in the samples and the predicted values ​​from the inversion model. The results are shown below. Figure 4 .

[0042] like Figure 4 As shown, the obtained linear equation is: y = 0.7611x + 2.2111, with a correlation coefficient r = 0.96. 89% of the data showed a relative error of less than 30% between the predicted values ​​from the inversion model and the laboratory measurements, with relative deviations ranging from 0.1% to 37.1%. There was no significant difference between the two sets of data. These results indicate that the model constructed in this invention has high accuracy, with a detection limit reaching 0.1 mg / L.

[0043] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the methods and techniques disclosed above without departing from the scope of the present invention to create equivalent embodiments. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for in-situ inversion of Kjeldahl nitrogen in water, characterized in that, The method includes the following steps: S1: Sample collection: Collect water as the sample to be tested, and at the same time obtain the water temperature of the sample to be tested; S2: Chemical analysis: Obtain the laboratory determination value of Kjeldahl nitrogen in the sample to be tested; S3: Three-dimensional fluorescence spectroscopy measurement and processing: The sample to be tested is scanned using an online ultraviolet-induced fluorescence spectroscopy monitor to obtain the original three-dimensional fluorescence spectrum of the sample to be tested. The original three-dimensional fluorescence spectra of the protein-like and humic acid-like samples are preprocessed. At the same time, the original three-dimensional fluorescence intensities of the protein-like and humic acid-like samples are Raman normalized to obtain the normalized fluorescence intensities of the protein-like and humic acid-like samples to be tested. Step S3 includes: S31: The original three-dimensional fluorescence spectrum of the sample to be tested is preprocessed using the Delaunay triangle interpolation method to remove Rayleigh scattering and Raman scattering; S32: Select the three-dimensional fluorescence characteristic peaks of protein-like and humic acid-like substances, and obtain the original three-dimensional fluorescence intensity of the protein-like and humic acid-like substances in the sample to be tested, respectively. The three-dimensional fluorescence characteristic peaks of the protein-like substance include: (i) Tryptophan-like characteristic peaks T1: Ex 275~285nm / Em 320~350nm and T2: Ex 215~237nm / Em 340~381nm; and (ii) Characteristic peaks of tyrosine-like compounds B1: Ex275~310nm / Em305~320nm and B2: Ex220~237nm / Em305~320nm; The three-dimensional fluorescence characteristic peaks of the humic acid-like acid include: (i) Characteristic peak A of fulvic acid-like substances: Ex 230~260nm / Em 400~480nm; and (ii) Humic acid characteristic peaks C: Ex320~360nm / Em420~460nm and M: Ex290~310nm / Em370~420nm; S33: Subtract the blank value of ultrapure water from the original fluorescence intensity of the protein-like and humic acid-like samples, and then divide by the Raman scattering value of ultrapure water at Ex350nm / Em398nm to obtain the normalized fluorescence intensity of the protein-like and humic acid-like samples, respectively. S4: Model establishment: Using the Kjeldahl nitrogen laboratory measurement value of the sample as the dependent variable, and the normalized fluorescence intensity of protein-like substances, the normalized fluorescence intensity of humic acid-like substances, and water temperature as independent variables, a multiple linear regression model was established. In step S4, a multiple linear regression model is established using Origin software. The regression model is as follows: KTN (mg / L)=β0+β1*D+β2*F+β3*T+ε Where: KTN is the Kjeldahl nitrogen concentration (mg / L), β0 is the intercept, β1 is the protein-like fluorescence intensity coefficient, D is the normalized protein-like fluorescence intensity value, β2 is the humic acid-like fluorescence intensity coefficient, F is the normalized humic acid-like fluorescence intensity value, β3 is the water temperature coefficient, T is the water temperature (°C), and ε is the random error value. The original three-dimensional fluorescence intensity of the protein-like substance is the sum of the fluorescence intensities at the strongest points of the tryptophan-like characteristic peaks T1 and T2, and the tyrosine-like characteristic peaks B1 and B2. The original three-dimensional fluorescence intensity of the humic acid-like substance is the sum of the fluorescence intensities at the strongest points of the fulvic acid-like characteristic peak A, and the humic acid-like characteristic peaks C and M.

2. The method according to claim 1, characterized in that, In step S1, the sample collection process is as follows: use a vertical sampler to collect 2L of water sample at a depth of 50cm below the water surface, let it stand for 30min, and then take 500ml~1L of supernatant into a brown glass bottle as the sample to be tested, and store it at 3~4℃.

3. The method according to claim 1, characterized in that, Step S2 includes: testing the Kjeldahl nitrogen value of the sample within 24 hours after collection, with the testing referring to the "Determination of Kjeldahl Nitrogen in Water" (GB 11891-89).

4. The method according to claim 1, characterized in that, In step S31, the original three-dimensional fluorescence spectrum of the sample to be tested is measured by an online ultraviolet-induced fluorescence spectrometer, and when the intensity of the original three-dimensional fluorescence spectrum exceeds the instrument's measurement range, it is diluted with ultrapure water.

5. The method according to claim 1, characterized in that, Step S4 is followed by a model validation process.

6. The method according to claim 5, characterized in that, The verification process includes verifying the accuracy of the model by comparing the laboratory measurements of Kjeldahl nitrogen from additionally collected actual samples with the model's predicted values.

Images